Technology Dependent Children: HospitalVs. Home Care

May 1987

NTIS order #PB87-194551

Recommended Citation:U.S. Congress, Office of Technology Assessment, Technology-Dependent Children: Hospi-tal v. Home Care—A Technical Memorandum, OTA-TM-H-38 (Washington, DC: U.S.Government Printing Office, May 1987).

Library of Congress Catalog Card Number 87-619805

For sale by the Superintendent of DocumentsU.S. Government Printing Office, Washington, DC 20402-9325

(order form on p. 107)

Foreword

This technical memorandum is about problems arising out of success. Recent ad-vances in medical technology have permitted sick children who once would have diedto survive with the assistance of sophisticated equipment and intensive nursing care.Often, the assistance is needed for just a short time, but sometimes the dependence onlife-sustaining technology is permanent. As technology for helping keep children alivehas improved, a new population of technology-dependent children has emerged.

This small but growing population of children raises new problems for the healthcare system. Because the care is expensive—often more expensive than most Americanfamilies can afford—children and their families depend on third-party payers—insurancecompanies, Medicaid, or philanthropists—to finance the needed care. But the structureof the health insurance system has not kept pace with the needs of these children. Mosttechnology-dependent children are eligible for Medicaid coverage in the hospital, butcoverage of home care is more limited. Consequently, some children have remainedhospitalized even when their families are able to provide good, lower cost care in thehome.

OTA was requested by the House Energy and Commerce Committee and the Sen-ate Labor and Human Resources Committee to examine the problems of health carefinancing encountered by technology-dependent children as part of a larger assessment,Technology and Children’s Health. The committees wanted to know how many chil-dren are technology dependent, how home care and hospital care compare in cost, andhow well private and public third-party payers cover the services needed by these children.

This technical memorandum provides a working definition of technology depen-dence and estimates the prevalence of technology dependence among American chil-dren. A principal finding is that the size of the population varies dramatically with theclinical criteria used in the definition. OTA has also found that the cost-saving poten-tial of home care depends to a great extent on attributes of the family and the homeenvironment. The ability and willingness of family members to provide ongoing nurs-ing care for a substantial part of the day are central to lowering costs to third-partypayers, although they may require great sacrifice on the part of the family.

The conduct of this study was guided by the advisory panel for the OTA assess-ment, Technology and Children Health, chaired by Harvey Fineberg. In addition, manygovernment officials and health care professionals were consulted. Information and in-sights provided by parents of technology-dependent children were also very helpful.Key OTA staff involved in the analysis and writing of the technical memorandum wereElaine J. Power and Judith L. Wagner.

Director

Advisory Panel on Technology and Children’s Health

Harvey Fineberg, ChairmanHarvard School of Public Health, Boston, MA

Lu Ann AdayCenter for Hospital Administration StudiesUniversity of ChicagoChicago, IL

Julianne BeckettChild Health Specialty ClinicUniversity of IowaIowa City, IA

Donald BerwickQuality of Care MeasurementsHarvard Community Health PlanBoston, MA

Alexander CapronMedicine and Public PolicyUniversity of Southern CaliforniaLos Angeles, CA

Norman FostProgram on Medical EthicsDepartment of PediatricsUniversity of Wisconsin HospitalMadison, WI

Robert GoldenbergDepartment of Obstetrics and GynecologyUniversity of AlabamaBirmington, AL

Michael GrossmanHealth Economics ResearchNational Bureau of Economic ResearchNew York, NY

Robert HaggertyW.T. Grant FoundationNew York, NY

Patricia KingGeorgetown Law CenterWashington, DC

Phyllis LeppertPerinatal ResearchSt. Lukes/Roosevelt Hospital CenterColumbia UniversityNew York, NY

Harvey LevyIEM-PKU Program, Childrens’ HospitalBoston, MA

Edward LisDivision of Services for Crippled ChildrenUniversity of IllinoisChicago, IL

Joanne MaconHolman Health RegionChicago Department of HealthChicago, IL

John MacQueenChild Health Specialty ClinicUniversity of IowaIowa City, IA

Janet ReisSchool of Nursing/SUNY BuffaloBuffalo, NY

Sarah RosenbaumChildren’s Defense FundWashington, DC

Barbara StarfieldDivision of Health PolicyJohns Hopkins UniversitySchool of Hygiene and Public HealthBaltimore, MD

NOTE: OTA appreciates and is grateful for the valuable assistance and thoughtful critiques provided by the advisory panelmembers. The panel does not, however, necessarily approve, disapprove, or endorse this technical memorandum.OTA assumes full responsibility for the technical report and the accuracy of its contents.

iv

OTA Project Staff–Technology-Dependent Children Hospital v. Home Care

Roger C. Herdman, Assistant Director, OTA,Health and Life Sciences Division

Clyde J. Behney, Health Program Manager

Judith L. Wagner, Project Director

Elaine J. Power, Study Director

Other Contributing Staff

David Alberts, Research Assistant

Pony Ehrenhaft, Senior Analyst

Marvin Feuerberg, Analyst

Mary Ann Hughes, Research Analyst

Virginia Cwalina, Administrative Assistant

Carol Ann Guntow, P.C. Specialist

Karen T. Davis, Secretary/Word Processor Specialist

Contractor

Harriette Fox, Fox Consultants, Inc., Washington, DC

ContentsPage

Chapter l: Summary . . . . . . . . . . . . . . . . . . . 3Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Summary of Findings . . . . . . . . . . . . . . . . . . . 3

The Population . . . . . . . . . . . . . . . . . . . . . . 3Relative Effectiveness and Costs of

Home v. Hospital Care . . . . . . . . . . . . . 5Sources of Financing for Home

Medical Care . . . . . . . . . . . . . . . . . . . . . . 7Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Chapter 2: The Size of the Technology-Dependent Child Population . . . . . . . . . . . . . 13Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Problems in Defining the Population. . . . . . 14Estimating the Prevalence of Technology

Dependence. . . . . . . . . . . . . . . . . . . . . . . . . . 16OTA’s Working Definition . . . . . . . . . . . . 16Data Sources and Sampling Problems . . . 17Estimating Prevalence . . . . . . . . . . . . . . . . . 19Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Trends in the Population . . . . . . . . . . . . . . . . 26General Trends in Chronic Illness

and Disability. . . . . . . . . . . . . . . . . . . 26Future Changes Due to New Maintenance

and Treatment Technologies . . . . . . . 28Changes in Medical Practice . . . . . . . . . . . 28

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Chapter 3: Comparing Effectiveness andCosts of Home v. Hospital Care . . . . . . . . . 3 5Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Evidence of the Effectiveness ofHome Care . . . . . . . . . . . . . . . . . . . . . . . . 35

Factors Influencing the Quality andEffectiveness of Home Care . . . . . . . . . . 36

cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37Components of Home Care Costs . . . . . . 38Issues in Comparing the Costs of

Hospital and Home Care. . . . . . . . . . . . 42Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

Chapter 4: Sources of Financing.. . . . . . . . . 51Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 51Private Health Insurance Coverage for

Technology-Dependent Children . . . . . . . 51Extent of Coverage . . . . . . . . . . . . . . . . . . . 51Adequacy of Coverage . . . . . . . . . . . . . . . . 52

Medicaid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55Background Issues. . . . . . . . . . . . . . . . . . . . 55

Page

Special Options for Financing theHome Care of Technology-DependentChildren Under Medicaid. . . . . . . . . . . . 59

State Use of Medicaid Options toServe Technology-DependentChildren . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Practices Limiting Use of the MedicaidOptions . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

State-Provided Services for Children WithSpecial Health Care Needs.. . . . . . . . . . . . . . 67A Comparison of Home Care Benefits inFive States: Three Hypothetical Cases. . . . . 71

Case l: “KM . . . . . . . . . . . . . . . . . . . . . . . . . 72Case 2: “M” . . . . . . . . . . . . . . . . . . . . . . . . . 73Case 3: “T’’. . . . . . . . . . . . . . . . . . . . . . . . . . 73

Department of Defense. . . . . . . . . . . . . . . . . . 75Other Public Programs and Services . . . . . . 7 6Charitable Organizations . . . . . . . . . . . . . . . . 77Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Appendix A. Acknowledgments . . . . . . . . . . 81Appendix B. Workshop Participants and

Agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . 83Appendix C. Implications of the

Population Definition . . . . . . . . . . . . . . . 85Appendix D. Alternative Settings of Care . 88Appendix E. The Educational System

as a Source of Health CareServices and Funding . . . . . . . . . . . . . . . 92

Appendix F. Glossary and Acronyms . . . . . 94References . . . . . . . . .

BoxesBoxA.

B.c.

D.

E.

. . . . . . . . . . . . . . . . . . . 99

Pageator-Dependent. . . . . . . . . . . . . . . . . . . 15

Profile of a VentiChild . . . . . . . . . . .Bronchopulmonary Dysplasia . . . . . . . . . 29Changing Technology in the NeonatalIntensive Care Unit . . . . . . . . . . . . . . . . . . 30Individual Benefits Management AmongPrivate Insurers . . . . . . . . . . . . . . . . . . . . . 56Reimbursement for Subacute Care inCalifornia . . . . . . . . . . . . . . . . . . . . . . . . . . 91

TablesTable No. Page1. Summary of OTA Estimates of the

Size of the Technology-DependentChild Population, 1987 . . . . . . . . . . . . . . 4

vi

Contents—continued

Table No. Page2.

3.

4.

5.

6.

7.

8.

9.

10.

11,

12.

13.

14.

15

The Population of Children CurrentlyServed in Programs EmphasizingAlternatives to Hospital Care . . . . . . . 1 6Major National Health Surveys andData Systems. . . . . . . . . . . . . . . . . . . . . . . 18Data Sources Used as Bases for OTAEstimates. . . . . . . . . . . . . . . . . . . . . . . . . 19Estimates of the Number ofVentilator-Dependent Children . . . . . . . . . . . . . . 20Estimates of the Number of ChildrenRequiring Parenteral Nutrition. . . . 21Basis for Estimate of the Population ofChildren Requiring ExtendedIntravenous Drug Therapy . . . . . . . . . . . 23Basis for Estimate of the Numberof Children Requiring Other Nutritionaland Respiratory Support . . . . . . . . 24Estimated Prevalence of SelectedChronic Conditions in Children,Ages 0 to 20, 1983 . . . . . . . . . . . . . . . . . . 25Estimated Incidence of BronchopulmonaryDysplasia, 1984 . . . . . . . . . . . . . . . . . . . . . 29Checklist of Respiratory and InfusionSkills for Home Care Patients andFamilies. . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Supplies Needed for Four HomeInfusion Therapies . . . . . . . . . . . . . . . . . . 39Sample Home Respiratory Care CostsThat Were Reimbursedby a Third-Party Payer, 1985 . . . . . . . . . . . . . . . . . . . 40Summary of Comparative AverageMonthly Charges Presented in theLiterature of Ventilator-DependentIndividuals in Hospital and HomeSettings . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Comparative Charges for Homev. Hospital Administration ofIntravenous Antibiotics as Reportedin the Literature . . . . . . . . . . . . . 46

Table No. rage16.

17.

18.

19,

20.

21.

22.

23.

24.

25.

26

Number and Percent of Children AgedO to 12 Years Covered Only by PrivateHealth Insurance, by Income Status,United States, 1986 . . . . . . . . . . . . . . . . . . 52Surveys of Employer-Sponsored GroupHealth Insurance Plans . . . . . . . . . . . . .54Medicaid Hospital Inpatient StayMaximums and Units of Payment inthe 50 States and District ofColumbia, 1985 . . . . . . . . . . . . . . . . . . . . . 58Summary of Medicaid Home HealthServices Provided in so States and theDistrict of Columbia, 1984 . . . . . . . . . . . 60Comparison of State Medicaid Optionsfor Expanded Home- and Community-Based Care . . . . . . . . . . . . . . . . . . . . . . . . . 61State Activity in Medicaid Home- andCommunity-Based Service Options asof Apr. 15, 1986. . . . . . . . . . . . . 65State Policies and Practices That LimitParticipation Under the RegularWaiver Programs That TheoreticallyCould Serve Physically DisabledChildren, April 1986 . . . . . . . . . . . . . . . . 68State Policies and Practices That LimitParticipation Under the Model WaiverPrograms, April 1986 . . . . . . . . . . . . . . . . 69State Policies and Practices That LimitParticipation Under the State PlanAmendment, April 1986 . . . . . . . . . . . . . 70Total Amount of Expenses Allocatedfor Programs of Selected Foundations,1979 and 1980 . . . . . . . . . . . . . . . . . . . . . . 77Some Conditions That May Lead toDependence on Respiratory orNutritional Support . . . . . . . . . . . . . . . . 86

vii

Chapter 1

Summary

Chapter 1

Summary

INTRODUCTION

Long-term dependence on expensive and so-phisticated health technology, and its use in set-tings other than the acute-care hospital, is notnew. The polio epidemics of the first half of thiscentury led to the use of respiratory rehabilita-tion centers (131 ), and by the end of the 1950sthere were over a thousand polio survivors requir-ing respiratory support living at home (102). Sincethen, sophisticated technologies such as hemodial-ysis, intravenous feeding, and now intensive ven-tilator care have been moved home. With eachhave come newly recognized needs for patient andfamily training and, increasingly, full-time com-plex nursing care.

Unlike the children who were part of the earlierpolio population, the present population of tech-nology-dependent children is a diverse group ofindividuals with a great range of medical diag-noses, many of them very rare. These childrenrequire a broad array of technologies and havesimilarly diverse care and nursing needs. With-out recent advances in medical technology, 1 manyof these children would not be alive. Positive-pressure ventilation, using machines that force airinto the lungs through a face mask or through asurgical opening directly into the trachea (wind-pipe), began to be used regularly on hospital pa-tients outside the operating room in the 1950s (31).

SUMMARY OF FINDINGS

The Population

“Technology-dependent” is a term used to de-scribe a small subset of the disabled child popu-lation who rely on life-sustaining medical technol-ogy and typically require complex, hospital-levelnursing care. In this technical memorandum, the

The sophistication of these devices and their man-agement to make them suitable for long-term useon infants fueled the subspecialties of neonatol-ogy and critical care pediatrics in the 1960s (69).New intravenous feeding technologies were addedto the neonatologists’ repertoire over the next dec-ade; the first person in the United States to re-spond to long-term total intravenous feeding wasan infant born without a functional intestine in1968 (46). But it is only since the beginning of the1980s that more than a handful of hospitals andphysicians have begun to consider the home envi-ronment appropriate for high-risk, technology-dependent children.

When these sophisticated medical technologiesshould be used, how and where they should beprovided, and who should pay for them are cur-rently subjects of public debate. To provide someof the foundation for this debate, this technicalmemorandum addresses four specific questions.They are:

1.2.

3.

4.

Who are the technology-dependent children?How many technology-dependent childrenare there in the United States?What services do these children require, andwhat are the costs and effects of receivingthose services at home rather than in institu-tional settings of care?To what extent does private and public in-surance cover the services needed by tech-nology-dependent children?

technology-dependent child is defined as one whoneeds both a medical device to compensate forthe loss of a vital body function and substantialand ongoing nursing care to avert death or fur-ther disability. This definition is independent ofthe setting of care or the particular credentials of

3

4

the caregiver. The ongoing nursing care, usuallyrequired for substantial parts of each day, maybe provided by a professional nurse or by atrained and skilled parent or other lay caretaker.

This definition can apply to a wide variety ofcases, ranging from children requiring the con-tinuous assistance of a device and highly trainedcaretaker to those requiring less frequent treat-ment and intermittent nursing care. Where onedraws the line on this continuum largely deter-mines the size of the population categorized astechnology dependent. OTA identified four sep-arate populations, distinguished from one anotherby their clinical characteristics, that might reason-ably be considered technology dependent:

●

●

●

●

Group I: Children dependent at least part ofeach day on mechanical ventilators.2

Group II: Children requiring prolonged in-travenous administration of nutritional sub-stances or drugs.Group III: Children with daily dependenceon other device-based respiratory or nutri-tional support, including tracheotomy tubecare, suctioning, oxygen support, or tubefeeding.Group IV: Children with prolonged depen-dence on other medical devices that compen-sate for vital body functions who requiredaily or near-daily nursing care. This groupincludes:—infants requiring apnea (cardiorespiratory)

monitors,—children requiring renal dialysis as a con-

sequence of chronic kidney failure, and—children requiring other medical devices

such as urinary catheters or colostomybags as well as substantial nursing care inconnection with their disabilities.

The first three groups are narrowly defined andlimited to children whose technology dependenceis both life-threatening and requires frequent andcomplex nursing tasks. The fourth group of chil-

2 In this technical memorandum, ventilators refer both to devicesthat apply negative pressure, such as the “iron lungs” that were usedto treat polio patients, and to devices that use positive pressure toforce air into the lungs.

dren is less susceptible than the others to long-term hospitalization, largely because the fre-quency or complexity of required nursing care issubstantially lower than for the first three groups.Under a very strict definition of technology de-pendence, this fourth group might not be in-cluded. OTA has included it to demonstrate howrapidly the technology-dependent populationgrows as additional groups are included in the def-inition.

Table 1 presents OTA’s estimates of the prev-alence of technology-dependent children in eachof the four groups. Precise estimation of preva-lence is impossible because of data limitations, soa range of estimates is provided for each group.Table 1 makes it clear that the number of tech-nology-dependent children is quite small (less than17,000 children) when the definition is limited toGroups 1-111 but increases dramatically whenGroup IV is included. Furthermore, a large num-ber of additional children not captured by thisdevice-based definition of technology dependencerequire at least as great a level of care as the chil-dren in Group IV. These children include the pro-portion of children with chronic diseases such asdiabetes, hemophilia, and epilepsy who requireconstant or very frequent nursing care as a con-sequence of the complexity and quantity of drugsand therapy they receive. If the definition of tech-

Table 1.—Summary of OTA Estimates of the Size ofthe Technology-Dependent Child Population, 1987

Estimated numberDefined population of children

Group 1:Requiring ventilator assistance . . . . . . . 680 to 2,000Group II:Requiring parenteral nutrition. . . . . . . . . 350 to 700Requiring prolonged intravenous drugs 270 to 8,275Group ///:Requiring other device-based

respiratory or nutritional support . . . . 1,000 to 6,000

Rounded subtotal (1+11+111) . . . . . . . . 2,300 to 17,000

Group IV:Requiring apnea monitoring . . . . . . . . . . 6,800 to 45,000Requiring renal dialysis . . . . . . . . . . . . . . 1,000 to 6,000Requiring other device-associated

nursing . . . . . . . . . . . . . . . . . . . . . Unknown, perhaps30,000 or more

SOURCE: Office of Technology Assessment, 1987

5

nology dependence were broadened to includethese chronically ill children, the population oftechnology-dependent children might be severaltimes again as large.

In large measure, medical practice and paren-tal attitudes determine how many technology-dependent children exist. In an area where par-ents and physicians are aggressive in managinghigh-risk infants, terminally ill children, and se-vere trauma victims, many children may survivewith long-term technology dependence. In con-trast, such children may die or may subsist withless intensive long-term support in areas wheretreatment is less aggressive due to social prefer-ence, customary practice patterns, a lack of fa-cilities, or low payment. The physician’s decisionregarding when to wean a child from a life-sus-taining device such as a mechanical ventilator alsoaffects the number of technology-dependent chil-dren, and there are large variations in weaningpractices among different physicians and differ-ent hospitals.

There is little evidence regarding the propor-tion of technology-dependent children who arehospitalized, except that it seems to vary widelyamong States. Children currently cared for athome generally meet discharge criteria such as acapacity for self- or family care; supportive, sta-ble home environments; and funding for neces-sary equipment, supplies, and professional nurs-ing services. Those who remain hospitalized areless likely to meet these criteria.

The population of technology-dependent chil-dren has increased in both size and visibility overthe past 25 years, and it will probably continueto increase for several more. In 1960, only 3 outof every 10 very-low-birthweight (under 1,500 g)newborns survived for at least a month; by 1980,nearly twice as many were surviving (170,171).Most newborn infants in this weight group requireat least temporary respiratory assistance, and theincreased survival rate has certainly increased therate of technology dependence. In fact, much ofthe survival is a result of that technology. Twenty-five years ago, the technology for long-term in-travenous nutrition did not exist; now, childrenwho have never been able to digest food are sur-viving to adulthood.

New technologies, such as improvements in theability to prevent chronic lung disease in new-borns, could reduce the size of the population,but they will not have substantial effects on theincidence of respiratory dependence for at least2 to 5 years. Meanwhile, the number of very-low-birthweight infants surviving will probably con-tinue to rise, increasing the total number of chil-dren with respiratory dependence. The spread ofacquired immunodeficiency syndrome (AIDS) inthe U.S. population will increase the number ofchildren with dependence on intravenous nutri-tion and medication. Aggressive treatment of pa-tients with ultimately fatal diseases such as cys-tic fibrosis and muscular dystrophy, and of infantswith intestinal tract disorders that would other-wise be fatal within days, also is becoming morewidespread. These developments will expand thepopulation of children who are dependent on res-piratory and nutritional technologies well intoadulthood. Payment policies that adequatelycover long-term care for these children willstrengthen this trend. Thus, it is likely that theincidence 3 of dependence on the technologies usedby children in Groups I, II, and III may as muchas double in the next few years, stabilizing or evendeclining somewhat in later years. Long survivalof those who are dependent, however, means thatthe total number of technology-dependent chil-dren will probably not decline.

Relative Effectiveness and Costs ofHome v. Hospital Care

Little objective evidence exists on the relativeeffectiveness of home v. institutional care on themedical status and development of technology-dependent children. Hospitals have generally beenconsidered the most appropriate and effective set-ting for complex medical care, while the familyhome has been considered the most appropriateand effective setting for child growth and devel-opment. Considerable experience has been gainedin moving complex medical care into the homein recent years, with much success. Many parentsand health care professionals now consider the

‘Incidence is the number of new cases during a specified periodof time, Prevalence, by comparison, is the total number of casesduring a period of time.

6

home preferable to an institution as a setting ofcare for even the most technology-dependent childwhenever home care is possible.

Effective home care is not an unqualified achieve-ment, however. First, and most importantly, ef-fective home care requires that parents want theirchild home. Second, families must be able to copewith living with the child and the intrusions ontheir own private lives as a consequence of themany other people also involved in the child’scare. Third, the effectiveness of home care de-pends on the quality of services that are providedto the family. These include:

●

●

●

●

●

●

●

●

●

adequate family training and preparation,professional caregivers trained in the relevantnursing skills,appropriately designed and well-maintainedequipment,adequate social and psychological supportservices,high-quality respite care,’appropriate home renovation,appropriate transportation,locally available emergency facilities, andcompetent case management services.5

Thus, while most family homes can be expectedto be appropriate and effective settings of care fortechnology-dependent children, a few will not beeffective for reasons inherent in the family situa-tion. Some others can be effective only if espe-cially strong social support and nursing servicesare provided. Reducing the level or quality ofthese services decreases the cost of home care tothird-party payers (at least in the short term), butit also decreases home care effectiveness. Inade-quately prepared families and home environments(as might sometimes occur in very aggressive earlydischarge programs) are likewise a threat to high-quality, effective home care.

The costs of caring for technology-dependentchildren are both high and highly variable. In thehospital, these costs depend largely on the dura-tion of dependence and the intensity of need formedical care. The care of ventilator-dependent

4Respite care is any care designed to give the family some relieffrom constant caregiving,

‘Case management is the coordination and oversight of the pack-age of health care and related services provided to an individual,

children tends to be most costly in acute-care hos-pitals because these children typically require thevery intense level of nursing found in intensivecare units; less expensive institutional care canusually be found only in special respiratory unitsof rehabilitation or long-term care facilities. Othertechnology-dependent children can be cared forin a variety of hospital settings, and some requirea level of care that can be provided by a skillednursing facility. However, nursing homes andother nonhospital facilities that accept young chil-dren and are equipped to serve their needs arerare,

The costs of home care depend less on thechild’s clinical condition and more on the attrib-utes of the family and home environment. In thehome, families have tended to bear a relativelyhigh proportion, and third-party payers a rela-tively low proportion, of the total costs to soci-ety. This situation has occurred because the fam-ilies of these children have provided most of thehighest cost services—nursing and housing—themselves.

The care of many technology-dependent chil-dren is likely to be least costly both to society andto public or private insurers when it is providedat home. Because the cost of home care dependsso heavily on social and environmental, ratherthan medical, factors, it is not possible to iden-tify a specific group of technology-dependent chil-dren based on clinical criteria alone for whomhome care will be cost saving to third-partypayers. However, if a child is medically stable,the home has a good potential for being a less ex-pensive setting of care than an inpatient facility.If family members are willing and able to providesome or most of the required nursing care, andif the child will be home long enough to offset theone-time startup costs such as training and reno-vation, the home is very likely to be the least ex-pensive setting of care for insurers. However, theuse of family members to care for these childrencan involve very high costs to the family in termsof lost income, career opportunities, leisure time,or time for routine household tasks. Reducingthese costs to the family—e.g., by paying for anurse when parents work outside the home—raises home care costs to the payer,

A few technology-dependent children cannotor should not live at home. For these children,foster care, hospital care, or other institutionalcare must be sought, and the relative costs andquality of care in these settings must be evaluated.Foster home care is often sought for childrenwhose natural parents cannot provide their care,although this setting raises costs to the govern-ment over care in an appropriate natural home.It may be preferred over institutional care, butit is likely to be difficult to find foster placementfor all technology-dependent children who needit. Other potential settings of care are:

transitional or subacute wards of acute-carehospitals,rehabilitation or chronic care hospitals (par-ticularly specialty wards of these hospitals),subacute care facilities,pediatric skilled nursing facilities, andspecialized community group homes (whichmay sometimes be considered “group” fos-ter homes).

In many areas, few or none of these alterna-tives may be available. Yet they are importantalternative sources not only of long-term care, butof transitional and respite care.

Sources of Financing forHome Medical Care

Both public and private payers have expandedthe coverage of alternative care options fortechnology-dependent children in the past 5 years.However, payment for nonhospital care is stillhindered by lack of coverage and poor coordi-nation between private and public payers.

Technology-dependent children are more likelythan other children to lack adequate private in-surance. When they are insured, their benefitsoften do not cover their extraordinary expenses,particularly in the home, and they are likely touse up their families’ insurance benefits rapidly.High lifetime maximum benefits (e.g., $1 millionrather than the still common $250,000) and casemanagement while under private insurance canextend private coverage, so long as the parentdoes not lose employment. Ultimately, however,virtually all very-long-term technology-dependent

7

children requiring a high level of nursing assis-tance will exceed the limits of their families’ pri-vate insurance policies, will be uninsurable in theself-purchase insurance market because they arepoor risks, and will end up on Medicaid. Poortechnology-dependent children, or those whosefamilies are uninsured, must turn to Medicaidfrom the start.

In most States, Medicaid does not routinely payfor full-time home nursing and other complexhome medical services. Nor are many technology-dependent children normally eligible for Medic-aid until their families have become impoverished.Since 1981, however, the Federal Government haspermitted States to waive certain Federal rules re-garding eligibility and services, allowing Statesto provide alternative mechanisms (separate fromStates’ regular Medicaid programs) to pay for in-tensive home care for technology-dependent chil-dren. Three alternative options’ are currentlyavailable to States:

1. regular 2176 “home- and community-basedservices” waivers, under which States canprovide augmented Medicaid services to spe-cified populations;

2. model 2176 waivers, a subset of the abovewaivers that can be targeted to very smalland specific populations; and

3. amendments to State Medicaid plans towaive certain restrictive eligibility income re-quirements for individuals who meet speci-fied criteria.

As of April 1986, 14 States had model 2176waivers directed specifically at technology-depen-dent or other severely physically disabled chil-dren. ’ Ten States have now amended their Stateplans to extend Medicaid eligibility to more chil-dren in this population (59).

--‘Between 1981 and 1Q84, individual waivers of Nledlcaid eltgibll-

ity restrictions were also awarded to a few technolog}~-dependen” tchildren acro~s the country. These I\’aI ~’er~ J re n (1 1 onger awardedalthough a number are still I n ctfect t(lr the ch 11 dren tvh (1 rec e]vedthem Approximately 14 States still have children ser~red un(ier ln -d]l]dual L\<ll\rt’rs (So)

“The exact number of States serving technology-dependent chil-dren under these wa]\er\ I\ unknown States may co~er ~u( h chll-dr(’n under tht,] r 2 17tJ ~~.a I ver> but nc)t actual 1}’ serve an}’, and man}States J% ere rene~iin~ the]r 2170 ~valvers In 1080, with some n e wpro~l<lons”

8—

Although these options have enhanced theavailability of Medicaid services to technology-dependent children, Medicaid still suffers fromtwo general problems regarding home care cov-erage for this population. First, the Federal Gov-ernment prohibits States from providing waiversof the usual Medicaid rules if program costs wouldincrease by doing so. If Medicaid hospital pay-ments in a State are routinely restricted (as, forexample, when Medicaid limits the number ofcovered hospital days), it can be very difficult toshow reductions in Medicaid costs when exten-sive home services are necessary. This restrictionprevents many technology-dependent childrenfrom receiving home services. Second, apart fromthe waivers, States cannot offer expanded bene-fits to a small, defined population; once covered,a service must be made available to any Medic-aid beneficiary who needs it. Both Federal andState governments have feared that expandingservices to technology-dependent children wouldmean greatly increased expenditures as other ben-eficiaries also use these services. Medicaid’s ex-perience with expanded home benefits for theelderly has been that these benefits tended to in-crease, rather than decrease, program costs.

Where States have used available options tocover home care for technology-dependent chil-dren on a case-by-case basis, they have had somesuccess in both increasing effective services anddecreasing costs. However, neither States nor theFederal Government are too willing to put in placemore general programs where costs will not beso tightly controlled. For the same reason, Stateshave even been cautious in applying the waiverand State plan amendment options.

On the other hand, States are sometimes usingMedicaid funds in ways that may not be strictlyin line with Federal regulations in order to servetechnology-dependent children more effectively.Many States find the Federal Medicaid rules in-creasingly complex and difficult to understand andimplement.

Because the federally supported State Servicesto Children with Special Health Care Needs(CSHCN) programsg offer more flexibility in im-

‘Until recently, these programs were referred to as Crippled Chil-dren’s Services (CCS).

plementation, a number of States have chosenthem as the primary vehicle to provide and co-ordinate home services to technology-dependentchildren. The role of CSHCN as the source of casemanagement and coordination for children servedunder Medicaid waivers has been particularlystrong in some States (47). The CSHCN programsare more commonly perceived as active sup-porters of care for the disabled than is Medicaid.However, the freedom that allows State CSHCNprograms to choose which groups of children theywill support (e. g., ventilator-dependent children)also allows for extreme variation among Statesin available services, and variation within Statesregarding which disabled children receive exten-sive assistance. Other public programs (such ashome-based social services) and services providedby charitable organizations supplement existingpayment for home-based medical care to vary-ing degress across States and localities. Thus, theavailability of home medical care and related serv-ices depends on the State in which the child livesand his or her particular medical condition. Atechnology-dependent child may receive adequateservices in one State through Medicaid, in anotherthrough the CSHCN program, in another througha combination of diverse sources, and in a fourthnot at all. A child requiring intravenous nutritionmay have access to adequate home services in oneState, while one who needs mechanical ventila-tion in that same State may receive no home serv-ices at all. And even if the child lives in a Statewhere home benefits to serve his or her medicalcondition are theoretically adequate, the child’sfamily may be given insufficient or conflicting in-formation regarding the availability of thoseservices.

Where adequate coverage of home medicalservices is available, other problems have begunto arise. As well-compensated alternatives to hos-pital care become more widely available, payershave incentives to limit the availability of hospi-tal care for technology-dependent children, andhospitals have increasing incentives to dischargethem, even if the family is not adequately pre-pared to take the child and no other options havebeen developed. This danger is both very real andvery great.

Y

IMPLICATIONS

Home care is not only feasible and desirable formany technology-dependent children but in manycases can also reduce costs incurred by insurers.Consequently, interest in extending home carebenefits to technology-dependent children is likelyto increase. A difficult question for third-partypayers is how to offer such benefits. Enhancedhome care benefits could be offered to all benefi-ciaries, but this strategy would substantially in-crease insurance costs and might discourage effi-cient use of such services. If insurers choose tooffer enhanced benefits to a narrowly defined setof beneficiaries, issues of fairness arise. There areno clinical criteria that can neatly separate chil-dren who deserve such benefits from those whodo not. The definition used in this technicalmemorandum which is based on the use of a med-ical device, does not capture all children who needsubstantial nursing care in the home. It is notnecessarily directly applicable in an insurancecontext.

Any expansion of home care benefits is likelyto increase the number of technology-dependentchildren at home and will have important second-ary

●

●

●

effects. These will include:

Increased early discharge from neonatal in-tensive care units. Some hospitals are begin-ning to encourage earlier discharge of prema-ture newborns (24). Increased payment forhome nursing, home phototherapy, apneamonitoring, and other services are likely tostrengthen the trend.Increased numbers of technology-dependentchildren discharged to homes before familiesfeel prepared to accept them. Overenthu-siasm in discharging children to the homecould have very serious consequences for thehealth of these children. Quality of care couldbe seriously impaired if children were dis-charged home without adequate long-termnursing support, equipment maintenance,and backup plans if home care becomes in-feasible.Problems in the quality of nursing care andequipment support in the home. A shortageof trained professional nurses and inadequateequipment-related support is already re-

ported in some places. The shortage couldget much worse if financing availability out-strips service availability. The lack of uni-form guidelines and technology-related skillcertification among home care nurses will ex-acerbate the difficulty in obtaining skilled,high-quality nursing.Increased charges for home services. Greaterdemand for high-technology home care serv-ices offers opportunities for home healthagencies to enter this field with high prices,particularly in geographic areas where thereis little competition or in areas where profes-sional nurses trained in these techniques arein great demand.Greater-than-anticipated costs to payers dueto the “woodwork effect. ” To at least someextent, enhanced home care benefits will re-place family care rather than hospital orother institutional care. This is certainlydesirable to most of the families involved andmay prevent later institutionalization ofmany children, Nonetheless, this factor willtend to increase program costs above whatwas originally anticipated.Increased demand for appropriate foster careor institutional care. Few options exist out-side of the acute-care hospital for childrenwho cannot return to a family home. Avail-ability and payment for care in small grouphomes, pediatric nursing facilities, and otherfacilities is likely to become a significant is-sue. The need for a source of respite care out-side of the home will add to the demand forappropriate facilities.Increasing numbers of technology-dependentchildren attending public schools. More chil-dren living outside of institutions will leadto more children in the schools. However,there are no Federal or State guidelines re-garding who pays for the health care neededby these children while attending school, orwho bears liability for any adverse effectsthey suffer in this setting. A lack of resolu-tion of these issues could needlessly preventmany technology-dependent children fromattending school.Need to better define the role of case man-

10.

ager and to ensure that the manager is in a nology and underlyingposition to balance the interests of the fam- its character. Childrenily, the third-party payer, and other involved ney failure do not raise

—

diseases continually alterneeding dialysis for kid-the same concerns as chil-

parties. dren needing ventilation, for the most part be-

The population of technology-dependent chil-cause payment for dialysis services is largely

dren is one with a constant undercurrent of assured and outpatient or home care has become

change. Although “technology-dependent” hasroutine in most cases. New approaches to medical

often been used as a euphemistic label for chil-practice and health care financing may yet accom-

dren whose home care was expected to be less ex-modate the most complex of today’s and tomor-

pensive than institutional care, changes in tech-row’s technology-dependent children as well.

Chapter 2

The Size of the Technology=Dependent Child Population

Chapter 2

The Size of the Technology-Dependent Child Population

INTRODUCTION

Although the technology-dependent populationis frequently discussed, it has never been defined.Simply put, technology-dependent’ children area vaguely defined subset of the much larger dis-abled child population. In this technical memo-randum, the term “technology-dependent” refersto those children who use a medical technology(embodied in a medical device2) that compensatesfor the loss of normal use of a vital body func-tion, and who require substantial daily skillednursing care to avert death or further disability.

This definition has four important character-istics. First, medical devices are used as a basisfor defining the population, because device useis observable. Second, OTA is including only life-sustaining technologies in the definition. A greatmany other children exist with extensive healthcare needs, but they are not included here. Third,the dependence is assumed to be prolonged. “Pro-longed” is not defined directly, and its meaningvaries somewhat with the type of technology, butit is assumed not to include situations such as apremature newborn who outgrows the need forventilation after only a few weeks. Finally, “skillednursing care, ” as used in this technical memoran-dum, means any care that requires highly tech-nical nursing skills, including care provided bynonprofessionals such as parents trained in these

] Some people prefer the term “technology-assisted” to the term“technology -dependent,” but the latter term has been more com-mon in recent legislation and is used in this technical memorandum.

2A medical device is any instrument, apparatus, or similar or re-lated article that is intended to prevent, diagnose, mitigate, or treatdisease or to affect the structure or function of the body (161).

skills. 3 Technology-dependent children often havemental, behavioral, or emotional disabilities inaddition to the above characteristics, but they areset apart by the level and nature of care—bothin terms of medical device support and skillednursing care—required by their chronic physicaldisabilities.

This chapter begins with a description of someof the problems encountered in defining technol-ogy dependence and the use of this term by others. q

The chapter then translates the general definitioninto a working definition for the purpose of esti-mating the number of technology-dependent chil-dren. Four clinically distinct groups of childrenare identified. Three are unquestionably technol-ogy dependent under the general definition; thefourth group meets the technical definition of tech-nology dependence but has nursing needs that aresubstantially lower than those of the first threegroups. (App. C presents some potential impli-cations of this working definition. Those impli-cations are not discussed directly in this chapter. )

The central part of this chapter presents the ex-isting evidence on how many children are in eachgroup. Finally, the chapter describes trends in thepopulation at risk of technology dependence, par-ticularly evidence on changes in the number andsurvival of children with chronic diseases, high-risk infants, and children with progressive, ter-minal illnesses.

3Nursing services are generally recognized as a group of medicalservices that cannot be performed by the average person withoutconsiderable training. They differ from custodial and personal careservices (e g., dressing, bathing, or feeding a patient ) which lesstrained people can perform safely.

‘The development of this chapter was greatly aided by the dis-cussion at a workshop conducted by OTA on the subject. The work-shop agenda and a list of participants are included in app. B.

13

PROBLEMS IN DEFINING THE POPULATION

The need to define the technology-dependentchild population arises not from any inherent at-tribute of this group but from the organizationand priorities of the health care payment systemin the United States. The home health care bene-fits of most third-party payers—private insurers,State governments, and Federal health programs—have been limited both in scope and in amount.The adaptation of sophisticated medical devicesand services to the home setting were likewiselimited. Over the past decade, however, the adap-tation of these technologies to the home has greatlyexpanded. Today, there are children with veryhigh long-term hospital costs who could be caredfor at home with such technology if payment forthat intense level of home care were available,Many third-party payers have come to pragmat-ically define technology-dependent children asthose children whose care is likely to be very ex-pensive, who could safely be cared for at homegiven sufficient services, and who are likely to costless to the payer if cared for in this environment.But such a definition does not provide clinical orfunctional criteria for identifying technology-de-pendent children. Rather, it is a criterion for case-by-case waivers of a payer’s usual limits to homehealth benefits and services.

This pragmatic, payment-based definition pre-sents a serious problem because it excludes seem-ingly similar technology-dependent children whomight benefit greatly from nonhospital care, butwho may cost more in the home due to their par-ticular family or home characteristics. Moreover,this pragmatic definition provides no basis for esti-mating the number of technology-dependent chil-dren, even the number who would fit the defini-tion as stated, because there exists no systematicway to count such children.

Legislation introduced in 1985 attempted to pro-vide more specific definitions of “technology de-pendent. ” S. 1793 defined a “medical technologydependent child” as “an individual under the ageof 21 who has a medical condition (specified bythe Secretary in regulations) which would requireinpatient hospital services in the absence of homeor community-based care, and who is dependentupon medical technology in order to avoid death

or serious injury” (emphasis added). In contrast,H.R. 2703 would have provided home care ben-efits only to ventilator-dependent people, irrespec-tive of age. These persons would be eligible forbenefits if they required a ventilator at least 6hours per day; had required this technology fora month while in a hospital or skilled nursing fa-cility; and would require institutionalization if thenecessary respiratory services were unavailableat home.

These definitions illustrate two congressionalconcerns regarding a definition of the technology-dependent child population. First, the populationof greatest concern is those children who, but forthe availability of special services and financing,could not be cared for at home. Second, “tech-nology” has been used to mean medical devices,rather than only skilled medical services. Both ofthese criteria reflect a desire to accommodate theneeds of technology-dependent children whilemaintaining control over Medicaid costs.

The prototype of the technology-dependentchild is one who cannot breathe without a me-chanical ventilator (see box A). The life of sucha child depends on an expensive and sophisticatedpiece of equipment, trained personnel to performthe necessary procedures that accompany its use,and a multiplicity of other devices, drugs, andtherapies, Until very recently, such a child wasnearly always cared for in an acute-care hospitaluntil the child died or could be weaned from theventilator, a process that could take months oryears.

While all agree that the child on a ventilatoris technology dependent, there is still a great dealof confusion over what other groups of childrenmeet this description. The population of childrenwho might be considered technology dependentis enormously diverse, Variations occur in thelength of dependence; a child may be ventilator-dependent for 10 years, or 2 years, or 2 months.Care needs vary in frequency and intensity acrosschildren as well. While one child may need skillednursing care 24 hours a day, another might needsuch care only 8 hours a day, or 2 times a dayfor 2 hours each. Some children require minimalmedical equipment but a great deal of skilled nurs-

15

Box A.–Profile of a Ventiilator-Dependent Child

Conditions leading to ventilator dependence may develop anytime in childhood or adolescence. Ateenager, for example, may suffer permanent breathing impairment due to chronic illnesses such as cysticfibrosis or muscular dystrophy. Spinal cord trauma, which can damage the nerves that enable breathing,is also a fairly common cause of ventilator dependence in children and adolescents. Or, ventilator depen-dence may be due to breathing difficulties present at birth.

A ventilator-dependent child sometimes begins life as a premature baby, initially given oxygen and24-hour ventilator support because he (or she) is unable to breathe adequately on his own. Since the babyis also unable to suck adequately, he must be tube fed as well. A physician may create a gastrostomy (asurgical opening into the stomach) to make feeding easier.

When attempts are made to reduce mechanical ventilation, it becomes clear that the infant cannotbreathe on his own even for a few minutes. The infant’s windpipe becomes irritated from having the nasaltube changed. The physician creates a tracheotomy (a surgical opening in the throat) so ventilation canbe administered more directly to the lungs. A tube, which can be connected directly with the ventilator,is inserted into the windpipe; this tube must be suctioned frequently, so it does not become clogged withsecretions, and changed regularly with great care to prevent infection at the tracheotomy site. Since aclogged tracheotomy tube would cut off all air, the infant must be watched constantly. The nurse mustperform these duties as well as administer nutrients several times a day through the gastrostomy tube, takefrequent blood samples to check the level of oxygen and other gases, administer aerosols and antibioticsto moisten the airway and prevent infection, and still offer all the normal comfort and care a newborninfant must receive. A physical therapist may begin exercises to help the infant maintain physical develop-ment; a respiratory therapist may perform procedures to help his breathing. The infant’s parents may spenda great deal of time with him to hold him and learn to care for him. If circumstances permit, they maytake him home.

As the infant matures, he may gradually be able to sustain breathing for longer and longer periodsof time on his own and may begin to learn to eat normally. Finally, if he becomes able to both eat andbreathe satisfactorily on his own, the tracheotomy and gastrostomy openings are surgically closed. Nolonger so dependent on equipment, he may still receive frequent treatments for respiratory infections andasthma-like attacks. He continues to receive physical and speech therapy to bring him up to the level ofother children his age.

ing (e. g., a child with both uncontrolled diabetesand severe epilepsy), while others may need so-phisticated medical equipment but only periodicsupervision (e. g., a capable older child receivingovernight intravenous nutrients).

Because of the lack of existing criteria and thediversity of the population that might be consid-ered technology dependent, it is difficult to dis-tinguish technology-dependent children from thelarger population of disabled children of whichthey are a part. As a child’s disability becomesgradually more (or less) immediately life-threaten-ing, and the frequency and level of skilled medi-cal intervention increases (or lessens), the bound-aries between technology dependence and lesslife-threatening disabilities blur. A child withmuscular dystrophy, for instance, loses muscle

strength gradually, first requiring braces, then awheelchair, then occasional supplemental oxygenor ventilation, and perhaps finally a full-time ven-tilator. The process may be reversed for an in-fant on a ventilator whose breathing problems re-solve over time.

Table 2 describes children who are presently

served by several programs that offer alternativesto hospital care for severely physically impairedchildren. These children display a wide range ofmedical problems. While many are obviously tech-nology dependent, requiring both highly sophis-ticated medical equipment and highly skilled andintensive nursing care, others require constantcaretaking and monitoring that depends neitheron expensive equipment nor on intensive medi-cal training.

16

Table 2.—The Population of Children Currently Served in Programs Emphasizing Alternatives to Hospital Care

Categories Description Services Sample diagnosesChildren who have acutemedical/surgical problems

Children who have a terminalillness

Children who are severelyintellectually disabled

Children who have chronicmedical problems

Children who have chronicrespiratory problems

Children who have CentralNervous System (CNS)dysfunction

Children with acute medical/surgical problems who aredischarged early from the hospitalbut who continue to needindividualized technical carefor limited periods of time

Children requiring technical carefor a terminal illness that isexpected to result in death within6 months.

Children who as the result of anillness, trauma, congenitalanomaly, or hereditary disease areseverely intellectually disabled sothat they cannot and will not inthe future be able to care forthemselves.

Children who will have chronicmedical problems for long periodsof time and are dependent ontechnical care.

Children who will be oxygendependent for relatively longperiods of time.

Children who need ventilationassistance for periods of time

Children who are completelyventilator dependent

Children who have CNS problem,either the result of trauma or CNSdisease so that they cannot andwill not be able to care forthemselves.

These children may requiremedications, unusual feedings,monitoring of vital signs, certainforms of technical treatment, etc.

These children may for a period oftime require oxygen, assistance infeeding, and/or medication forcomfort.

These children require varyingdegrees of assistance in feeding,defecation, urination, positioning,and other personal care.

These children may requirecomplex alimentation, certainmedications, suctioning,catheterization, intravenoustherapy, tracheotomies,equipment monitoring, prescribedtherapy regimens, and/orcolostomies/ ileostomies.

These children will require oxygenand may require suctioning orcardiopulmonary monitoring

These children will requireventilator care and bronchialsuctioning. They may requirecardiopulmonary monitoring andgastrostomy feeding.

These children require constantventilator care, bronchialsuctioning, and cardiopulmonarymonitoring and may requiregastrostomy feeding.

These children may requireassistance in physical positioning,feeding, defecation, and/orurination. (Some may also beventilator dependent.)

Severe infectious diseasePostoperative conditionsLow-birthweight infants

Terminal cancerRenal failure

Severe microcephalySevere post meningitisSevere hydrocephalus

Chronic malabsorptionsyndrome

Severe cystic fibrosisMultiple congenital anomaliesSevere seizure disorderDystrophiesAtrophiesMyastheniaChronic aspiration syndromeShort gut syndrome

Chronic bronchopulmonarydysplasia (BPD)

Chronic BPDPost encephalitisProgressive CNS diseaseTracheo-bronchial malaciaOndine’s curse

Chronic BPDPost encephalitisProgressive CNS disease

Progressive CNS diseaseSpinal cord trauma

SOURCE J MacQueen, “Alternatives to Hospital Care, ” unpublished, Aug. 5, 1986

ESTIMATING

OTA’s Working

THE PREVALENCE OF TECHNOLOGY DEPENDENCE

Definition sources must be available whose categories areconsistent with the definition. The most easily

To estimate the size of a population quickly and identifiable aspect of technology-dependent chil-with reasonable accuracy, criteria are needed that dren is their continual dependence on a medicalcan easily distinguish this population from others, device to replace or compensate for a vital bodyTherefore, concrete characteristics (e.g., a particu- function or avert immediate threat to life. Thus,lar diagnosis or the use of a very visible technol- in this study, four groups of children are identi-ogy) should be the basis of the definition, and data fied whose reliance on medical devices and nurs-

ing care for maintenance of life make them can-didates for classification as technology dependent:

●

●

●

●

Group I: Children dependent at least part ofeach day on mechanical ventilators. s

Group II: Children requiring prolonged in-travenous administration of nutritional sub-stances or drugs.Group III: Children with daily dependenceon other device-based respiratory or nutri-tional support, including tracheotomy tubecare, suctioning, oxygen support, or tubefeeding.Group IV: Children with prolonged depen-dence on other medical devices that compen-sate for vital body functions who requiredaily or near-daily nursing care. This groupincludes:— infants requiring apnea (cardiorespiratory)

monitors,—children requiring renal dialysis as a con-

sequence of chronic kidney failure, and—children requiring other medical devices

such as urinary catheters or colostomy bagsas well as substantial nursing care in con-nection with their disabilities.

The groups are designed to be mutually exclusive.If a child requires technologies from more thanone group, he or she is considered only as partof the applicable group with the lowest number.For example, a child requiring both ventilationand parenteral nutrition would be placed in GroupI.

Groups I, II, and III comprise children whosecharacterization as technology dependent is gen-erally accepted in discussion among parents, pro-viders, payers, and policy makers (although therange of service needs of such children varieswidely). In contrast, Group IV encompasses abroad range of children whose technology depen-dence is less life-threatening and requires less fre-quent or less complex nursing tasks. The childrenin this group are less susceptible than children in

In this technical memorandum, ventilators refer both to devicesthat apply negative pressure, such as the “iron lungs” that were usedto treat polio patients, and to devices that use positive pressure toforce air into the lungs.

17—.

the first three groups to long-term institutionali-zation as a consequence of their disabilities, andthey are not universally recognized as technologydependent. They are included here because theydemonstrate how the numbers of technology-de-pendent children change as additional groups areincluded in the definition.

Data Sources and Sampling Problems

The lack of a formal definition of technologydependence, its rarity, and the difficulty in de-tecting it have thus far prevented any reasonableestimation of the size of the population from ex-isting common health surveys. Table 3 summa-rizes a number of these surveys and the popula-tions they describe. They are generally of twotypes: institution-based surveys, such as surveysof hospital discharge records (which list items suchas age, diagnosis, and surgical procedures for alarge sample of hospital patients); and householdinterview surveys, in which family members areasked about various aspects of their health. Noneof the information from these surveys is directlycorrelated with technology dependence as definedin this technical memorandum.

Approximately 2 percent of noninstitutional-ized children (over 1 million children) are limitedin their major daily activity (e. g., attending school)(123,124). An additional group of mentally andphysically handicapped children reside in insti-tutions. Whatever the exact size of the technology-dependent child population, it must be consider-ably less than this total disabled population. Thereare two basic approaches to estimating the sizeof such a small population: counting it directly,and statistical estimation based on a sample ofchildren. Because technology dependence is rare,a sound statistical estimate would require a verylarge sample. As table 3 shows, there are no ma-jor national health surveys that are comprehen-sive enough or detailed enough to support a prev-alence estimate for this population.

The primary sources of data used as the basesfor the OTA estimate of the number of technol-ogy-dependent children are State-based programs(in most cases relating to home care provided un-der public medical aid programs) and nationalhome nutrition program registries. Table 4 sum-

Table 3.— Major National Health Surveys and Data Systems

Survey or data system Population surveyed

National Health Interview Survey(NH IS)

NHIS Child Health Supplement

National Medical Care Utilization

. .Approximately 40,000households (about 30,000children).

1 child per above household(about 15,000 children)

and Expenditures Survey

National Hospital DischargeSurvey

National Ambulatory MedicalSurvey

Care

National Health and NutritionExamination Survey

Birth Defects Monitoring Program

office of Special Education

Survey of Institutionalized Persons

Census of the Population

Approximately 6,600households (about 4,500chiIdren)

Discharges from approximately420 short-stay hospitals

Office visits to approximately3,000 physicians

Households, Including about6,000 to 7,000 children

Discharges from 928 hospitals,about 22% of U.S. births

State-reported data on childrenserved in special educationprograms

Persons living in facilities withaverage stays over 30 days

All households; sample ofinstitutionalized persons

Periodicity Relevant data elements Selected limitations. ——.Annually

1981; may bedone in futureagain

1977, 1980,planned 1987

Annually

Annually from1983-1981 ; 1985

1971-1975;1976-1 980;planned 1988

Annually

Annually

1976 only

Every 10 years

. .

Activity limitations, certainchronic conditions, hospital andphysician use

Detailed perinatal and child care,child development, child healthproblems

Same as NHIS plus additionaldata on income, Insurance,medical expenditures

Age, race, sex, medical diagnoses,procedures done in the hospital

Age, race, sex, reason for visit,diagnoses, procedures performed

Data from physical exam andlaboratory tests

Discharge abstract data for 161birth defect categories

Number of children served byhandicapping condition categow

Age, race, sex, cost of care,condition treated, physicallimitations

Age, race, sex, education, region,type of institution

Institutional population excluded, sampletoo small to detect very rare conditions,functional limitation measures verygeneral

Same as NHIS

Same as NHIS

Sample too small to detect very rareconditions, not an unduplicated count ofpersons, no data on outpatients,nonhospitalized children

Sample too small to detect very rareconditions, excludes clinic andinstitutional visits, not an unduplicatedcount of persons

Small sample, institutionalizedpopulation excluded

May not be representative sample ofbirths, newborn data only, cannot directlydetect technology dependence

Handicapped categories very broad,categories not consistently definedamong States, do not include childrennot served by programs

Limitation categories very broad,noninstitutionalized population excluded,data old, analysis excluded someinstitutions

No health-related functional dataincluded, institutional categories verybroad

SOURCES: F M. Ellman, National Association of State Directors of Special Education, Inc , Washington, DC, personal communication. January 1976, M A. McManus, S.E. Malus, C H Norton, et al , Guide toNational Data on Maternal and Child Health (Washington, DC: McManus Health Policy Inc., 1966), U S. Department of Commerce, Bureau of the Census, 1976 Survey of Institufionalfzed Persons”A Study of Persons Receiving Long-Term Care, Current Population Reports Special Studies, series P-23. no. 69, June 1978, U S Department of Education, Office of Special Education, 9th AnnualReport to Congress on the Implementation of the Education of the Handicapped Act, 1987

Source

State dataI I I I nols

Louisiana

Maryland

Massachusetts

New Mexico

North Carolina

Wlsconsln

American AssociatlonRespiratory Care

for

Table 4.— Data Sources

Population Included

19

Used as Bases for OTA Estimates

All ventilator-dependent children in State; allother children served In State program forhandicapped children

Ventilator-dependent children served in specialState home care program

Children dependent on respiratory supportdevices who are served in special State homecare program

People dependent on ventilators for longer than3 weeks

Ail children served by State Medicaid waiver fortechnology-dependent children; other similarchildren identified in State but not eligible forthe program

All children in State who are ventilatordependent and have been medically stable forat least 2 months

Children eligible for Medicaid home services onthe basis of being disabled and at a level thatwould otherwise require institutionalization

Res Respiratory therapists nationwide via their Staterepresentatives (37 States responded); asked toprovide information on all ventilator-dependentpatients they were serving

Commercial nutrition reaistries Individuals served by companies ororganizations maintaining ‘the registries betweenOctober 1984 and April 1985

OAISIS registry, OleyFoundation

Hambrecht & Quist homeinfusion market analysis

Abbott Laboratories homeinfusion market analysis—

Patients served by hospital and community -based programs responding to a 1985 survey ofsuch programs

National hospital discharge data and detailedinformation from a nonrandom sample ofhospitals

Not speci f ied

Original purpose of information collection

State Information; evaluation program for similarState programs

State Information, evaluation program for similarState programs

State Information, evaluation program for similarState programs

Survey to determine the numer of ventilatordependent Individuals

State Information, Medicaid requirements

Demonstrate potential need for pediatricrespiratory unit

State information, Medicaid requirements

Document the number of ventilator-de~endentpersons and the degree of institutionalization

Develop a database of persons on homenutritional support technologies

Develop ongoing database of characteristics ofpersons using home nutritional support

Provide estimates of the current and futuremarket for home Infusion technologies

Provide estimates of the current and futuremarket for home Infusion technologies

SOURCES M J Altken and L A Aday, Home Care for the Chron/ca//y /// and/or Dmab/ed Technology Ass/sfed Ch//d An Eva/uat/on Model, unpublished. November 1985E LIS, Crippled Children’s Serwces Chicago, IL, personal communlcatlon, April 198Q K Valdez, Human Services Department, Santa Fe NM, personalcommu nlcat!on July 22, 1986, P Tschumper, Department of Health and Soc Ial Serv!ces, M adtson. WI, personal communl cat ton J u I y 22. 1986. G Worley,Duke Unlverslty Medical Center, Durham NC, personal commun!catlon, July 1986, Care for Life, paper prepared for U S Congress Off Ice of TechnologyAssessment 1985, Oley Foundation, paper prepared for U S Congress, Off Ice of Technology Assessment 1985, B B Rucker and K A Holmstedt Home/ndustry Therapy Industry (San Francisco CA Ham brecht & Qulst, April 1984), Blue Cross and Blue Shield Assoclat!on, /n fusion Therapfes In Home Hea/thCare (Chlcaco, IL BC/BSA January 1986)

marizes these data sources and some of their char-acteristics. The OTA estimates are not derivedfrom large random samples; their validity restson the fact that very different and independentsources of information yield estimates that arewithin an order of magnitude of each other.

Estimating PrevalenceThe number of cases of a disease in the popu-

lation can be described in three ways:

• the number of new cases during a period oftime (incidence),

● the total number of cases during a givenperiod of time (period prevalence), and

● the total number of cases at a single pointin time (point prevalence).

The size of the technology-dependent popula-tion depends on which of these measures are usedand, for incidence or period prevalence, the lengthof the period. Point prevalence is analogous toan instantaneous total count of the population.Period prevalence is more relevant to surveys,which often take several months to conduct, andto programs, which usually estimate budgets forserving a population over a period of a year. Thus,period prevalence—specifically, the estimated to-tal number of technology-dependent children dur-ing 1987—is used in this technical memorandum.

20

A major problem with the data sources usedin this chapter to estimate the prevalence of tech-nology dependence is that they enumerate casesduring different time periods. Some of the sur-veys present the prevalence of a condition (e. g.,ventilator dependency) over one or more months,rather than over a year. To obtain the total prev-alence in a year, one should add to this monthlytotal the number of new cases that arose duringthe succeeding months in that year. However,there is no basis for estimating how many of thosenew cases would arise. In such cases OTA as-sumes an incidence and duration of technologydependence consistent with what few data areavailable. That assumption is stated in the dis-cussion of the estimate.

In deriving consistent prevalence figures fromthe data, OTA also uses the implicit assumptionthat the incidence and duration of technology de-pendence are stable. However, there are indica-tions that duration of technology dependence maybe increasing as children on these technologies sur-vive longer. Incidence may also be rising with in-creased survival of extremely premature babiesand the advent of acquired immunodeficiencysyndrome (AIDS), two conditions that can leadto technology dependence. An increase in eitherincidence or duration of technology dependencewould increase the prevalence of the population.These and related issues are discussed further inthe second half of this chapter.

EstimatesGroup I: The Ventilator-Dependent Population

The most soundly-based estimates are those forventilator-depedent children. To estimate the sizeof this population, OTA used the numbers ob-tained from each of the States and organizationsthat have attempted to identify such children.From these numbers, OTA obtained a rate perchild under the appropriate age group (e. g., un-der age 18) in the State. That rate was then ap-plied to the entire U.S. child population to obtainan estimate of the number of ventilator-dependentchildren that would exist if every State’s medicalpractice patterns and other relevant characteris-tics were similar to the reference State. To accom-modate differing age boundaries, OTA providesestimates both for children under age 18 and chil-dren under age 22. ’

Table 5 presents the estimated number of venti-lator-dependent children in six States and onemulti-State survey documented during the past3 years. The populations varied considerablyamong the States; different States identified or re-ported children in different age groups, rangingfrom children under age 16 (Massachusetts) to— .——

bIn extrapolating estimates to the different age groups, the lowerbound of an estimate assumes that no individuals between the agesof 18 and 22 require the technology, while the upper bound assumesthat these individuals have this attribute at the same rate as thoseunder age 18,

Table 5.—Estimates of the Number of Ventilator-Dependent Children

Number Rate per Extrapolation to US. Extrapolation to U.S.

Survey Survey Age ventilator million per survey period per yeara

Percent inState year period group dependent children Under 18 Under 22 Under 18 Under 22 institutions

Illinois . . 1985 1 year 0-21 74 19,0 1,191 1,500 1,191 1,500 3 6 %Louisiana . . . . . 1986 1 year 0-21 35 23.8 1,305 1,643 1,305 1,643 1 3 %

Maryland . . . . . . 1985 1 yearc 0-17 26 23.9 1,498 1,886 1,498 1,886 2 3 %

Massachusetts . 1983 1 month 0-15 14 13.5d 843 1,062 1,096 1,381 8 6 %N e w M e x i c o 1986 <1 month 0-21 4 74 577 726 753 948 7 5 %

North Carolina . . 1986 1 month 0-17 7f

4.3 268 337 421 530 43 ”/0

AARC survey(37 States) . . . . 1985 1 month 0-17 445 8.3 520 655 679 845 5 5 %

asee footnote 7 in text for explanation of conversion from monthly to annual Prevalencebllllnois, Louisiana and Maryland have active programs to place ventilator-dependent children at homecNOt reported, apparently at least a YeardAdjusted for 82 percent response rate Remalnlng lflstltutions were assumed slmliar to responding OneseFigure applles t. ail patients in the survey, Including adultsfFour of the seven ~hlldren had been discharged home on ventilators during tfle past s years lt IS unknown Wflettler all four children cared for at home are still alive

and ventilator-dependent, but they were assumed to be so for the purposes of this table Thus, (n converting from monthly to annual prevalence, 4/7 of the U.S extrapo-lation was not converted up, stnce this part of the number represents a 3-year prevalence rather than a I.month one

SOURCES. Office of Technology Assessment, 1987 Data from K Klrkhart, Children’s Hospital, New Orleans, LA, personal communication, January 1987; M.J Altkenand L A Aday, Home Care for the Chron/ca//y /// and/or Dsabled Technology Ass/steal ChI/d An Eva/uaf/on Model, unpublished, November 1985, K Valdez,Human Services Department, Santa Fe, NM, personal communication, July 1986, G. Worley, Duke Un!verslty Medical Center, Durham, NC, personal commu.nlcatlon, July 1986, Care for Life, “Life Sustaining Technologies and the Elderly Prolonged Mechanical Ventllat!on, ” paper prepared for U S Congress, Officeof Technology Assessment. 1985

21

children under age 22 (Maryland and New Mex-ico). The operational definition of individuals on“prolonged” ventilation also varied.

The lower bound of the ventilator-dependentestimate is based on a survey conducted duringone month in 1985 by the American Associationfor Respiratory Care. This survey yielded datafrom 37 States, which when extrapolated to theU.S. population as a whole yielded an estimated520 ventilator-dependent children under age 18that month, or roughly 680 children per year.7

This estimate is slightly higher than the lowestState-based estimate. It is used instead of thatnumber because of the evidence that at least a fewStates have much higher prevalence. The high-est estimate is based on data from Marylandwhich imply a nationwide population of 1,886ventilator-dependent children per year under age22. OTA has arbitrarily increased this number by

“TO de-r~ve an annual prevalence from a monthly one, OTA as-sumes an average duration of 3 years for ventilator dependence( Prc)bab]y ]OW but consistent with pre-1 984 data from Louisiana).The monthly incidence is then multiplied by 11 and added to themonthly prevalence to yield annual prevalence.

10 percent, to 2,000, to obtain an upper estimate.This upper bound accounts for both a possibleundercount of the universe of ventilator-dependentchildren in Maryland and for any increases in thepopulation between 1985 and 1987.

Group II: The Intravenous Therapy Population

Parenteral (Intravenous) Nutrition.—To esti-mate the number of children requiring parenteralnutrition, OTA extrapolated from the availableState data to the United States as a whole andcompared those numbers with nutritional regis-try data. Extrapolations and registry figures aresummarized in table 6.

The State data from Illinois, New Mexico, andWisconsin are underestimates of the children onparenteral nutrition in these States, because theyinclude only home patients who are monitoredby these programs. However, since the universeof ventilator-dependent children is known in 11-linois r and the proportion of those children servedby the Services to Children with Special HealthCare Needs program is also known, an estimate

Table 6 .—Estimates of the Number of Children Requiring Parenteral Nutrition

Source Basis for estimate Comments on manipulation Extrapolated U.S. estimate

Commercial registries, 373 children under age 18 on1984-85 home parenteral nutrition

documented on one of tworegistries supported by homenutrition companies.

Illinois, 1985 5 children requiring parenteralnutrition served by Stateprogram (compared to 22children in program onvent i Iators).

New Mexico, 1986 2 children on parenteralnutrition served by Stateprogram (compared to 5children in program onventiIators).

Wisconsin, 1986 4 children on parenteralnutrition served by Stateprogram (compared to 5children in program onventilators).

Assumed to be a nationalminimum estimate.

Total of 74 ventilator-dependent children known inentire State. Assumedchildren on parenteralnutrition are represented inproport ion.

Probably not total Statepopulation of children onparenteral nutrition, Usedsimple extrapolat ion.

Probably not total Statepopulation of children onparenteral nutrition, Usedsimple extrapolation.

373 children on parenteralnutrition under age 18 (per7-month period).

341 children on parenteralnutrition under age 22 (at timeof program documentation).

232 children on parenteralnutrition under age 18; 292under age 22 (at time ofsurvey).

At least 192 children onparenteral nutrition under age18 (at time of documentation).Fewer children on parenteralnutrition than on ventilators,

About 13°/0 of patients incommercial registries underage 18; apply to this figure.

Hambrecht & Quist Estimated U.S. home caremarket estimate, 1983 market of 2,700 patients per

year requiring parenteralnutr i t ion,

SOURCES Oley Foundation “Nutriltional Support and Hydration for critically and Terminally Ill Elderly, ” paper prepared for Office of Technology Assessment, September1985, E Lisj Crippled Children’s Services, Chicago, IL, personal communication, April 1986, G Cleverly, Human Services Department, Santa Fe, NM, personalcommunication 1986 P Tschumper, Department of Health and Social Services, Madison, Wl, personal communication, April 1986, B B Rucker and K AHolmstedt, Home Infusion Therapy Industry (San Francisco CA Hambrecht & Quist, Inc , April 1984)

351 children on parenteralnutrition under age 18 in 1983(for 12-month period); marketassumed growing.

22

of the universe of children on intravenous nutri-tion in Illinois can be derived by assuming thatthe latter children are represented in the programat the same rate as the former. If children onparenteral nutrition are less likely than ventilator-dependent children to be served by this program(e.g., if Medicaid or private insurers cover homeparenteral nutrition costs more comprehensivelythan ventilation costs), this method will under-estimate the nutrition population. g

Equivalent calculations cannot be performedwith the Wisconsin or New Mexico estimates.Nonetheless, these data provide additional evi-dence that there are somewhat fewer children onprolonged parenteral nutrition than on prolongedventilation.

The commercial registries provide the mostcomprehensive data on children who receivedhome parenteral nutrition. The registry data avail-able cover only 7 months, however, and not allpatients served are represented by these data. Thecumulative total from this source roughly agreeswith the estimate from the market analysis report.Both sources are underestimates of the prolongedparenteral nutrition population, because they as-sume that no such children reside in institutions.Based on these figures and the State extrapola-tions, OTA’s estimated lower bound for the num-ber of U.S. children receiving prolonged paren-teral nutrition is 350 children per year.

An upper bound for this population would ac-commodate several assumptions: 1) that the pop-ulation documented in the registry would havebeen higher had the registry covered a full year,2) that all children served at home even duringthat 7 months were not documented on the regis-try, 3) that some additional children on paren-teral nutrition reside in institutions, and 4) thatthe population has increased somewhat since1985. An upper bound of 700 (double the mini-mum estimate) accommodates these hypothesesto a reasonable degree. However, even this up-per bound may soon be an underestimate given-—— —-—

“Children requiring parenteral nutrition would be more likely tohave adequate home care insurance coverage than ventilator-depen-dent children, for example, if shift (e. g., 8-hour) nursing were anuncommon benefit. Lack of nursing is more likely to absolutely pro-hibit a ventilator-dependent child from going home than a nutri-tional-dependent child.

current trends in diseases and therapy (discussedlater in this chapter).

Intravenous Antibiotic Therapy and Chemother-apy.—Intravenous drug therapies are generallyadministered for weeks or months, rather thanmonths or years as is the case for other technol-ogies. They are included here because they aretechnologies that require substantial skilled nurs-ing and involve issues in nonhospital care that arevery similar to the issues surrounding parenteralnutrition.

Market analyses and literature reports on thenumber of individuals served in various home in-travenous drug programs are used as the basis ofestimates of the size of this portion of the Group11 population, because they are the only sourcesavailable. Table 7 summarizes these sources andthe estimate derived from them. The foundationof the estimate is a market analysis figure. Theprimary data sources and reliability of the mar-ket analyses are unreported in detail. Data fromspecific programs are used to estimate the propor-tion of the relevant population that is children.To the extent that these programs are gearedtowards adults rather than children, they under-estimate the population. Home program numberswere adjusted by OTA to account for equivalentchildren not served at home.

An estimate of the number of children who re-ceive intravenous drugs and chemotherapy is par-ticularly sensitive to whether one is consideringpatients per year or patients actually receiving in-travenous therapy at a single point in time. Thenumber of cases per year is estimated here. Basedon the information presented in the table, between268 and 8,275 children receive prolonged intra-venous drug therapy per year.

Group III: Children Dependent on OtherNutritional or Respiratory Support

Group 111 children are similar in many ways toGroup I and 11 children. Their nursing needs areoften less intensive and complex than those of chil-dren in the first two groups, however, and thesechildren may be more likely to be served at homeor in other nonhospital settings, particularly chil-dren with very-long-term dependence.

2 3— .

Table 7.—Basis for Estimate of the Populationof Children Requiring Extended Intravenous

Drug Therapy

/intravenous antibiotic therapy:Total home intravenous antibiotic therapy

market, 1984 (patients/year)a . . . . . . . . . . . 2,000 to 5,000Proportion children (range given

in Iiterature reports of individualprograms) b . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3°/0 to 46.6°/0

Implied total number of children per yearon home therapy . . . . . . . . . . . . .

Inflation for past exclusion of patients forhome care due to financial, medical, orpsychosocial reasonsc . . . . . . . . . . . . . . . .

Total number of children per yearreceiving prolonged antibiotic therapy . .

Intravenous chemotherapy:Total home intravenous chemotherapy

market, 1984 (patients/year)a . . . . . . . . . . .Approximate proportion childrend . . . . . . . .Total number of children per year,

minimum estimate . . . . . . . . . . .California hospital discharges of children

with leukemia undergoing venouscatheterization (discharges/year)e . . .

Extrapolation to U.S. (discharges/year) . .Leukemia as proportion of all childhood

cancers f. . . . . . . . . . . . . . . . . . . . . . . . . .Extrapolated U.S. number, all childhood

cancers . . . . . . . . . . . . . . .

Total number of children per year under-going chemotherapy . . . . . . . . . . . . . . . .

Total intravenous drug therapypopulation, children per year . . . . .

. .

. .

. .

. .

. .

.—

86 to 2,330

166 ”/0

143 to 3,868

2,5005%

125

1601,469

33%

4,407

125 to 4,407

268 to 8,275aSee reference 137bsee references 50,78,%,130,151CUp to 40 percent of all patients I n the studies cited here were rejected for home

therapy for these reasons Since 4 of every 10 original pattents were excludedand 6 of every 10 were Included, the figure for potential home antibiotic therapymust be re-inflated by 1660/0 to estimate the total maximum number of childrenthat would be eligible if these barriers did not exist

dA Pennsylvania report on 139 patients receiving outpatient (not home) chemo-

therapy gives the range of ages of these patients as 16 to 86, with a mean ageof 57 (86) It IS unlikely that more than 5 percent of these pattents were underage 21

‘See reference 15‘See reference 98

SOURCE Off Ice of Technology Assessment, 1987

Estimates of the number of children in GroupIII are derived primarily from two sources. First,the size of this population is estimated based onthe prevalence of these children relative to venti-lator dependence and other categories of disabil-ity in the various States. Second, registries of in-dividuals on home enteral and parenteral nutritionprograms are used as baselines to compare extrap-olated estimates for tube feeding. In this case,however, comparisons are somewhat uncertainbecause many tube-fed patients may also be de-pendent on respiratory support.

From the data presented in table 8, the mini-mum number of medically stable children requir-ing Group III respiratory and nutritional supportin the United States could be as low as 1,000. Thiswould be the case if one assumed that most of thechildren on enteral nutritional support also requirerespiratory support. The upper bound, however,is much higher. Maryland data suggest that thereare over 3,500 children on respiratory support(other than mechanical ventilation) alone; the highrelative prevalence of Group 111 children in NorthCarolina, Wisconsin, and New Mexico indicatesthat these children may be more than 10 times asprevalent as ventilator-dependent children in someStates. The Illinois-based extrapolation of about2,500 Group 111 children is a more moderate mid-dle estimate. Based on these numbers, a range of1,000 to 6,000 Group III children seems reason-able. The actual number could easily reach thehigher estimate if early hospital discharge of pre-mature infants becomes more common.

Group IV: Children Requiring OtherLife-Sustaining Medical Devices andAssociated Skilled Care

Group IV comprises children who require life-sustaining medical devices but whose nursing careneeds are generally less complex, less prolonged,or less frequent than the needs of children inGroups I through III. It includes three subgroups:1) infants requiring apnea monitors, 2) childrenrequiring renal dialysis, and 3) children requir-ing other life-sustaining medical devices in con-junction with substantial nursing care.

The Food and Drug Administration has esti-mated that approximately 40,000 to 45,000 homeapnea monitors for infants are currently in use(173). There is considerable controversy regard-ing the appropriate indications for monitoring,and many of these children may be monitored forreasons not considered by all physicians to besufficient. A National Institute of Health panelestimated that approximately 6,800 to 17,000 ofhome monitors are prescribed as a result of anapparentl y life-threatening episode in an infant(173). OTA has used 6,800 as the lower boundand 45,000 as the upper bound for an estimateof the number of medically necessary home ap-nea monitors in use and makes the simplifying as-

24

Table 8.—Basis for Estimate of the Number of Children Requiring Other Nutritional and Respiratory Support

Information source Data Implications CommentsOASIS registry, OleyFoundation

147 children ages 0-10 inregistry; 92 require parenteralnutrition

Ratio of enteral to parenteralnutrition is 1:1.67

Proportion of children alsousing respiratory supportunknown; proportion of tube-fed population covered byregistry unknown

Same as Oley Foundationregistry

Commercial registries 368 children in registryrequiring enteral nutrition (i.e.,tube feeding)

7,500 persons in U.S. receivedhome tube feeding in 1983

Ratio of enteral to parenteralnutrition is 1:1 .01

990 tube-fed childrenin the U.S. at home

Hambrecht & Quistmarket analysis

Based on discharge data andsample of hospitals.Extrapolation assumes that13.2% of tube-fed populatonare children (from commercialregistry proportion)

Unknown basis for estimate.Same assumption of 13.2%children as above

Of net addition to populationof 82 gastrostomies, assumeseach child received only onegastrostomy and required itfor one year

Ratio of ventilator: Group Illchildren in program 22:36;apply this to extrapolation of1,500 ventilator-dependentchildren in U.S. to yield totalGroup Ill estimate

Assumes Maryland identifiedthe universe of such childrenin the State

Presumably is anunderestimate if not allsimilar children are served byState program. Probablyconsiderable overlap betweentube feeding and respiratorysupport groups. Prevalence ofGroup II probably overstated

One-month survey,hospitalized children only

per year

per year

through

Abbott Laboratoriesmarket analysis

5,500 persons in U.S. receivedhome tube feeding in 1983

726 tube-fed childrenin the U.S. at home

777 children tube-fedCalifornia hospitaldischarge data forchildren

97 gastrostomy procedures,15 closures in 1983 gastrostomies each year

State data:Illinois

2,445 Group Ill children peryear in the U.S.36 children on Group Ill

technologies served by homecare program

Maryland

Wisconsin

87 children in State requiringrespiratory support; 61 requireother than ventilators

3,513 children in the U.S. peryear requiring respiratorysupport other than ventilators

2,401 U.S. children requiringtube feeding at any one pointin time; up to 4,800 requiringrespiratory support. Ratio ofventilator: Group Ill supportsabout 1:10

49 children served in Stateprogram require tube feeding;49 require respiratory assistdevices (other thanventilators)

North Carolina

New Mexico

8 hospitalized children inState with prolonged oxygendependence (compared to 3on ventilators)

Ratio of ventilator: oxygensupport about 3:8

Ratio of ventilator: Group Illsupports about 1:18

Prevalence of Group Illprobably overstated due to

1 ventilator-dependent child;18 other children requiringrespiratory and nutritionalsupport

small number of ventilator-dependent children served

SOURCES: M,J. Aitken and L.A. Adav. Home Care for the Chrorrica//v Ill and/or Disabled Tecfrno/oav Assisted Child: An Evaluation Model, untwblished. November 1985.J Bates, San Diego Children’s Hospital, San Diego, CA, personal communication, July-{986; Blue Cross/Blue Shield Association, infusion Therapies in HomeHealth Care (Chicago, IL: Blue Cross/Blue Shield Association, January 1966); G. Cleverly, Human Services Department, Santa Fe, NM, personal communication,August 19&; L.L Heaphey, The Oley Foundation, Albany, NY, personal communication, August 1986, E. Lis, Crippled Children’s Servjces, Chicago, IL, personalcommunication, April 198B; Oley Foundation, Inc , “Nutritional Supporf and Hydration for Critically and Terminally Ill Elderly” Utilization in the Home,” contractpaper prepared for the Office of Technology Assessment, U S Congress, Washington, DC, September 1985; B B Rucker and K A. Holmstedt, l+orrre InfusionTherapy Industry (San Francisco, CA Hambrecht & Quist, April 19B4), and P Tschumper, Department of Health and Social Services, Madison, WI personalcommunication, April 1986.

—..— .

25—

Table 9.—Estimated Prevalence of Selected ChronicConditions in Children, Age O to 20, 1983

Approximate numberPrevalence per of children in the

Condition 100,000 children, 1980 United States, 1983

Mental retardation ., . . . . . . . . . .:. . . . . . .—— —

2,500 1,781,300Asthma (moderate and severe) . . . . . . . . . . . 1,000 712,500Diabetes mellitus ... . . . . . . . . . . . . . . . . 180 128,300Congenital heart disease (severe) . . . . . . . . 50 35,600Spina bifida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 28,500Sickle cell anemia . . . . . . . . . . . . . . . . . . 28 20,000Cystic fibrosis . . . . . . . . . . . . . . . . . . . 20 14,300Hemophilia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 10,700Leukemia (acute Iymphocytic leukemia) . . . . . 11 7,800Chronic renal failure . . . . . . . . . . . . . . . . . . . . . 8 5,700Muscular dystrophy. . . . . . . . . . . . . . . . . . 6 4,300Traumatic brain injury . . . . . . . . . . . . . . 5 3,600SOURCE Prevalence rates from G L Gortmaker and W. Sappenfleld “Chronic Childhood Disorders: Prevalence and Impact,

Pediatric Clinics of North America 31(1) 318, February 1984 Population size estimates calculated by OTA basedon prevalence rates and U S Census population data

sumption that this range represents the numberof monitors in use per year.

The minimum estimate of the number of chil-dren requiring renal dialysis is from Medicare End-Stage Renal Disease Program data. The HealthCare Financing Administration, which adminis-tered Medicare, documented 1,713 patients age Oto 19 receiving dialysis in 1985 (110,166). Add-ing a minimum of 171 patients to this figure toaccount for patients age 20 to 21 (one-tenth, or2 average age years, of the initial figure) and in-flating the total figure by 3.6 percent per year (theincrease documented from 1983 to 1984), yieldsa minimum of 2,022 children under age 22 on di-alysis during 1987. This number underestimatesthe number of children requiring dialysis, sincesome are covered by private insurance. A maxi-mum estimate assumes that all children withchronic renal failure require dialysis. The preva-lence of this condition has been estimated at 8 per100,000 children, or nearly 6,000 children underage 22 in the United States (see table 9).

There exist no appropriate data at all to esti-mate the number of children requiring other de-vices and associated nursing care such as urinarycatheterization and colostomy care. 9 This group

— —-.911ata from the National Center for Health Statlstlcs, which sur-

\’e}rs a large sample ok ho~pital d]schar~es every year, suggett thatthe annual incidence of colost<~rnies and ileostornles {n children maybe a tew thousand per year ( 172). The sur~’ey ]s not large enoughto estimate an accurate number of these proceciures for children,but it IS certainly less than 10,000 per }rear and probabl> less than5,000

is very large; indeed, it may be larger than allother groups combined. If this group is includedin the population of technology-dependent chil-dren, the size of that population will increase dra-matically. Many children with spina bifida andother spinal conditions, for example, require uri-nary catheterization. The total number of childrenin this group could easily be 30,000 or more (seetable 9).

If the definition of technology dependence usedin this technical memorandum were not limitedto children using medical devices, this group couldpotentially include a substantial proportion ofchildren with hemophilia, insulin-dependent di-abetes, and many other chronic diseases. Whilemost such children require periodic injections ofmedications and a relatively modest amount ofnursing care by family members, a few have moreintensive needs for monitoring and nursing. It isonly the lack of dependence on a major medicaldevice, not necessarily a difference in nursing andcare needs, that distinguishes this population fromthose children included in Group IV.

Table 9 presents prevalence estimates for sev-eral serious chronic illnesses in children. Unfor-tunately, no quantitative information on the levelof technology and nursing needs for this or anyother subpopulation of children with chronic ill-ness exists. Some of these children have alreadybeen included in groups mentioned above; for ex-ample, children with cystic fibrosis, muscular dys-trophy, traumatic brain injury, or severe asthma

2 6

who meet the definition of technology dependencelikely need respiratory or nutritional assistanceand would be included in Groups I through III.Children with rheumatoid arthritis and leukemiawould be included under Group 11 if they requiredperiodic intravenous drug therapy.

Wisconsin data illustrate how large the popu-lation of “technology-dependent” children couldbe if the definition did not require dependence ona device that compensates for a vital body func-tion as a necessary criterion (but retained the “sub-stantial nursing needs” criterion). Wisconsin oper-ates a program that, among other criteria, enableschildren to receive certain health care benefits ifthey would be permanently institutionalized with-out these benefits and could be served less expen-sively at home. Of 181 children served by this pro-gram in mid-1986, one-third of the total requireda very high level of care but did not require nutri-tional, respiratory, or other mechanical support.

What sources of information might be tappedin the future for more precise estimates of thenumber of technology-dependent children? Onepossibility might be a school-based survey, tar-

TRENDS IN THE POPULATION

Future changes in the size of the technology-dependent population will depend on three sepa-rate

1.

2.

3.

factors:

changes in the number of children who havethe diseases and conditions that lead to tech-nology dependence;technological change, which can either in-crease the size of the population (if new tech-nologies lead to increased survival dependenton long-term life-saving equipment), or de-crease population size (if new technologiesallow less intensive equipment and serviceneeds, or prevent the development of dis-abling conditions); andchanges in medical practice and social atti-tudes, which are themselves affected by fac-tors such as the emergence of new technol-ogies and the availability of third-partypayment,

geted at the population most likely to include asignificant proportion of technology-dependentchildren—those children who have been individu-ally assessed prior to educational placement. Asimilar approach is currently being used in an on-going study to estimate more accurately the num-ber of children with hemophilia, cystic fibrosis,and spina bifida (73). The approach is fraughtwith its own problems, not the least of which isthat very young children and children living inhospitals or long-term care institutions would notbe captured. Also, technology-dependent childrenare rare even among children assessed for possi-ble special education placement .’” A very largesurvey would be required to produce a reliableestimate of size of the population. Still, this sourceoffers one possibility for estimating future appar-ent or real changes in the prevalence of technol-ogy dependence.

I ~In Fairfax county, Virginia, for instance, 700 children Wereserved in home or school-based special education preschool pro-grams (ages 2 to 4 ) in August 1986 (14). Of these 700 children, 6might have qualified as technology dependent (4 served in class-rooms had tracheostom ies or gastrostom ies, and 2 served at homehad special medical problems).

General Trendsand Disability

in Chronic Illness

Trends in chronic illness and disability overtime are somewhat difficult to identify. Nationalsurveys show that the proportion of children withreported major activity limitation has increasedsubstantially in recent years, from approximately1.1 percent in 1967 to the present 2 percent (124).However, this finding may be caused by any ofa number of influences. Some of the apparent in-crease may be due to changes in survey method-ology and in families’ awareness of illness overtime (123,124), rather than to real changes in dis-ability rates. Another explanation is increased sur-vival of children with certain chronic illnesses,such as cystic fibrosis and spina bifida. A thirdpossible explanation is that new technologies andnew systems of care, such as intensive care unitsfor newborn infants, are resulting in more chil-

2 7

dren who survive birth or trauma but with severelong-term disabilities. The absolute number of dis-abled and chronically ill children will increase asthe children survive longer, even if the rates ofonset of various disorders are unchanged .1’

The number of children with inherited chronicdiseases that can lead to technology dependenceis unlikely to change significantly due to changesin the incidence of these disorders. Gortmaker andSappenfeld conducted an extensive review of theliterature in 1980 to investigate the prevalence ofa number of common childhood chronic diseases(72). They noted that the incidence of most suchdiseases has been stable over time. It is now pos-sible that new technologies permitting prenataldiagnosis of muscular dystrophy and cystic fibro-sis may decrease the incidence of these diseases,if couples choose to terminate pregnancies whena fetus has been identified as having a geneticmarker associated with the disorder, However,these prenatal diagnostic technologies are unlikelyto have a major effect on overall incidence ofchronic disease.

A more important factor affecting the numberof children with chronic diseases, and one withimplications for technology dependence, is the sig-nificant improvements in survival for childrenwith many life-threatening diseases. Improve-ments over the past two decades in survival ratesfor children with leukemia, diabetes, certain heartdefects, sickle-cell anemia, and chronic kidney dis-ease have greatly increased the number of suchchildren who live to adulthood (72). Better andmore aggressive treatments for spina bifida andmuscular dystrophy have also increased the sur-vival of children with these disorders (35). Chil-dren with cystic fibrosis who would have died inearly childhood two decades ago are now surviv-ing, and over 50 percent of them live into adult-hood (109). Long-term survival of children withintestinal malformations will greatly increase the

1 I Evidence frc)m pub] ic scho(>] records tends to supwrt the Premise

that the number {~t children with severe dlsabilitles, or at least thenumber bein g served b}’ pub] ]{ sc h(~~~ls, has I ncreawd. The pr[~p(]r-t](~n of multihandlcapped children, F(>r Instance, increased from 0, 12t(> 0.16 percent of sch(~(}l enrollment between 1 Q70-77 and 1982-83( 187) , Ho~vcver, >ch[)(]l data (In ct]sabillty IS genera l ly c(~n~]deredunre]iab[[’ because 01 the ~reatl} \ar}’ing det Irutlon< dltterent scho(~ld]str]c t\ L1$L’

total number of children requiring parenteral nu-trition.

Changes in head and spinal cord injury rates,and changes in the survival of severe trauma pa-tients, could affect the size of the population.More important, however, is the rising incidenceof acquired immunodeficiency syndrome (AIDS).AIDS is likely to continue to spread in infants asit spreads in women, since the virus can be com-municated from mother to fetus at or before birth(40). This disease is likely to increase the numberof children receiving prolonged intravenous drugtherapy, nutritional support, and considerablenursing care. A few hospitals have experienceddramatic increases in the number of babies withAIDS and in the number of those babies grow-ing up in the hospital (22).

The most profound changes in the incidenceand prevalence of disorders leading to technol-ogy dependence seem to be occurring in neona-tal care. It has been widely asserted that the in-creased survival of very-low-birthweight (lessthan 1,500 g) infants, due to improved, aggres-sive neonatal intensive care, has resulted in asharp surge in the number of ventilator- and othertechnology-dependent children. If this is true, con-tinued advances towards survival of very-low-birthweight infants can be expected to increase thenumber of such children. If, on the other hand,technologies are successfully developed that canprevent prematurity y or moderate the developmentof chronic lung disease in newborn infants, thenumber of infants on long-term ventilation andnutritional support may be considerably reduced.

Most medical evidence thus far supports thecontention that the increased survival of very-low-birthweight infants has not increased overall ratesof disability, but it may have increased the ac-tual number of severely disabled children. A studyof changes in infant morbidity and neonatal mor-tality between 1976 and 1978-79 found that ne-onatal mortality decreased by 18 percent duringthis time, while infant morbidity also decreasedby 16 percent (144). Overall, therefore, newbornsurvival did not lead to an increase in long-termdisability. However, the detected decrease in dis-ability was among the minor disability categories;“the proportion of children with severe or mod-

28

crate congenital anomalies or developmental de-lay did not change” (144). Other evidence sup-ports three conclusions:

1.

2.

3.

within groups of infants of a given birth-weight, handicap rates remain stable or de-crease over time;handicap rates are greatest in the lowestbirthweight groups; andinfants in the lowest birthweight groups aresurviving in increasing numbers (162).

If the incidence of severe disorders is unchangedor even declines somewhat, but the total numberof neonatal survivors increases, then the numberof infants with severe disabilities increases over-all. Box B describes a common source of respira-tory disability in infants and the differing ratesof incidence of this disorder in infants of differ-ent birthweights.

Future Changes Due toNew Maintenance andTreatment Technologies

The most promising technologies to decreasethe incidence of long-term technology-dependenceare those aimed at preventing the need for long-term respiratory support in infants. Current ef-forts to combat chronic lung disease in newbornsare described in box C. One or more of these tech-nologies may eventually greatly reduce the num-ber of infants with long-term technology depen-dence. However, significant changes are not likelyto be apparent for a few years yet.

New technologies may have other direct effectson trends in the technology-dependent popula-tion, aside from their effects in reducing the under-lying disorders that lead to technology depen-dence. For example, advances in implantableinfusion pumps for long-term chemotherapy andimplantable phrenic nerve pacers to stimulatebreathing could reduce the constant, complexnursing needs associated with many technology-dependent children.

Enhanced access to transplant technology mayeither increase or decrease the number of childrenrequiring intensive long-term nursing services. Asthe number of infants and children receiving bone,

liver, heart, and other organ transplants growsdue to increased transplant experience and en-hanced insurance coverage, children recuperatingfrom transplants may become a group for whomintensive home medical care is both socially andfinancially desirable. Access to intensive medicalservices in the home setting might allow these chil-dren to leave the hospital earlier than wouldotherwise be possible, and they may have ongo-ing nursing and technology needs. On the otherhand, increased transplant success could obviatethe need for very-long-term dependence on tech-nologies such as insulin (through pancreas trans-plants), parenteral nutrition (through bowel trans-plants), and dialysis (through kidney transplants).

Changes in Medical Practice

The wide range of prevalence estimates for tech-nology dependence suggests that medical practicepatterns may vary considerably among regions,States, and medical centers. Some of these differ-ences may be in simple treatment protocols. Forexample, there is some evidence that differencesin medical practice can inadvertently affect theincidence of chronic lung disease in newborns. Inan examination of treatments and rates of bron-chopulmonary dysplasia in eight hospitals withregional neonatal intensive care units, Avery andcolleagues found that the rates of this newbornchronic lung disorder varied considerably amongcenters, even after adjusting for differences in thenewborn populations (11). They concluded thatthe differences in routine treatment practicesamong these centers were probably responsiblefor the differences in the rates of this disorder,implying that changes in the routine practices ofhospitals with higher rates could reduce the inci-dence of dependence on long-term respiratorysupport.

Other researchers have documented the varia-tion in routine treatment patterns among physi-cians treating people with fatal chronic diseases.A 1981 study demonstrated that positive-pressureventilation may extend the lives of children andadults with muscular dystrophy by an average of7 years (12). Only one-third of the patients in thestudy had tracheotomies. By comparison, in a1985 survey of Muscular Dystrophy Association

—

29

Box B.-–Bronchopdrnmary Dysplasia

One of the most common sequelae of neomtal intensive care, and one with particular implicationsfor technology dependence, is bronchopuhnonary dysplasia (BPD). First remgnized in the early 1960s (154),this condition sometimes occurs in infants requiring mechanical ventilation soon afterbirth. An infant withBPD is unable to be weaned from ventilation during the first month after birth due to certain changes inthe lung that can often be detected by X-ray (71).

Pneumonia, meconium aspiration, patent ductus arteriosus, and apnea of prematurity are among themany conditions that can lead to the initial need for assisted ventilation (and, thus, sometimes BPD) innewborns (71). The most common reason for initial ventilation, however, is respiratory distress syndrome.This syndrome, characterized in its initial stages by an increasing need for oxygen, is often experiencedby very premature infants because art essential lining layer in the lung (pulmonary surfactant) has not yetdeveloped (99).

Table 10 presents estimates of the annual incidence of BPD by birthweight category. There are nonationally representative data on the incidence of BPD. A multi-center study of 700 to 1,500 g babies in1983 and 1984 found that one-third of the survivors had chronic lung disease (11). OTA used this 33 per-cent estimate for very-low-birthweight infants, although other researchers reported rates of BPD incidenceamong their institutional populations varying from 25 to 75 percent of respiratory distress syndrome sur-vivors under 800 g at birth, and from 13 to 62 percent of survivors weighing less than 1,000 g at birth(17,25,45,74,85,138).

Researchers have not reported in the literature on BPD incidence among the larger Iow-birthweightinfants, but the authors of a recent review article about BPD estimated its incidence at 10 to 20 percentamong infants with RDS who receive mechanical ventilation and survive (71). OTA adopted the low endof this estimate, 10 percent, in calculating the BPD incidence among babies weighing 1,501 to 2,500 g.

Only a relatively small proportion of the babies developing BPD are obvious candidates for technology-dependent home care. In their eight-center study, Avery and colleagues found that about 4 percent of in-fants weighing less than l,500 g at birth still needed supplemental oxygen at 3 months of age (althoughthe range among institutions was considerable) (11). BPD can take mild, moderate, or severe forms, andinfants are weaned from the ventilators and/or oxygen support after variable periods of time.

In the future, the incidence of BPD will likely decline, although extremely Iow-birthweight babies sus-ceptible to BPD—including babies weighing less than 500 g at birth-are increasingly surviving (162). Refine-ments of existing techniques and newly introduced neonatal technologies might substantially reduce BPDin premature infants within a few years.

Table 10.-Estimated incidence of Bronchopuimonary Dyspiasia, 1984

Birthweight U.S. births U.S. neonatal Neonatal Percent survivors Total infants with(grams) (1964) mortality (1960) survivors with BPD BPD per year

500-1,500g 39,045 43.1 % 22,217 33% 7,332l,501-2,500g 202,606 2.4% 197,743 1 %a

1,977Total . . . . . . . . . . 9,309

%oldberg and Bancalari (71) estimate that approximately 10 percent of Inftmt$ with ra$pkatoty dktreas syndrome (RDS) get BPD. If approximately 10 percentof all surviving infants get RDS (174), then approximataty 1 parcent of etl suwlvora oat 8PD.

SOIJRCE: Office of Technology Assessment, 1S87. Numbers of U.S. btrtha from U.S. DepertrneM of He8ith and Human Services, Public Health Service, NationalCenter for Heafth Statistics, “Advance Report of Finai Netaflty Statlstica, 1SS4, Table 24,” W@ St8t/st/cs Repofl 35 (4, SUPP.):JUIY 18, 1988. Neonatalmortality rates from U.S. Department of Health and Human Sewices, Publk Heetth Service, Oantere for Disease Control, “National Infant MortalitySurveillance (NIMS),” unpublished tablea, May 1988. BPO Incidence rates approximated from M,E. Avary, Boston Children’s Hospitai, Boston, MA,personal communication, July 23, 1988; and ranges presented in J.D. Horbar, “A Multicenter Suwey of 28 Day Survival end Suppiementat OxygenAdministration in infants 701-1500 Grams,” paper presented at the Rosa Laboratories Special Conference on Topics on Neonataiity, Washington,DC, Dec. 7-9, 1s88.

30

Box C.–Chang@g Tedmolo& in the Neonatal Intensive Care UnitPreven~ the coqdidbts of mechanical ventilation in newborns has been a focus of research for

some time. Changes in the way artificial ventilation is dministered to newborns have been an importantpm of that rew~ (~,~~). However, medkd _ and the use of t@mokwY a vw *h -OIWperinatal centers. Avery and her colleagues surveyed eight centers in 1983 and 1%!4 for their experiencewith chronic lung disease in infants weighing 700 to 1,S00 g, The msearc hers found that some institutionsdid significantly better than others, and that routine mmagemmt techniques used for the very small in-fants might explain the differences (11). Refinements in existing techniques may thus hold promise for re-ducing the development of brcmchopuhmmary dysplasia (BPD) in ventilated infants in the future. Somenew technologies, such as the high frequency ventilator (which delivers multiple smaII breaths instead ofslower, larger ones) and extracorporeal membrane oxygenation (essentially a heart-lung machine for new-borns with severe asphyxia), may also have some effect.

Other technologies under investigation focus on preventing respiratory distress syndrome (RDS), therecurser of BPI) in most infants. The administration of steroids to mothers in preterm labor in order toP

accelerate infant lung maturation has been used and studied for 16 years (9), but concerns about the long-terrn effects of the therapy have prevented its routine use (43). Recent large-scale studies are somewhatcontradictory buts- that the technology can rwha th~ incidence and severity of RDS, and may im-prove survhd in some very prenwture infants, with no evidence of negative Iong-tem e-s (43,174).Even if antenatal steroid therapy does become generally accepted as useful, however, it will have severallimitations. It clearly does not work for all babies. And in addition, because the therapy must be initiatedat least M hours before delivery in order to be effective, many women in preterm labor cannot be candi-dates for its use.

Treating surfactant defkiency by administering artifidal or natural (animal lung) surfactant to the lungsof very premature babies at or soon after birth has the potential to greatly reduce the incidence of severeRDS. The bask chemistry of lung surfactant has beeh known for a lo% time, but research is ongoing re-garding the best mixture, the optimum dose, and the timing and frequency of administration. At least fiverecent clinical trials testing natural surkctants document that surfactant-treated infants have less severeRDS (and, pl’W3Wi ably, less Mkelihood of developirqjllP13) than control infants (66,89,100). Studies withartificial& produced surf~ant, on *he other hand, have shown essentially no benefit to respiratory func-tion (76,183). ‘

Lqe+cale, multi-center trials are being uncktaken h Ihwope, Jq=, and the United states to COn-tinue to test surfactant experimentally. It is possibl~ that surfactant therapy could become generally avail-able for preterm babies within 2 to 5 years (10,143).

clinics around the country, 43 percent of venti-lated patients were found to have permanenttracheotomies (35). This difference may repre-sent an increasing willingness over time to treatend-stage muscular dystrophy patients aggressively,

The 1985 clinic survey also revealed that 24 per-cent of the responding physicians did not providerespiratory support systems to individuals withdegenerative neuromuscular disorders, while 33percent prescribed such supports routinely and theremaining 42 percent provided them only underspecialized circumstances (35). The researchersfound no standard patient-selection processes orestablished protocols for respirator use. If all phy-sicians applied the same criteria for ventilator

support as those physicians who prescribe thistreatment routinely, the number of people withend-stage muscular dystrophy using mechanicalventilation could triple.

Other differences in medical practice and socialattitudes may also be reflected in rates of tech-nology dependence. Some centers now treat new-borns weighing less than 500 g (1.1 lbs) aggres-sively, although these babies are highly unlikelyto survive (74). The promulgation of “Baby Doe”regulations and accompanying social attitudes hasprobably had at least some marginal influence onphysicians’ decisions to treat severely prematureor disabled infants aggressively in the UnitedStates (14.5). It is likely that the trend toward ag-

31

gressive treatment of very small newborn babieswill continue, at least in the short term.

Whatever the reasons, differences among med-ical centers and geographic areas do exist. As hasbeen noted by one clinician, “There are somecenters that just don’t seem to have children thatrequire home ventilation” (91 ). Whether this dif-ference is due to more aggressive attempts to weanventilated children in those centers, or other fac-tors that result in fewer infants who both surviveand require long-term ventilation, is unclear.

Finally, improved access to funding for non-hospital long-term care, particularly home care,could result in more technology-dependent chil-dren and their more visible participation in soci-

CONCLUSIONS

Defining the population of technology-depen-dent children is a necessary first step for bothenumerating the population and providing healthcare benefits directed at this population. That ar-riving at such a definition is not easy is clear fromthe fact that, after 5 years of public debate aboutthe issue, no satisfactory definition exists. Froma clinical perspective, the crucial distinguishingcharacteristic of these children is that they requirespecial equipment and an intense level of medi-cal services that are beyond the normal capabil-ities of untrained families. From an insurance pro-gram perspective, the crucial characteristic ofthese children is that it may be possible to carefor them more appropriately and less expensivelyat home if the funding and services are made avail-able. Although the two populations described byeach of these characteristics overlap considerably,in that both require some form of hospital careif services in other settings are unavailable or un-affordable, they are not identical.

OTA’s estimates of the size of the medically sta-ble, technology-dependent child population at anyone point in time, based on available sources ofdata for four alternative groups, are as follows:

1. Approximately 680 to 2,000 children peryear in the United States are substantially orcompletely dependent on mechanical venti-lation.

ety. It is possible that providing opportunities forchildren to be in home or home-like settings, com-bined with enhanced funding for long-term care,eliminates some of the social, financial, and med-ical disincentives to initiate and maintain long-term technology dependence. The three Stateswith the highest identified prevalence of venti-lator-dependent children all have aggressive homecare programs to serve such children; North Caro-lina, a State with few ventilator-dependent chil-dren, does not. More families may consider itworthwhile to maintain the life of a terminallyill child as long as possible if they can afford totake the child home, and more physicians mayconsider it appropriate medical care to prescribelong-term ventilation for children.

2.

3.

4.

Approximately 600 to 9,000 children requireintravenous therapy each year, including 350to 700 children dependent on intravenousnutrition.Approximately 1,000 to 6,000 children aredependent on some other kind of device-basedrespiratory or nutritional support, such assuctioning, tracheotomy care, oxygen, ortube feeding. The cumulative number of chil-dren in the above three groups is betweenapproximately 2,300 and 17,000 technology-dependent children per year.Expanding the definition of technology-de-pendent children to include children requir-ing apnea monitors and kidney dialysis wouldincrease the size of the technology-dependentchild population to between approximately11,000 and 68,000 children per year. Add-ing children requiring urinary catheterizationand colostomy/ileostomy care to this pop-ulation could raise the upper bound of thisestimate to as high as 100,000 children.

There is no evidence of overall increase in theincidence of most severe chronic disabling con-ditions. However, the number of technology-depen-dent children appears to have been increasing overthe past ten years, due primarily to increased sur-vival of very-low-birthweight infants, who havea high incidence of chronic lung disease, and in-

32

creased survival of children with certain inheritedand congenital chronic disorders, particularly cys-tic fibrosis, muscular dystrophy, and congenitalanomalies. AIDS is also increasing the technology-dependent population. This trend is likely to con-tinue for several years. In the long run, it is notclear whether the trend will continue, level off,or represent a “bump, “ analogous to the polio andrubella epidemics that produced many severelydisabled children earlier this century.

Factors that may increase the size of the popu-lation include:

●

●

●

●

●

●

wider acceptance of medical practices suchas aggressive treatment of individuals withend-stage disease, severe trauma, and severenewborn disabilities; and less aggressive at-tempts to wean ventilated children;consequent increased survival of childrenwith conditions that would be fatal if not ag-gressively treated, and are highly likely toresult in technology dependence if they aretreated;new severe chronic diseases, such as AIDS;lessened acceptance of abortion;sufficient financing to encourage aggressivemedical practices;technologies that improve survival outcomesfor burn patients or transplant patients, butat the cost of extended recuperative care; and

● increases in the apparent size of the existingpopulation due to new opportunities to ob-tain funding for home care.

Factors that may decrease the size of the popula-tion

●

●

●

●

It

include:

improved prenatalchronic disease;technologies thatdent prevention),

diagnostic tests for severe,

can prevent trauma (acci-premature births (prena-

tal care), and specific neonatal disorders;technologies that can lessen the intensity orduration of technology dependence, such asoral insulin or transplants to correct diabetes;andincreased acceptance of palliative care for fa-tal disorders.

is likely that the expansion of current medi-cal practices tending to increase the size of thetechnology-dependent population will continue inthe short run, perhaps for a decade or longer.Thus, the population size estimates given in thischapter will probably be lower than the actualpopulation size within a short time. In the longerterm, opposing factors—most significantly, tech-nologies to prevent premature birth or its compli-cations—may eventually lead to stabilization oreven a decrease in the technology-dependent pop-ulation.

.- —

,/

Chapter

ComparingcostsEffectiveness and

of Home v. Hospital Care

Chapter 3

Comparing Effectiveness and Costsof Home v. Hospital Care

INTRODUCTION

The purpose of this chapter is, first, to sub- chapter presents evidence regarding the costs ofmarize existing evidence on the effectiveness of home v. hospital inpatient care. More discussionhome care for technology-dependent children and of alternative settings of care can be found in ap-the components that may affect it. Second, the pendix D.

EFFECTIVENESS

Evidence of the Effectiveness ofHome Care

There is a broad assumption in the home careliterature that, compared to institutional care,home care is both more desirable and more ef-fective in promoting the mental, emotional, andphysical health of children. This assumption seemsa reasonable one for many, perhaps most, tech-nology-dependent children. There is little reasonto believe that when a family wants its child athome, and adequate medical services and supportare available, the home is not an effective settingof care.

Unfortunately, concrete evidence on how muchmore effective home care is, or the circumstancesunder which it is as effective, is lacking. Thereis also virtually no evidence in the literature onthe relative effects, either medical or psychologi-cal, of care across various alternative settings forthe technology-dependent population, includingalternatives such as board and care homes.

A few studies have attempted to document thebeneficial effects generally ascribed to pediatrichome health care, or to special services associatedwith this care. The largest and most rigorouslydesigned of these studies was performed as partof an evaluation of the Pediatric Home Care (PHC)unit at the Albert Einstein College of Medicine-Bronx Municipal Hospital in Bronx, New York(150). In that evaluation, chronically ill childrenwere randomly assigned either to the PHC unit

or to standard outpatient clinic-based care (withno special services). Standard inpatient care wasnot compared. PHC services included delivery ofhealth services, patient and family training, serv-ice coordination, and patient advocacy. Serviceswere delivered by an interdisciplinary team whosecore was the pediatrician, pediatric nurse practi-tioner, and family, but which also included phy-sician specialists, a psychiatrist, a social worker,and a physical therapist.

Children enrolled in the PHC program had bet-ter psychological adjustment, families reportedsignificantly higher satisfaction with care, andmothers showed reduced psychiatric symptomscompared to the control group (150). There wasno difference between the experimental and con-trol groups in the impact on the child’s functionalstatus, or in the impact of the child’s illness onthe family. The researchers in this study hypoth-esized that the social and psychological supportoffered by the program balanced the family’s bur-den of caring for the child at home, resulting inno change in net impact on family activities, struc-ture, or burden.

An evaluation of the impact of the home carecoordination and support programs for ventilator-dependent children in three States (Louisiana, ll-linois, and Maryland) is currently being conductedby researchers at the University of Chicago (3).Results from this evaluation are expected in late1987.

3 5

36

Numerous parents and clinicians have assertedfor some time that home care is superior to hos-pital care for medically stable, technology-dependent children. Children cared for in thehome are believed to make faster medical and de-velopmental progress, and have better psychoso-cial development, than children in the hospital(69,95,141,156). There are no published objectiveinvestigations of this hypothesis. It is a difficultone to test, since one cannot compare a child’sprogress at home with his or her progress in thehospital during the same time period.

The rate of chronic rehospitalization of childrenplaced in family homes can be considered a meas-ure of effectiveness of home care. It is to be ex-pected that technology-dependent children willhave occasional brief hospitalizations for acuteepisodes of infections and illnesses and for evalu-ation, but the ability of a program to reduce oreliminate avoidable rehospitalizations and emer-gency room visits could be a useful measure ofits effectiveness.

Factors Influencing the Quality andEffectiveness of Home Care

Two sets of factors influence the effectivenessof home care: those internal to the home envi-ronment, including parental abilities and attitudes;and those external to the home, such as the avail-ability of trained professional nurses and the qual-ity of equipment available.

Internal Factors

A crucial condition for effective home care isthat the family wants the child at home, and thatit is willing and able to help care for the child (1)or to accept and support a professional, full-timecaregiver into the household. Parents have ex-pressed strong desires to have their child at home,and they commonly take over much of the child’snursing needs (60,156). They may become soproficient in providing the necessary nursing thatthey train some of the the professionals who as-sist in their child’s care (156).

However, there are families for whom full-timelong-term home care may not be the best alter-native for either the child or the family. The fam-

ilies of chronically ill children can sometimes en-counter severe and ongoing psychological andemotional stress (54,173,180). Some families maybe simply unable or unwilling, for physical, psy-chological, or financial reasons, to cope with in-tensive home care for the child. Other familiesmight want to have the child at home, but mightneed a long adjustment period, or might need tofeel confident that respite or long-term care out-side the home is available if the stress becomestoo great. In a few cases, a parent might be will-ing to care for a child but be unable to do sosafely. These factors lead to less effective homecare and the need for alternative settings of care.

External Factors

The availability of services in various settingsis also crucial to the quality and effectiveness ofhome care relative to institutional care. Home caremay not be more effective than hospital or otherinstitutional care if the appropriate range of serv-ices are not provided. If a child can receive ther-apy or other vital services in the hospital but notat home due to lack of insurance coverage, homecare is likely to be relatively ineffective. Con-versely, if a child receives more intense, individu-alized therapy and education at home, home careis likely to be more effective than hospital care.

Professional nursing skills are a particularly im-portant factor in the effectiveness of care, butskilled nurses are not always available for homecare. In an acute-care hospital, nursing servicesfor a technology-dependent child are most likelyto be provided by a registered nurse (RN), andoften by an RN with extensive training in pedi-atrics or intensive care. In home settings, on theother hand, professional nursing is much morevariable. Some home care agencies specialize in“high-technology” home care or pediatric homecare. Others, however, may not have nurses (what-ever their certification level) trained to providethe specialized care needed by technology-depen-dent children. Most licensed practical nurses (LPNs)and many RNs, for example, are not trained tooperate ventilators and provide respiratory care.

Third-party payers may place restrictions onthe services that a nurse with a particular set ofcredentials can provide, although there is consid-

37

erable controversy over the relationship betweencredentials and quality of care. In New Mexico’sMedicaid program, for example, RNs may pro-vide any respiratory- or nutritional-support homecare; LPNs can provide most such care, but notventilator-related care or parenteral nutrition; andparaprofessional attendants may not provide anysuch complex care (34). In Louisiana, on the otherhand, families needing assistance at home relyheavily on trained paraprofessionals even forventilator-related care (97). These different con-ventions are associated with differences in serv-ice availability, but they also reflect different im-plicit evaluations of the relationship betweencredentials and quality. The differences in thequality of home care provided by family mem-bers, paraprofessionals, LPNs, RNs, and specialtynurses has not been addressed explicitly in theliterature. Undoubtedly, the level of skill-specifictraining is an important variable, regardless of thegeneral certification of the provider.

Certain quality issues arise at home that rarelyarise in the hospital because of established rou-tines or protocols in the latter, but not the former,setting. For example, home chemotherapy intro-duces substantial concerns regarding the use anddisposal of very toxic chemicals (87), Widespreaduse of and dependence on home ventilators intro-

COST

Cost considerationsrole in the evolution of

have played a substantialhome care for technology-

dependent children. When the Federal Govern-ment first waived certain Medicaid rules to per-mit hospital-bound, technology-dependent chil-dren to receive Medicaid payment for equivalenthome services, it did so on two grounds: that thehome was equal or preferable to the hospital asa setting for a child’s care and development, andthat home care would be a fraction of the costof hospital care to Medicaid. These criteria, andparticularly the second, have endured. From 1981,when the first exception was granted, to 1986,Medicaid (and other third-party payers) has con-tinued to require a showing of program cost sav-

duces concerns regarding the quality of mainte-nance of equipment and issues of how the widelydispersed users are informed about potential me-chanical defects (118).

Concerns about monitoring the quality of homehealth care have been raised before (160). Theseconcerns are particularly relevant in the contextof widespread emphasis on early hospital dis-charge. In a number of States, Medicaid pays hos-pitals a preset rate per discharge regardless of theactual length of hospital stay of a child (101).Once home care for technology-dependent chil-dren is widely accepted in an area, and fundingbecomes available, hospitals may be very reluc-tant to keep these children, whose length of stayis generally quite long. If the third-party payer’sinterests also lie in encouraging home care, par-ents could be forced to take a child home, possi-bly with insufficient services, before they are ade-quately prepared. Or, parents could be forced intohome care when they are unable and unwillingto provide the service at all. Anecdotal reportssuggest that, in some cases where home care isa funded option, hospitals or payers are indeedputting pressure on families to take these childrenwhen the families are not ready to do so (104,120).These circumstances could have serious negativeimplications for the quality of home care.

ings before paying for home care for many tech-nology-dependent children. Of course, programcost savings and social cost savings are not nec-essarily the same.

This section first describes the components ofhome care costs—i.e., the factors that influencethe costs of home care for different technology-dependent children. It then presents the issues andproblems involved in comparing the costs of careacross alternative settings. Finally, it presents ex-isting evidence from the literature and from homecare programs regarding comparative average to-tal costs of technology-dependent children acrosssettings.

38

Components of Home Care Costs

Startup Costs

Startup costs are one-time costs that are usu-ally incurred before the child is placed in thehome. They include the costs of home improve-ments, major equipment, and caregiver training.

Home improvements are often a necessary pre-requisite to home care, particularly for ventilator-dependent children. Home modification needs caninclude:

●

●

●

●

wiring and other electrical work;construction (storage and preparation space,wheelchair ramps, equipment accommoda-tion);special needs (e.g., a generator for emergencypower in rural areas) (178); a n dgeneral upgrading that requires the family tomove (e.g., moving from an unsanitary apart-ment or one inaccessible to a wheelchair toother housing where appropriate modifica-tions can be made).

The costs of necessary home modifications canvary substantially; in the first 3 years of Loui-siana’s home care program for ventilator-depen-dent children, home modification charges rangedfrom $0 to $13,500 (97).

Equipment can be a major component of start-up costs, particularly for children on ventilatorsor oxygen. (If the child is not expected to be tech-nology-dependent at home for long, or if an in-surer will only pay for rented equipment, muchof the child’s equipment will be rented rather thanpurchased. ) Special equipment for a ventilator-dependent child might include two ventilators (aprimary and a backup ventilator), an emergencybattery, an oxygen tank, a suction machine, anebulizer (to deliver aerosol medication), a man-ual resuscitator, and an infusion pump (to con-trol the administration of nutrients). The childmay also need other supportive equipment suchas a wheelchair, a commode, a special bed, andvarious other adaptive furniture and devices. Achild dependent on intravenous feeding, by com-

‘Extra space may be needed, for instance, for preparing nutrient

so lu t ions or for c leaning and s ter i l iz ing equipment

parison, might have startup equipment costs thatare a negligible proportion of total home costs.

Training in the necessary medical procedures,which may take days or weeks, is a vital first stepfor families. Even if they will have professionalnursing help, they must learn to perform the nec-essary procedures as a guard against emergencies(e.g., resuscitation) or in situations where theregular nurse might be incapacitated or absent.Table 11 shows a list of skills the family of a childon respiratory support or infusion therapy (intra-venous drugs or parenteral or enteral nutrition)might need to learn.

Training time and costs differ by training in-stitution, by level of care the child requires, andby the family members’ ability to assimilate in-formation and perform the necessary tasks. Insti-tutions training families in respiratory care maydo this in several days of intensive training (13).Or, they may gradually encourage family mem-bers to provide care while the child is in the hos-pital, perhaps requiring that the family providetotal care for 48 hours before discharge (57,65).Training in intravenous techniques can also betime-consuming; one program reported a 3-weektraining period for home patients (181).

Ongoing Supplies and Services

Supplies are often purchased monthly throughthe hospital or home care agency. Table 12 de-tails the supplies needed for intravenous therapyand tube feeding. The highest ongoing supplycosts are probably incurred by children requir-ing total parenteral nutrition, because their nu-trient formulas—consisting of “pre-digested” fat,carbohydrate, and protein solutions—are individ-ualized, require special handling and storing, andhave expensive components. In a 1982 survey,average charges for nutrition supplies and solu-tions were reported as $3,059 per month for hos-pital-supplied solutions and $4,615 per month fornonhospital-supplied (possibly pre-mixed) solu-tions (122). The range of charges, however, wasvery wide; the highest charges were nearly dou-ble the average in both categories.

Ventilator-dependent children also have highongoing supply costs. As is shown by the child

39

Table 11 .—Checklist of Respiratory and

Respiratory skills:1. The disease process:

● lung disease and Its treatment● short- and long-term prognosis and goals

2. Pulmonary hygiene measures:

Infusion Skills for Home Care Patients and Families

6. Educational and diversional activities:● encouraging chiId self-care● sedentary activities

7. Access to services:● nurses

● avoidance of infection (hand-washing and sterile ● physicianstechnique) . respiratory equipment suppliers

● adequate systemic hydration ● therapists● chest physiotherapy procedure ● emergency power● steriIe suctioning procedures ● other services● tracheostomy care procedure Infusion skillsb

● tracheostomy tube cuff care procedure 1. Understanding of components of home lnfusion● signs of airway infection and cor pulmonae that therapy:

should be reported to the doctor 2. Sterile procedures:3. Use and maintenance of the equipment: ● caring for medications and solutions

● daiIy maintenance of the ventiIator ● preparing medications and solutions for infusion● oxygen use, abuse, and hazards 3. Infusion techniques:● cleaning and changing of ventiIator circuits ● measuring components, using syringes, bottles,● resuscitation bag use and cleaning and bags● suction machine use and cleaning ● setting up the infusion

4. Nutrition counseling:a● starting the i n fusion

● maintenance of ideal body weight ● discontinuing the infusion● special dietary restrictions as needed ● operating the infusion pump

5. Physical therapy: 4, Recognizing complications:● ambuIation, where possible ● of the catheter● general strengthening exercises ● of the infusion● relaxation exercises ● of the medications

aThls table IS based on one for adults Most In fants on ventilators requl re enteral lube feeding for at least the beg[nnlng weeks or months Thus the ‘aml I Ies of thesechildren must also be tral ned In I nfus!on skills relatlng to tube feeding and care

bNot all SKI115 are applicable to all k!nds of lnfuslon therapy

SOURCES List of respiratory skills adapted from J Feldman and P G Tuteur Mechanical Ventilation From Hospital Intensive Care to Home Heart & Lung 11 (2)162 165March April 1982 Infusion skiIls adapted from Blue Cross and Blue Shteid Association In fusion Therapies in Home Health Care (Chicago, I L BC/BSA, Janu-ary 1986)

Table 12.—Supplies Needed for Four Home Infusion Therapies

Therapies for” which supplies are needed

Parenteral Enteral IntravenousSupplies nutrition nut ri t ion antibiotics

Intravenous catheter . ... ... . ... . x - xIntravenous tubing . ... x xMedications . . . . . . . . . ... . x xNutrient solutions (e.g., lipids) ... . . . . . . ... . . . . . xIn t ravenous so lu t ions (dex t rose or sa l ine) . , . . . . x xInfusion pumps . . . . . . . . . ... ... . x x xHeparin lock and dilute solution ... . . ... ... ... ... x xNeedles and syringes . . . . . . ... . . . . . . . ... . . . . . . . . x xDressings (gauze and tape or transparent . . . . . . . . . . . . x x xNasogastric, gastrostomy, jejunostomy tubes ... . . . . . . xEnteral bag and tubing ., . . . . ... ... . . x

Cheroot herapy

xxx

xxxxx

Enteral feeding preparations . ... . . . . . ... . . . . . . . xSOURCE “Blue Cross and Blue Shield Association Infusion Therapies in Home Health Care (Chicago IL: BC/BSA, January 1986)

40

whose reimbursable expenses are detailed in ta-ble 13, monthly costs for major supplies may to-tal over $1,600 (116). Medications, special nutri-ent solutions, and equipment maintenance canadd substantially to this cost.2

Nursing needs are highly varied and, for manychildren, are the most expensive component ofhome costs. Outlays for nursing care are inverselyrelated to the amount of unpaid care that the fam-ily is willing and able to provide. As has beennoted (148), most of the reduction in charges re-ported for ventilator-dependent patients at homeresults from shifting the burden of nursing costsfrom the payer to the family. Similarly, a signifi-cant part of the reduction in home charges ob-served by numerous intravenous therapy pro-grams (96,130,132,151) is due to the fact that thepatients in these programs receive little or noprofessional nursing at home.

‘Inadequate or poorly coordinated equipment maintenance canpresent a major problem to a home care program and a significantexpense to families or third-party payers (104). Programs are stillaccumulating experience in working with manufacturers and sup-pliers to minimize problems.

Table 13.—Sample Home Respiratory Care Costs ThatWere Reimbursed by a Third-Party Payer, 1985

one-time Durchase of eauipment Unit cost

Suction equipment. . . . . . . . . . . . . . . . . . . . . . . $ 714.29Manual resuscitator. . . . . . . . . . . . . . . . . . . . . . 157.31Emergency 12V battery . . . . . . . . . . . . . . . . . . 70.00Heating nebulizer. . . . . . . . . . . . . . . . . . . . . . . . 324.00

Total one-time cost . . . . . . . . . . . . . . . . . . . . $ 1,265.60

Monthly services and supplies Monthly cost

Home assistance:Nursing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $ 7,320.00

Rentals:Backup ventilator . . . . . . . . . . . . . . . . . . . . . . 200.00Ventilator. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450,00Suction device . . . . . . . . . . . . . . . . . . . . . . . . 50.00Apnea monitor . . . . . . . . . . . . . . . . . . . . . . . . 200.00Oxygen system . . . . . . . . . . . . . . . . . . . . . . . . 130.00

Supplies:Ventilator tubing . . . . . . . . . . . . . . . . . . . . . . 100.00Oxygen masks . . . . . . . . . . . . . . . . . . . . . . . . 55.20Liquid oxygen . . . . . . . . . . . . . . . . . . . . . . . . . 816.00Nebulizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.60Sterile water . . . . . . . . . . . . . . . . . . . . . . . . . . 79.50Tracheotomy tubes . . . . . . . . . . . . . . . . . . . 143.70Suction catheter w/ gloves . . . . . . . . . . . . . . 315.00Cardiac leads . . . . . . . . . . . . . . . . . . . . . . . . . 10.00

Total monthly cost. . . . . . . . . . . . . . . . . . . $9,993.00SOURCE’ M Mikol, SKIP of New York, Inc , New York, NY, personal communica.

tion, June 1986

Actual nursing care expenditures for a technol-ogy-dependent child depend on three factors: thecomplexity of care required, the amount of paidnursing care required (e. g., 3 hours per day v. 24hours per day), and the certification level of thenurse. In general, 24-hour ventilator-dependentchildren with tracheotomies need the most con-stant and complex nursing care. Other children,such as those receiving intravenous nutrition ortherapies, may need complex or intensive care forseveral specified hours per day; or they may needless complex care but need it constantly in orderto avoid a life-threatening event (e.g., children re-quiring trachea suctioning).

Professional home nursing costs can be substan-tial and vary considerably with the certificationlevel of the nurse. In New Mexico, for example,Medicaid pays $17 per hour for RNs and $13 perhour for LPNs, slightly more than the charges ofthe lowest priced home nursing agency in thatState (34). If all care were provided by profes-sional nurses, the monthly Medicaid payments fora child requiring an 8-hour professional nursewould range from $3,120 to $4,080; payments fora 24-hour nurse would range from $9,360 to$12,240. If Medicaid paid agency charges, as someinsurers do, payments could be as high as $18,000per month for a 24-hour RN.

Other factors can also affect nursing costs. Forexample, the need for an escort to accompany anurse to work in a high-crime area would raisecosts. Or, nurses might demand higher pay whenworking in such areas.

Specialized therapy is needed by most technol-ogy-dependent children in order to progress.Speech therapy, physical therapy, and occupa-tional therapy are commonly provided to thesechildren in one or more weekly visits. Basic res-piratory therapy, however, often becomes a re-quired skill of the primary caregiver, both becauseit is often required so frequently and becausemany third-party payers do not pay for home res-piratory therapy visits.

Outpatient services to technology-dependentchildren in home care can include regular visitsto one or more specialty physicians (e.g., a pedi-atric pulmonologist), frequent laboratory work-ups, and visits to a local pediatrician or family

41

physician who oversees thecare. ” A local physician who

child’s “well childis familiar with the.

child’s health status and medical needs is particu-larly important for families who live a great dis-tance from the specialty clinic or tertiary care hos-pital. Even with these outpatient visits, childrenin home care may have frequent hospitalizationsfor in-depth assessment, initiating new treatments,respite care, or complications or sudden emergen-cies relating to their conditions. These rehospitali-zation are often overlooked when comparinghome and institutional care costs.

Although outpatient visits and laboratory testsare often relatively minor compared with otherhome care costs, they are ongoing and by nomeans negligible. For example, the individuals in-cluded in a 1982 survey of home parenteral nutri-tion programs averaged $23 to $32 per month inphysician and clinic costs (for patients served byhospital- and nonhospital-programs, respective-ly), with extremes ranging from $6 to $83 (122).Laboratory work for stable patients in this sur-vey averaged $69 and $82 per month for hospital-and nonhospital-supplied patients, respectively,and ranged from $4 to $350 per month (122).

Transportation to outpatient services and toschool is required by technology-dependent chil-dren in home care and can be a substantial costfor some children. Children receiving intravenousantibiotic therapy or chemotherapy may requirephysician visits as often as twice a week (130).For stable ventilator-dependent children, visitsmay be less frequent—weekly or monthly—buttransporation costs may be very high because thetransportation vehicle must be spacious enoughto accommodate respiratory equipment and awheelchair. Children served by the Louisianahome program for ventilator-dependent childrentravel up to 385 miles round-trip for physician andclinic visits (97). Of 23 children whose transpor-tation needs were detailed by this program as ofJune 1985, 7 used the family car; 3 used specialvans purchased for that purpose; 3 used publictransportation; 1 child used an institutional van;and 8 children used ambulances.3

‘Three of the children using ambulances did so for emergency orlnterhospital transfer purposes only, while they resided in institu-tions or nursing homes. One other child required an ambulance twicefor tracheotomy tube changes. Of the remaining 4 children usingambulances for transportation, 3 eventually died. No other formsof transportation are recorded for any of these 8 children,

Respite care is care that gives the family somerelief from ongoing nursing care. It may be in theform of an occasional professional nurse or otherperson who provides care in the home. Or, it maybe in the form of a nearby hospital or other facil-ity that cares for the child while the family is onvacation or pursuing other activities. Respite careraises the immediate costs of home care, but itmay lower total costs if the assurance of occa-sional respite enables the family to provide mostongoing care.

Case management-coordination and oversightof the package of services provided to an indi-vidual—is a vital service to most technology-dependent children because of the multiplicity ofstartup and ongoing services needed, Case man-agement may be performed by a health care pro-fessional, such as the child’s pediatrician (112) ora specially trained pediatric nurse (129). Or, third-party payers may provide case managers (as partof an individual benefits management program)to ensure, first, that the appropriate mix of serv-ices are available to enable the child to receiveappropriate care at home; second, that those serv-ices continue to be provided as arranged; andthird, that the child’s progress is monitored, sothat appropriate changes in service are made.

Case management can sometimes minimize thecosts of care for children already being cared forat home. The Florida “Rural Efforts to Assist Chil-dren at Home” (REACH) program, a Medicaiddemonstration project, succeeded in reducing re-hospitalizations and emergency room visits ofchronically ill children in that program as com-pared to equivalent children not served by theprogram (129). The program, targeted at high-costchildren (not necessarily technology-dependentchildren) eligible for both Medicaid and Servicesto Children with Special Health Care Needs, usedcommunity-based pediatric nurses as case man-agers, coordinators, and consultants to help fam-ilies make the most appropriate use of medicalservices (129).

The most important aspect of case managementis that it can serve both the interests of the fam-ily and of the third-party payer. At present, it isclosely linked with the expectation of cost savings,and the process of ensuring those savings. If a pri-vate insurer expects to eventually pay out themaximum lifetime benefit, however, the incentive

4 2—

to provide case management may be weakened.There is also a danger that case management willbe superficial. Managers whose activities arelimited by the payer’s interests, or by a largecaseload, may be hampered by not being inti-mately acquainted with the child’s needs, or byignorance of important resources that could bemade available.

Issues in Comparing the Costs ofHospital and Home Care

1. Cost to whom?—The relative costs of hos-pital and home care depend on whether one isconsidering costs to Medicaid, costs to private in-surers, costs to the family, or total resource costs.Home care might often involve fewer total re-source costs than hospital care, but it might notcost the third-party payer less. For example, insome States, Medicaid pays hospitals a set rateper patient or per admission, regardless of the ac-tual length of stay for that patient or the servicesprovided. In these States, paying for adequatehome care for technology-dependent childrencould cost the State and the Federal Governmentmore than financing those children’s care in a hos-pital at the fixed rate. (Of course, the uncompen-sated costs to the hospital caring for children inthis situation are very high. )

Home costs may be lower to third-party payersthan institutional costs because of certain uncom-pensated costs incurred by the family when thechild lives at home. The two most notable exam-ples of such uncompensated costs are the costs ofbasic room and board, and the unpaid time of par-ents or other volunteer caregivers.

For some children, paid home services may bereplacing not institutional care but care previouslyprovided by the family at great expense. Theseservices reduce the costs to the family while in-creasing the net cost to the third-party payer.

2. Comparing equivalent costs.—Differentsources of information on costs of technology-dependent children use different concepts of cost.For example, one may report hospital charges,while another reports third-party payments to thehospital. Hospital charges themselves may not beequal to the actual costs associated with provid-ing a service. For example, one analysis of inten-

sive care unit (ICU) services found that the chargesin one hospital for room and board in the ICUwere only slightly more than one-half of calcu-lated actual costs to the hospital (77a). Even com-paring average hospital charges (say, per month)with average home care charges for a technology-dependent child can be misleading if the two in-clude different services. For example, hospitalcharges often include acute-care services such assurgery, while home charges do not. Inconsistencyin comparing the appropriate equivalent costs(i.e., maintenance and recuperative treatmentcosts) in each setting is a major problem in theliterature.

3. Biased sources of data.—Most third-partypayers offering intensive home care services re-quire that it be cheaper for them to pay for carein this setting. Consequently, most data on tech-nology-dependent children served at home will,by definition, show that home care is cheaper,There may be some children who are not dis-charged home because it will not be cheaper tothe payer, but since these children are not servedat home their presumably higher home costs arenot recorded.

4. Different costs for different children. -In thehospital, costs depend largely on medical need andon the physician’s judgment and style of medicalpractice. At home, however, the costs of care varynot only with the type and severity of disability,but also with the family and home environment.Factors such as the ability of family members toprovide most nursing care, and the extent of homecare renovations needed, have great impact oncosts of home care.

5. Cost in which setting? —Given sufficientequipment and services, many technology-depen-dent children could, if necessary, receive care inany of a number of settings that are intermediatebetween the acute-care hospital and home (seeapp. D). In some urban areas, several of these in-termediate options may actually be available; inother areas, none may be. Thus, while for onechild it may be appropriate to consider the rela-tive costs and effectiveness of care in a long-termrehabilitation hospital as well as home and acute-care hospital costs, for another child the rehabili-tation hospital may be unavailable, even if it theo-retically offers the lowest cost care.

43

The primary reason for variations in averagecost across institutional settings is that differentsettings offer a different mix of services and differ-ent levels of service intensity. For example, onesetting may cost more because it offers full-timerespiratory therapists on staff. However, cost isnot a direct measure of effectiveness or services.Costs and effectiveness are related (e.g., addinghome services to enhance effectiveness may in-crease costs), but, given no other information, onecannot deduce the relative effectiveness of differ-ent settings from their relative costs.

Evidence

There are two types of evidence on the costsof home care for technology-dependent childrencompared to care in other settings. First, there isevidence on the costs of care for children requir-ing ventilators and other device-based respiratorysupports. Since the evidence is sparse, some in-formation of the costs of care for ventilator-dependent adults is also included here. Second,there is literature regarding the cost of home carefor people receiving home intravenous nutritionor drug therapy. In this case, there is little evi-dence specifically for children, and the programsgenerally do not send home individuals whosefamilies cannot provide the necessary nursingservices. Published comparisons of home carecosts with the costs of nonhospital institutionalcare (e. g., nursing homes) are nonexistent fortechnology-dependent children in either group.

There is some evidence regarding costs of carefor children with less intensive needs, such asthose on dialysis or apnea monitors. It is not dis-cussed here because the relative costs of care forthese children is less of an issue than the costs ofcare for children requiring respiratory and nutri-tional supports.

Evidence on Relative Costs of Care forVentilator-Dependent Children

The evidence on the relative costs of caring forventilatordependent children in alternative set-tings is incomplete. Cost estimates are typicallybased on charges or payments and are availablefor small numbers of children enrolled in a par-ticular program or discharged from a particular

institution. All comparisons of home and hospi-tal costs for these children show that their careis almost inevitably less expensive at home. Thesecomparisons demonstrate that there are a consid-erable number of children for whom home careappears to be less expensive, often dramaticallyso. However, up to now no payer has reimbursedfor an exceptional 1evel of home care unless it isless expensive than hospitalization, so the venti-lator-dependent children now on home care areby definition less expensive to care for in thissetting.

Table 14 summarizes comparative hospital andhome charges for ventilator-dependent peoplefrom the literature. Because of the few reportsavailable, adults as well as children have been in-cluded, The figures in this table suffer from manyof the problems discussed above. The services in-cluded in hospital costs are generally much moreextensive than those included in home costs, andthe fact that the figures are averages disguises highvariations in the amount of paid nursing the pa-tients required. Despite these methodologicalproblems, however, table 14 still provides com-pelling evidence that for some technology-depen-dent children the home care charges are substan-tially less that those for intensive or intermediatehospital care,

The primary reason for this difference is sim-ple: when a child is cared for at home, the costsof housing and much of the nursing are borne bythe family rather than by the hospital or healthcare payer. If a child required 24-hour paid nurs-ing at $20 per hour (slightly higher than Medic-aid pays in New Mexico), monthly home nurs-ing costs alone would total $14,400—nearly asmuch as monthly hospital costs in several of thestudies in table 14. Thus, the extent to which to-tal home care charges are less than total hospitalcharges depends largely on the extent to whichthe family is able and willing to provide nursingcare and appropriate facilities for the technology-dependent child. The difference in charges alsodepends on whether less expensive nursing can besubstituted for more expensive care, and on themedical effectiveness of home care. Home care be-comes relatively cheaper if it speeds the time un-til a child achieves minimal dependence (or de-lays total dependence), and if it minimizes the

Table 14.—Summary of Comparative Average Monthly Charges Presented in the Literatureof Ventilator-Dependent Individuals in Hospital and Home Settings

— . , , - —- . - - . .— - , J

Source

Banaszak, et al., 1981 . . . . . . . . . . .

Burr, et al., 1983. .

Cabin, 1985 ., . .

Care for Life, 1985

Dorm, 1982 . . . . . . .

Feldman, et al., 1982 . . . . . . . . . . . .

Goldberg, 1983 . . . . . . . . . . . . . . . . .

Goldberg, et al., 1984 . . . . . . . . . . . .

Kahn, 1984 . . . . . . . . . . . . . . . . . . . . .

Lee, 1983 . . . . . . . . . . . . . . . . . . . .

Perry and Lierman, 1985 . . . . . . . . .

Sherman, et al., 1982 . . . . . . . . . . . .

Sivak, et al., 1983 . . . . . . . . . . . . . . .

Splaingard, et al., 1983 . . . . . . . . . .

Averagehospitalcharge

$15,469

19,613

29,113

25,063

19,071

17,500

27,435

24,590

22,000

40,590

51,517

40,332

15,600

15,000

Average Home COSts Incluoehome Number of Adults (A)/ some professional

charge patients children (C) shift nursing Comments

$3,535 2

2,388 6

5,201 1

1,853 —

389 12

5,704 2

7,310 2

7,425 2

5,400 26

5,490 8

7.361 2

1,943 9

1.760 10

775-16,900 47

A

c

c

A,C

cA

c

c

c

c

c

c

A

c

No

Not given

Not given

Not given

No

Yes (1 patient)

Yes

Yes

NO (if included, homecharges are $8,000)

Not given

Yes (1 patient)

Not given

Yes (4 patients)

Yes (patient represented byhigh home charge figure)

—All hospital charges (not just maintenance charges)are included In hospital figure.

Hospital charges include charges before patient wasmedically stable. Home charges include first monthhome.

Based on survey of State respiratory therapist repre-sentatives’ estimation of average charges.

Received nasal oxygen only (not ventilation).

—

.

—

Hospital charges from intermediate care unit.

Charges are averages during one year, not neces-sarily before initial discharge home. Two patientswere not hospitalized that year.

Hospital charges include surgery, acute care. Home“charges” Include estimated cost of some servicesprovided at no charge.

Charges presented here are averages of ranges givenin Iiterature report,

Hospital charges are estimated and include physiciancharges. Home charges do not include physiciancharges or initial costs. If startup charges areincluded, home charges rise to $1,894.

Hospital is a rehabilitation hospital. Home care costdoes not Include drugs.

SOURCES See references 13, 26, 28, 32, 42, 55, 68, 69, 92, 103, 128, 146, 147, and 149

4 5

frequency of rehospitalization for infections orother problems.

An analysis of more detailed charge and pay-ment data from children who have been servedby high-service home care programs is illuminat-ing. Maryland has computed the monthly hospitaland home payments for 25 children served by theState’s home care program for children requiringrespiratory support (93). For these children, themean third-party payment for the last month ofhospitalization (a proxy for typical hospital costsof a child when medically stable) was $24,715(range: $210 to $41,057). These children were hos-pitalized an average of 421 days before discharge.The average monthly home care payments forthese children, excluding the first month home,was $9,267 (range: $300 to $25,000). First-monthpayments averaged $9,798, or an average of $531per child for one-time expenses (93).

In Louisiana, average third-party payments forthe last month of hospitalization were compara-ble to those in Maryland ($25,995 for 19 children),but average monthly home care costs to the Loui-siana program were considerably lower (97).Home payments (computed for 21 children) aver-aged only $3,012 per month, excluding startupcosts. Children in this study were all at least partlyventilator dependent. Louisiana Medicaid, thesource of support for many of these children, doesnot pay for home shift nursing, which may ex-plain why the home care payments are so low.

Data from Illinois demonstrate the potentialdifferences between private payers and Medicaidin cost-savings of home care to third-party payers.For children whose care was compensated onlyby Medicaid, payments for the last month of hos-pitalization averaged $16,984, while monthlyhome care payments averaged $6,358 (104). Forchildren who had at least some private coverage,the payment for the last month of hospitalizationaveraged $26,616, while equivalent home carepayments were $6,922. Thus, while the monthlyhome care payments for children in these twogroups are roughly equivalent, the cost savingsof home care experienced by private payers (orthat would have been experienced had they cov-ered home care) was substantially greater thanthat experienced by Medicaid alone. (Illinois’

Medicaid program pays for a maximum of 45 hos-pital days per year, which limits program hospi-tal expenditures. )

Evidence on Relative Costs of Home v.Hospital Administration ofIntravenous Therapies

Intravenous therapies—prolonged parenteralnutrition and drugs—have followed the patternof renal dialysis and hemophilia treatment, inwhich treatments once received exclusively in thehospital have been adapted to the home. (Intra-venous drug therapy may also be administeredin outpatient settings. ) Drug therapy usually takesa few weeks, compared with the months or yearscommon with parenteral nutrition, but in otherways the two therapies have many similarities,

There is little information in the literature re-garding relative costs of care in different settingsfor individuals receiving chemotherapy, but allprograms reporting their experiences with homeantibiotic therapy have reported substantiallylower charges for home treatment compared tohospital inpatient treatment (table 15). In all casesin these programs, patients or their familiesadministered the infusion at home. Patient selec-tion was a vital component of these programs,because inadequately administered infusions canresult in ineffective treatment and rehospitaliza-tion. In one study, 40 percent of the patients re-ceiving prolonged antibiotic therapy were rejectedby the home care program for reasons includinginability to administer the antibiotic, poor fam-ily support, and poor motivation. Thus, patientswho would have required substantial professionalnursing to receive home treatment (had it beenavailable)—and thus might have had more expen-sive home care—could not participate in the pro-gram. The major differences in charges for homeand hospital treatment in the programs reportedin the literature are the need for patient training,planning, and clinic or nurse visits for home pa-tients; and the hospital room charge for hospi-talized patients.

Most individuals on parenteral nutrition, too,go home only after they or their families havemastered the techniques and can provide all homecare. One case has been reported in which a 58-

46

Table 15.—Comparative Charges for Home v. Hospital Administration of Intravenous Antibioticsas Reported in the Literature

Study Home charges

Antoniskis, et al., 1978 ... .. .$69 per day

Eron, 1984 ... ... ... ... ... .$10 per day in chargesincurred only by homepatients (training clinicvisits)

Harris, et al,, 1986 . . . . . . . . .$207 per day

Hospital charges

$243 per day

$170 per day in chargesincurred only by hospitalpatients (room charge incommunity hospital)

$428 per day

Rehm and Weinstein, 1983 .. .$1,652 per illness $7,380 per illness

Stiver, et al., 1978...........$40 per day $137 per day

NOTE All home lnfusions in these studies were administered by patients or their families

SOURCES See references 6, 50, 78, 132 and 151

year-old patient was discharged home on paren-teral nutrition under the supervision of full-timehome nurses (105), When this patient first wenthome, requiring 12 hours per day of parenteralnutrition administered by a nurse, her home carecharges were comparable to charges for hospitalcare. Her need for parenteral nutrition and theassociated nursing care diminished over time,however, lowering the home care charges (105).

The previously-mentioned 1982 survey of homenutrition programs found home care charges forparenteral nutrition that were roughly $3,400 permonth for hospital-supplied patients and $4,900per month for nonhospital-supplied patients (122).Individual programs have reported program costsof serving patients of approximately $1,800 permonth (in 1976-78) and approximately $2,700 permonth (in 1982-83) for the first year, when costsare highest (48,181). Even after adjusting for in-

CONCLUSIONS

There is no concrete evidence regarding the rela-tive effectiveness of home and hospital care,Home care is generally considered more effectivein promoting the psychological and emotionalhealth of children; hospital care is generally con-sidered more effective at providing medical and

Comments

Separate home and hospitalgroups studied.

Other charges (for servicesprovided to both home andhospital patients) are assumedequal.

Charges are for patients treatedinitially in the hospital, then athome. Hospital charges mayinclude surgery.

Hospital charges are estimates(patients all got home care).Charges are averages over 4years of the program.

Hospital charges are estimates(patients all got home care).

flation, these costs are probably substantially lessthan the costs of a patient receiving parenteral nu-trition in an acute-care hospital. A third study hasreported per-patient monthly home care chargesof $1,445, compared with hospital charges thatwould have been approximately $6,170 (23).

Thus, for both intravenous drugs and nutrition,the literature suggests that home treatment pro-vided under an organized program is substantiallyless expensive to the payer than care in an acute-care hospital, provided that patients are carefullyselected and can perform all necessary proceduresthemselves or with the help of family members.The literature also suggests, however, that homecare charges for adult patients who require sub-stantial professional nursing may approach hos-pital care charges. None of this literature specifi-cally addresses the relative costs of offering thesetherapies to children in different settings.

nursing care when necessary to promote physi-cal health. With the adaptation of sophisticatedtechnologies and care systems to the home, how-ever, there is no reason to believe that home set-tings cannot be equally effective at promotingphysical health. The desire of the family to have

—

47

the child home, however, and the availability andquality of these sophisticated technologies andcare systems, are critical determinants of effec-tiveness.

Cost savings to third-party payers have becomean important factor in their willingness to financeintensive home care. Such cost savings are likelyto exist for most children whose families are will-ing to bear some of the costs of home care, par-ticularly nursing costs, by providing those serv-ices at no cost to the payer. However, becausethe technology-dependent child population is sodiverse, and the nonmedical characteristics of thechildren and their families are so important a fac-tor in nonhospital health care costs, there are nomedical or clinical criteria that can be used to clas-sify children according to their expected cost sav-ing. Payers can ensure a high probability of over-all cost savings only on a case-by-case basis,where each child is evaluated and the likely totalcosts of care for that child in alternative settingsestimated. Still, some general factors that tend toincrease or decrease relative costs (and, some-times, relative effectiveness) can be identified.

1. Nursing Costs: Home care becomes morefavorable to third-party payers as familynursing can be substituted for professionalnursing, as less expensive professional orparaprofessional help can be safely substi-tuted for more expensive help, and as inten-sive nursing needs decline. Paid 24-hournursing may make home care as expensiveas hospital care.

2. Expected Duration of Dependence: The highstartup costs associated with home care canbe more easily justified if a child is expectedto be technology-dependent for a very longtime and ongoing home care costs are rela-tively low.

3. Family and Other Environmental Factors:Some children cannot or should not returnto a family home. If foster care cannot befound, there are often few options other thanacute-level hospital care for such children,although other appropriate options (e. g., re-habilitation hospitals, pediatric skilled nurs-ing facilities, or group homes) may be lessexpensive when available.

4. Availability of Services: Inadequate avail-ability of respite care or caregivers trainedin the appropriate skills can make hospitalcare the only viable option even where homecare might be less expensive if those servicesexisted. Or, choices in home services maybe so constrained that families and third-party payers may be forced to pay highprices for the services.

5. Substitution of Care: Home care will belikely to reduce third-party payments onlyif it can substitute for institutional care,rather than augmenting the care of childrenalready being cared for at home by their fam-ilies. In many cases, however, augmentedcare—particularly respite care and case man-agement—may be effective in increasing thequality of care and of life for these children,and may reduce rehospitalizations.

Chapter 4

Sources of Financing

Chapter 4

Sources of Financingl

INTRODUCTION

Technology-dependent children typically incurhigh annual treatment costs that may go on fora number of years. As chapter 3 has shown, thesecosts are substantial in any setting. For example,in Maryland’s program for children dependent onventilators and other respiratory supports, theaverage third-party payment for long-term carein the hospital was approximately $24,700 permonth in 1985 (93). Home care for such childrenwas about $9,3oo per month. Although the num-ber of children who become technology depen-dent each year is small, the financial burden ontheir families can be staggering. In the absence ofpublic or private insurance, other public funding,or private philanthropy and charity care, thesecosts would exceed or severely strain the resourcesof all but the wealthiest families.

The extent to which technology-dependent chil-dren have access to adequate financing for health

J I’arts (~t this chapter are adapted tr(>m a background p~per prc-p~r~~ under con t rac t t(]r OTA b} I{arrlette F(IX and R u t h }’oshpeent lt led, ‘ T e e hn[)lt~gy-[leptn~ient C“h]ldren \ AC ces> tt~ N!edlcalli1 l(~me Care Flnanc In~, ” Augu\t ] Q8~

care, and particularly for home care, depends onthree factors:

1.

2.

3.

the degree to which this population is cov-ered by private insurance or public healthcare programs,whether the insurance or program coverslong-term care at home for this population,andwhether the home care benefits are sufficientto finance most of the medical needs of thesechildren.

This chapter first discusses the extent of privateinsurance coverage among technology-dependentchildren and the adequacy of insurance for thosewho are covered. It then discusses the extent ofpublic payment for home care services across thespectrum of available sources of public third-partypayment. These sources include not only Medic-aid, but also diverse programs targeted to specificgroups or providing specific services.

PRIVATE HEALTH INSURANCE COVERAGE FORTECHNOLOGY-DEPENDENT CHILDREN

This section examines the adequacy of privatehealth insurance in covering the costs associatedwith the care of technology-dependent childrenin the hospital and at home. The question isframed as follows: what is the likelihood that atechnology-dependent child will have private in-surance coverage that is adequate to cover thecosts of care in the setting that is most appropri-ate for child and family?2 The answer to this ques-

—‘Note that this question ci]fters from the usual approach to dc>fln -

Ing the adequ~clr (JI lnsurancc c[~verage, where health} pe~~ple makt,]ucigment~ ab(>ut their willingness to pa} to co~er the risk. (~t givenlevels ot expense That a frjori que~ti[~n ma} have ~ ~ery ciit terentan~wer t r(]m the (>x p( Kt ~ac (CJ quest Ion addressed I n t h ]t wc t ltln ~ \eereterenct’ 521

tion depends on two factors: the likelihood thata technology-dependent child will be privately in-sured at all; and, once insured, the amount andscope of coverage that the child is likely to have.

Extent of Coverage

Private health insurance is available to childrenthrough two avenues: employer-based group healthpolicies offering coverage of employees’ depen-dents, and self-purchase by the family. Self-pur-chased insurance is generally a great deal moreexpensive to the family than employer-based groupinsurance and covers only about 6 percent of allprivatel y insured children (41).

51

52—

Table 16.—Number and Percent of Children Aged O to 12 Years Covered Only by Private Health Insurance,by Income Status, United States, 1986

Income status

Less than 100% 1000% to 199% More than 2000/.of poverty of poverty of poverty Total

Total number of children(in thousands) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10,861.6 9,997.7 23,658.9 44,518.2

Number with private health insurance(in thousands) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,520.6 6,198.6 20,346.6 28,065.8

Percent with private health insurance . . . . . . . . . . 14 ”/0 620/o 860/0 630/oSOURCE U S Department of Commerce, Bureau of the Census, Current Population Survey, unpublished data, March 1986.

A substantial majority—62 percent—of Amer-ican children between the ages of O and 12 haveprivate health insurance as their sole source ofthird-party coverage (table 16). The likelihood ofhaving private insurance depends primarily on themarital status of the mother and on income sta-tus, Almost 86 percent of children with family in-comes above 200 percent of the Federal povertylevel have private health insurance, while only 14percent of poor children do (41).

Of course, lack of private health insurance doesnot necessarily imply lack of coverage. Many chil-dren (16 percent of those under age 13 in 1986)are covered by Medicaid, Medicare, or the Civil-ian Health and Medical Program of the UniformedServices (CHAMPUS). A small proportion of chil-dren (4 percent in 1986) have a mixture of publicand private insurance. Nonetheless, about 19 per-cent of all American children under 13 years ofage—8.5 million young children—had no healthinsurance of any kind in 1986. 3

Technology-dependent and other severely dis-abled children may be less likely to have privatehealth insurance than children without majorhealth problems. The 1982 National Health In-terview Survey found that 53.8 percent of nonin-stitutionalized children with severe limitations ofactivity had private insurance, compared with74.3 percent of children without any limitations(58). The reasons for these disparities may includeunavailability of individual coverage for severelydisabled children, lack of coverage of preexisting

‘This estimate is based on the Current Population Survey (4]),which asks about health insurance coverage for the previous year,It is not clear whether the estimate relates tc~ a point in time or tothe entire previous year, The estimate is in line with point-in-timeestimates of other surveys (see reference 152).

conditions under group plans, lack of coveragebeyond lifetime maximum limits, differences inthe employment status of parents in the twogroups, and parental decisions to rely on publicsources of support.

Adequacy of Coverage

The fact that a technology-dependent child iscovered under a private health insurance policydoes not necessarily mean that the insurance pro-vides adequate financial resources. The insuredperson’s exposure to out-of-pocket expenses de-pends on the following aspects, which vary widelyamong insurance plans:

●

●

●

●

●

First Dollar Deductible: the amount that theinsured must pay each year before he or sheis eligible for coverage. This amount mayvary by type of benefit (e. g., hospital v.medical).Coinsurance Rate: the percent of the cost ofcovered services for which the insured is re-sponsible.Catastrophic Stop-Loss on Out-of-Pocket Ex-penses: typically an annual upper limit on thebeneficiary’s out-of-pocket payments for in-sured services.Overall Plan Maximums: limits on the totalamount the insurer will pay out on the pol-icy, calculated either as annual, per episode,or life- time limits.Limits on Covered Services: limits on thetype or number of insured services, such asmaximum hospital days or home care visitscovered.

Because expenses for technology-dependent chil-dren are typically catastrophic in nature, and be-

53

cause these children often require complex care,the three most important features of private healthinsurance plans for these children are the overallplan maximums, the catastrophic stop-loss pro-visions, and the covered services.

Overall Plan Maximums and Stop-Loss Provi-sions. —The most direct information on the ex-posure of privately insured children to overallplan maximums comes from the 1977 NationalMedical Care Expenditure Survey. In that year,approximately 50 million children under 18 yearsof age had private health insurance, and about41 million (84 percent) had major medical cover-age. 4 Of those children with major medical cov-erage, only 23 percent had overall plan maximumlimits above $250,000 (53).

Several more recent surveys of employee grouphealth plans, summarized in table 17, show howplan maximums and stop-loss features are distrib-uted among private sector group plans. These sur-veys indicate that over three-fourths of plans (andemployees) are subject to some kind of overallplan maximum, with more than one-half of allemployees under policies with life-time maximumlimits of $500,000 or less.5

Data on new group health insurance policieswritten by insurance companies in 1984 show adefinite trend toward higher overall maximumlimits. Ninety-three percent of the employees cov-ered by a sample of new group policies writtenby commercial insurance companies had maxi-mum limits of $1 million or more (80). Althoughthe data from all sources taken together suggestsubstantial improvement in adequacy since 1979,it is probable that over one-half of privately in-sured families still have insurance that is inade-quate for the catastrophic expenses associatedwith long-term technology-dependent conditions,

Maryland and Illinois programs for children re-quiring respiratory support. If hospitalized, Mary-land children would, on average, exceed a $250,000maximum in about one year (even if the insurerpaid only 80 percent of charges). Yet the averagelength of hospital stay for these children was 14months (93). Even at home, these children wouldexceed this lifetime maximum within about 3 years.Of 63 children in Illinois’ home care program, 42were supported exclusively through Medicaid, 23because their private insurance had lapsed (104).

The situation is somewhat better with respectto catastrophic stop-loss coverage. In 1984, morethan three-fourths of plans and employees had anannual catastrophic limit on out-of-pocket ex-penses. Catastrophic limits ensure that familieswill not be wiped out by coinsurance requirementsin the early months or years of expenditures fora technology-dependent child, but they do notlessen the exposure of families to lifetime maxi-mum benefit limits.

Limitations on Covered Services.—Of particu-lar interest to technology-dependent children isthe availability of home care coverage. Almostone-half of employees in medium and large busi-ness establishments were without any home healthcare benefits in 1984 (175). Although home healthbenefits have been introduced increasingly in therecent past as a cost-containment measure (e. g.,11 percent of plans in a survey of large firms re-ported adding home health care benefits between1980 and 1982 (121)) it appears that many chil-dren would not be eligible under their currentplans. (Note that although health maintenanceorganizations are often thought of as providingcomprehensive coverage, they too may have lim-ited home health benefits. )

Not only is home health frequently an unco-

The impact of low lifetime maximums can be vered service in private insurance policies, but as

demonstrated with preliminary data from the structured, these benefits typically do not meetthe needs of a child requiring continual nursing

‘Nla j (}r medica l c(lverage prc~vldes fOr a wide array ok serlrices care. For example, although over 90 percent ofand u~ually Include> an annual deductible, coinsurance requirements, Blue Cross/Blue Shield plans covered home healthand m~xtmum benetit llmlt~, By comparison, ba$ic benetlt plansu~uallt’ prov lde f}rst-dc~ll ar cot’era~e, but cover (~nlt’ a \’er}, nar - services in 1984, 95 percent of such plans limitedrow set (lt ser~ri ce~, the number of professional nursing visits, b About

‘The only exception ]s reported bv Fox and }’[)shpe, who surve}eda small ;ample (60 f i rm> t ~~t employer-s, This sample may be biased

one-half of the plans had annual limits on thebecause it was drawn from a data source ]isting firms with net asset —values ab{>ve a threshold, suggest in~ that e~’en the smaller t lrms ‘Blue Cross Blue Shield covers about 13 percent <Jt employees]n the sample are dlsprop(lrt ionately wealth}’ (.58). In medium and large firms ( 175 I

Study

U.S. DOL-BLS . . . .

W y a t t C o . . .

H e w i t t A s s o c i a t e s

Fox & Yosphe. . . . . . .

B a t t e l l e / E B R l

Surveyyear(s)

1984

1978-84

1979-84

1986

1977-78

Table 17.—Surveys of Employer-Sponsored Group Health Insurance Plans

Sample

Probability sample of 1,326 -

business establishments meetingindustry-specific minimum sizerequirements (ranging from 100to 240 employees)

In 1984, 1,115 firms of all sizes(but mostly large) participatingin the study

250 major employers; 680/0 inFortune 100, 32% in Fortune 500

Random sample of 60 firms ofall sizes selected from Dunn &Bradstreet’s U.S. BusinessDirectory (small firms) andBusiness Insurance Directory(medicine and large firms)

Probability sample of smallnonagr icu l tura l bus inessestablishments (less than 250employers)

— — - — — .

Employeegroups covered

by survey

Full-timeemployees

Salariedemployees

Salariedemployees

All employees

All employees

Percent of plans(p) or employees

(e) with maximums

1984: 82% (e)a –

1984: 870/0 (p)1980: 880/0 (p)

1984: 82-870/0 (p)1979: 89-900/0 (p)

1986: 67% (p)

1978: 75-830/. (e)c

Percent of plans(p) or employees(e) with lifetimemaximums of

$500,000 or less

1984: 52-570/o (e)

1984: 52% (p)1980: 600/0 (p)

1984: 560/0 (p)b

1979: 750/o (p)

1986: 16,30/, (p)

Percent of plans(p) or employees Percent of plans(e) with lifetime (p) or employeesmaximums of (e) with stop-loss

less than catastrophic$1 million coverage

1984: 53-58°/0 (e) 1984: 76°/0 (e)

1984: 880/0 (p)

1984: 870/0 (p)1979: 590/, (p)

1986: 25.50/. (p) 1986: 800/0 (p)

—%alculated as percent of employees with major medical coverage who are subject to overall plan maximum About 90 percent of plan participants In this sample had major medical coverage The rematnder

had bas!c benefits only, which may not be subject to lifetime Ilmlts but which are often subject to spec!flc maximum Ilmlts on servicesbcalculated as the percent of all employees w!th Ilfetlme maximums less than this amount An additional 4 percent of emplOyeeS who were not subject to Ii fetlme maximums In 1984 were subject to annual

or ‘‘per cause” maximumcThls ,s an overstatement, because employees subject to more than one maximum are double-counted

SOURCES See references 33, 58, 83. 175. 186

5 5

number of visits, with a median limit of 90 visitsper year; only 7 plans covered at least 2 visits perweek (20). These plans also vary in the specifichome health services covered. For example, phys-ical therapy is covered by all plans with homehealth benefits, but respiratory therapy is not acovered service in 22 percent of plans (20). Noneof the Blue Cross/Blue Shield plans in the 1984survey included hourly (“shift”) nursing as a regu-lar home health benefit.

Increasingly, States are using their regulatoryauthority to require health insurers to offer homehealth benefits. At present, 13 States have lawsrequiring coverage of home health services underhealth insurance plans (5). These State laws coveronly those policies written by health insurancecompanies and do not apply to health plans pro-vided by employers on a self-insured basis. Thelatter are exempt from State regulation by Sec-tion 514 of the Employee Retirement and IncomeSecurity Act of 1974 (ERISA, Public Law 93-406).The exemption from State regulation has been apowerful spur to self-insurance by employers, andfurther increases in mandated benefits are likelyto increase the proportion of employers who self-insure (7). Thus, to the extent that families areinsured through employer self-insurance, State ac-tion to mandate home health benefits is not likelyto be an effective mechanism to increase cover-age of services to technology-dependent childrenin the home.

Recent initiatives within the insurance indus-try itself are more promising. Several health insur-ance companies have initiated individual benefitmanagement programs, in which the contractuallimitations on covered services are waived for cer-tain high-cost patients. Under these programs, theinsurance company will pay for services in home

MEDICAID

Background Issues

Medicaid” provides health insurance to verypoor people who are also aged, blind, disabled,

and other settings that would normally not becovered, provided that by doing so the companywill reduce the rate of outflow of total benefit pay-ments. Four examples of such individual manage-ment programs are presented in box D. In a re-cent survey of employer-sponsored health plans,Fox and Yoshpe found that 53 percent of employ-ers had an individual benefits management pro-gram (58), ’ although these programs may not alloperate to encourage nonhospital care for tech-nology-dependent children.

As promising as they are, individual manage-ment programs by insurers do not eliminate theproblems caused by low overall plan maximums,for the insurer typically will not pay beyond thosecontractual limits. Individual case managementcan extend the length of time before the maxi-mums are reached. However, insurers may havelittle incentive to offer this important service ifthey think they will still end up paying out themaximum amount.

A handful of private insurance plans have con-sidered increasing coverage of specific complexhome services to beneficiaries as a group. For ex-ample, three of the plans responding to the BlueCross/Blue Shield 1984 survey reported that theywere developing or implementing pilot programsspecifically for chronically ill children, includingventilator-dependent children. Another three planswere implementing programs for expanding high-technology services in the home, such as intra-venous nutrition and drug therapy, but these pro-grams were not specifically targeted at children(20).

—— ..—

56

m’?% m--d ~ * Aan4m# Prhmt8 hsarara- ad@!wn $@protide cam b e n e f i t s to tWhdOgy._* *X ~ ~“~oti * * &in# ~ b * la Blckly to save roomy, or at leastto &tt&d the kqgth of tim t& !ckild v@ k covered befora re&hhw a Wtirdi maximum. Po- are “brief ~

ke Cro#atti $$tak3 RMkdf ml#6$@4as BqMftt$ Pra$ram “

‘n!w ~U@ Gom#BIue $$@&t Jba@doa &$@MiatM, VX$ bddf (lfthe 90 independent pk’MJ, a benefitsPacka$e 4m’%tm~ Wh’h”tk U*!?+ O!mm d NkW@mmt. This cmtract is one of many heaith in-surance options that a Rderal pm#+~ maydwhoae. Ml WMhws @ Blue Shield plans agree to providethb benefit p+a&e, to elW@&!d -~ qnqdpyees m their ]oad @5sdktions. ‘!k present cmktmct pM3-VJkk employe&a Withti Qp#Un in!a ‘%@h*’ plan, with knefits including 90 home ~th days~y~d~ ~nij Wt; W ~ ‘%W’’@4lkM , with a kwmr premium, Mgher

d&kctibks @ CMfri&&ab, a W M&m l@it on covered benefits, and ma home health bendfits.hkr a pikt pro-; bq#tm when the fh%t ease was brou@t to the atteqtion of &@ program in 1983,

* Bh! CI’os#/Elw can provide llc%rnebenefit$ in excess of the contractedb e n e f i t s t o? iiwkling chikk, who: 3) would b e digibk for cover~if in the hospital, and-s) @be km ~ to care fdr a! h-e @55). A number of individuals whohave been extettdmf speci# Mmefits uncle? W plkt project have been children, pdmarily infants with res-piratory dkrders, hea?tdfsease, tube keCMn@ or parented rmtr’ition requkements, and multiple handicaps.

Aetna Life /k Casualty is a natimal fwq?4’K@t kurance Cc?fnpany wkh regkmal offices that adminkterits plans. The wmpany @m *-M X bmwfits to any policyholder who would be eligible forcoverage if in the hospital and will b h ~ to care for at hmne. Aetna has provided this servicednce 19$3 to all age @oup#. m campy esWMMes that thll? pro#am saved $&S million in the first yearof impkrmmtation, $M million in the sectmcl VW, ad $26 dkn iR tk tid W).

Aetna had served 26 chikkm ander w 16 in tk program as of March 19$6. Saviqgs for each case underthe program are cakukte#”at kast every 6 rnomthsky~ed cases over $6,000 in cost or with care lastinglonger than 6 monthg are tmmitmed and revkwed Cumpany’s nldical director. Benefits can rangefrom an apnea monitor phmhase, home nwdifkatkms, nursing, and therapy. Bene-fits cease when the patient is compktely rekbilitated or the lifetime benefit level is reached. In the lattercam, the patien$ may ke@p purehad @quip!nent (W.

The Equitable

The Equitable Life AsWranm Society uf thelikited States, &for-profit insurance company, offers a Med-i~ c= ~=t ~a$ an b@mud Wwn&rWtt to *P health insurance policies. The policyamendment MM Carr@l ma W@ Cha%$e, @tin agmmrqj* t~ h ammdment the group policyholder (theemplckyar) agrees to pay the curer of a.- mordinamr for my mlevartt Ca=s ~bt *9 ~ of J-w1, 194M, 2S5 group pdicyhokiers had a medicai case man~t agreement with The ikpdtabfe, coveringapprmdmatdy 910,m en@Oyem abont Z,*,OQO total persons (employees plus @@le depemkmts) (M).

Patients may be referred to the cam m amgement program by the company’s pm-admission review serv-ice, the wurce of eli#billity wdfkakn, tk employer, the employee or dependent, claims personnel, orprwiders. The cmnp~$ Wes a ~W-r for of appropriate cases to ensureearly referral and th-+potity to mtxtblkh a rapport with the patient, family, and provider. Includedin the diagnostic profik “m’ * fimnba d diqymses amnciated with neonatal problems (e.g., congenitalbart anOmali@s,w mapkWwy _ ~drom4 and trauma. Mare diagmatic categories can be addedto accommodate the C&k’s id CM)* “

5 7

Under the case management program, case coordinators at The Equitable screen potential cases, assessthe medical and other needs of accepted patients, prepare care plans, coordinate the necessary care, andmonitor progress. Patients in the program can receive services that would not be reimbursed under theusual insurance contract, such as home modification, family counseling, and transfer to a special rehabili-tation hospital (136).

John Hancock

John Hancock Mutual Life Insurance Co., which has a health insurance component, operates a medicalcase management program that is very similar to The Equitable’s. It concentrates on serving trauma pa-tients, high risk infants, and (in the case of older patients) stroke. The program brings a case consultantto certain of these cases to coordinate care and provide benefits not normally available to beneficiaries,such as specialized rehabilitation services and home services, in order to reduce costs while providing appro-priate care (56).

members of families with dependent children, orfirst-time pregnant women (“categorically eligi-ble”). In 35 States and the District of Columbia,people in these categories can also qualify forMedicaid if their medical expenses are sufficientlyhigh that they become poor as a consequence(“medically needy”). Each State has an approvedMedicaid plan that details eligibility, coverage,and reimbursement features in that State.

Two features of the Medicaid program are par-ticularly important in the context of care for tech-nology-dependent children. First, eligibility is avital issue because the Medicaid program is oftenthe third-party payer of last resort for a technol-ogy-dependent child. Second, the coverage ofcomplex health services under the State’s usualMedicaid rules, and the way these services arepaid, affect the setting and amount of care thechild receives. Over the past 5 years, concernsabout these two features of the program have ledto changes in the Federal statute and regulationsregarding coverage and reimbursement under spe-cial Medicaid rules. Many States have taken ad-vantage of these changes, described later in thissection, to enhance coverage for community serv-ices provided to technology-dependent children.

Eligibility

All persons receiving payments under the Aidto Families with Dependent Children program(AFDC) are automatically eligible for Medicaid.(Note that in some States, two-parent families

cannot qualify for AFDC even if they are verypoor. ) In addition, Medicaid eligibility in mostStates is extended to all aged, blind, and disabledindividuals (including children) who receive cashassistance under the Federal Supplemental Secu-rity Income (SSI) program.9 Medically needy per-sons—those who would qualify for these pro-grams but for their incomes, and who have veryhigh medical expenses—can also be made eligi-ble if the State opts to include them.

To be eligible for SSI, an individual must havea disability that is expected to last at least a year(or until death) and must have available incomeand resources no higher than established limits.By statute, the income and resources of certainrelatives, specifically a parent or spouse if he orshe is living in the same household as the indi-vidual, must be deemed available to the individ-ual. After one month in an institution, however,the individual is considered to be not living in thefamily household and the relative’s income andresources are irrelevant to the eligibility determi-

91n providing Medicaid coverage to SS1 beneficiaries, States mayselect one of two options. They can make all SS1 recipients eligibleand, if they choose, also provide Medicaid to individuals receivingonly optional State payments; or they can limit Medicaid eligibil-ity to individuals who meet requirements more restrictive than thoseunder SS1. The State may be more restrictive in setting financialrequirements for income or resources, more restrictive in definingblindness or deafness, or both. Each requirement, however, maynot be more restrictive than that in effect under the State’s Medic-aid plan on Jan. 1, 1972. As of 1983, 14 States required SS1 recipi-ents to meet eligibility standards more restrictive than the Federalstandard (168),

58

nation (20 U.S.C. 416). In some circumstances,the application of these rules may encourage theinstitutionalization of individuals who could becared for at home if Medicaid financing wereavailable.

The linkage of Medicaid eligibility to SSI pay-ments, and SSI payments to institutionalization,allows a child with a long-term disability and in-adequate private insurance to receive hospitalservices under Medicaid, regardless of the incomeof the child’s family. A number of technology-dependent children who would not otherwise beeligible for Medicaid can thus receive hospitalservices under this rule without their families hav-ing to become impoverished. However, until veryrecently these children were almost invariably in-eligible to receive Medicaid reimbursement forequivalent medical care at home, because oncehome, their families’ resources would be deemedto be available to them. This situation receivednational attention in 1981, after the family of ahospitalized ventilator-dependent child appealedher case to Congress and the President. Limitedoptions for the States to avoid the link betweenhospitalization and Medicaid benefits now existand are described later in this section.

Basic coverage and reimbursement

States may pay for hospital care in a numberof alternative ways, and they may place restric-tions on the amount of hospital care they will payfor (see table 18). States pay hospitals accordingto a variety of methods, including:

●

●

●

●

the costs incurred in serving Medicaid pa-tients;prospectively set rates per day, or per ad-mission;prospectively set rates arrived at throughcompetitive hospital bidding or through pre-dicted Medicaid caseloads as a proportion ofhospital budget; orprepaid health plans, in which a health careprovider is paid a set amount per enrolledMedicaid individual, regardless of the actualmedical care use of that individual.

Under the prepaid or prospectively set rate sys-tems, hospitals have an incentive to reduce thelength of hospital stays as much as possible, be-

Table 18.—Medicaid Hospital Inpatient StayMaximums and Units of Payment in the

50 States a and the District of Columbia, 1985

Inpatient hospitalState stay maximum

Alabama . . . . . . . . . . . 12 days/yearAlaska ., . . . . . . . . . .Arizona . . . . . . . . . . .Arkansas . . . . . . . ..35 days/yearCalifornia . . . . . . .Colorado ... . . . . . . .Connecticut . . . . . . . .Delaware ... . . . . . . . .District of Columbia .Florida . . . . . . . . . . . . . . .45 days/yearGeorgia . . . . . . . . . . .Hawaii . . . . . . . . . . . . .I d a h o . . . . . . . . . .40 days/yearIllinois ... ... ... .. .45 days/yearIndiana ... . . ... . .lowa. . . . . . . . . . ... .Kansas . ... . . . . .Kentucky ... ... .. ,14 days/spell of

illnessLouisiana ., . . . . . . . . ..15 days/yearMaine. ... . . . . . . . . .Maryland . . . . . . . separate

maximums foreach case type

Massachusetts . . . .Michigan ., ... .. ....18 days/yearMinneso ta . , . . . . . .Mississippi ., ... ... ,15 days/yearM i s s o u r i . . . . . . . . .Montana . . . . . .Nebraska. ... . . . . .Nevada ... ... ., .,N e w H a m p s h t r e . ,New Jersey. . . . . . . . . . .New Mexico . . . . . . . . .New York . . ... .North Carolina . . . . . maximums per

diemNorth Dakota . . . . .Ohio . . . . . . . . . . . . . . . . .30 days/spell of

illnessOklahoma . . . . .......10 days/spell of

iIlnessOregon . . . . . . . .. ..18 days/yearPennsylvania . . . . . . . .Rhode Island . . . . . . . . .South Carolina . .......12 days/yearS o u t h D a k o t a . . .Tennessee . . . . .......20 days/yearTexas . . . . . . . . . . . . ..30 days/spell of

iIInessUtah . . . . . . . . . . . . . . . . .V e r m o n t . . . . . . . . .Virginia . . . . . . .......21 days/spell of

illnessW a s h i n g t o n . . .West Virginia . . . . .. ..20 days/yearWisconsin . . . . . . . . . . . .Wyoming . . . . . . . . . . . .

Inpatient unitof payment

per diempercentage of chargescompetitive biddingper diemper diemper caseper casecost-basedper caseper caseper caseper caseper caseper diemcost-basedper diemper diemper diem

per caseglobal chargesper case

percentage of chargesper case (DRGs)per case (DRGs)per diemper diemcost-basedper diemper case/per diemcost-basedper case (DRGs)cost-basedper diemper diem

cost-basedper case (DRGs)

per diem

per case (DRGs)per case (DRGs)global chargescost-basedper case (DRGs)per diemcost-based

per case (DRGs)per diemper diem

per case (DRGs)cost-basedper casecost-based

aArlzona~ ~rogram IS a statewide Medlcald demonstration Pro9ram.bsome States with Ilmlted covered hospital days allow longer stays for EpSDTellgibles

c Early and perlodlc Screening, D!agnosis and Treatment program

SOURCE US Department of Health and Human Serwces, Health Care FtnanclngAdministration, Hea/fh Care Financing Program Sfatisfics ” Arra/ys/s ofState Medicaid Program Characteristics, 1984 (Baltimore, MD DHHS,August, 1985), and S.S Laudecina, A Cornparat/ve Survey of MedicaidHospita/ Refrnbursernerrf Systems for /rrpatient Services, Slate by State,1980-1985 (Washington, DC George Washington Unlverslty 1986)

—

cause they do not recoup any extra payment forlonger stays or extra services. Furthermore, evenin some States with cost-based reimbursement,hospitals are paid by Medicaid for care only upto a limited number of days. The net effect of thesepayment methods and limits is to provide hospi-tals with an incentive to discharge patients as soonas possible, or as soon as the day limit has beenreached. If a child cannot be cared for outside thehospital, the hospital is faced with providing in-definite charity care. Medicaid payment thus maycover only a small fraction of the total hospitalcosts of caring for a technology-dependent child.

States are not required to cover either pediatricnursing home stays or pediatric home care in theirMedicaid programs. If they do cover the former,however, they must also cover the latter; and,covering home care means that certain minimumservices must be provided. 10 States covering pedi-atric home care as a normal part of their Medic-aid programs (all but three do) must provide somebasic services, such as home nursing visits, med-ical equipment, and supplies (167). States mayalso cover numerous optional services. As is evi-dent from table 19, the result is considerable var-iation in the services covered (and the limits tocoverage) across States.

Very few States cover the full range of serv-ices and technologies needed by a technology-dependent child in a nonhospital setting as a partof their regular Medicaid benefits. For example,30 States provided no home shift nursing (i.e., pri-vate duty nursing) at all in 1984 (167). Further-more, Medicaid home services vary dramaticallyin amount even where they are provided. AllStates covering home services under Medicaidmust offer intermittent or part-time home nurs-ing, for instance, but the number of covered nurs-ing visits varies from 50 to 300 visits per year(167).

‘“As of 1980, States may, at their option, pr(>t]~e case manage-ment and hc~me resplrat(jry care services under hfedica id ~ [>LIb] ic

Law QQ-453 ~

5 9-—

Special Options for Financing theHome Care of Technology -DependentChildren Under Medicaid

Four special options have been available underthe Medicaid program for States to use in extend-ing eligibility and expanding the range of coveredservices for technology-dependent children whocan be cared for in their homes. Three of theseoptions require the States to obtain a federally ap-proved waiver of usual Medicaid rules in orderto provide additional services, while the fourthallows changes in eligibility rules but not services.The

1.

2.

3.

4.

A

options are: -

the individual “Katie Beckett” waiver (phasedout after 1984),the Section 2176 regular home- and commu-nity-based waiver,the Section 2176 model home- and commu-nity-based waiver, andan amendment to a State’s Medicaid plan.

summary of the various provisions of eachof these options is presented in table 20.

Individual Waivers

An individual waiver program, created in 1982by the Secretary of the Department of Health andHuman Services (DHHS), was the first Medicaidoption designed to address the problems of indi-viduals who remained institutionalized becausereturning home for less costly medical treatmentwould result in the loss of SSI and Medicaid eligi-bility. Commonly referred to as “Katie Beckett”waivers (after the first child to receive one), theywere intended as a temporary strategy to permitspecified individuals to have Medicaid coverageat home while States pursued the longer range op-tions of 2176 waivers or State plan amendments(47 FR 24274).

Requests for these waivers were accepted fromState Medicaid agencies between June 1982 andDecember 1984. A DHHS interdepartmental re-view board determined whether or not the usualSSI deeming rules should be applied in each par-ticular case. 11 For each nominated child, the board

‘ ‘Due to the large number [~i applicatic~ns that t~ere not rew~lved,the board continued to act into IQ86.

60Table 19.-Summary of Medicaid Home Health Services Provided in 50 States and the District of Columbia, 1984

"'rivate duty "'art-tlme Home r-ersonal MeOical supp"eSI r-rosrneTlc State nursina nursina TheraDiesa health aide care services equipment devices Transportation

Alabama, , N A,L N A,L N L A,L L

Alaska, , , , . , . N A A A N A A A,L Arkansas .. ,. N U L U L U A,L A California ... N A,L A A,L N A A,L A Colorado ",. N A A A A,L A L A,L Connecticut, . L L U L N U L U Delaware .... A,L U U U N U L A District of Columbia .... U U L U U A,L A,L N Florida,.,., ...... , .. , .. N U N U N L L A,L Georgia .. ,' . . . . . . . . . . . . . N L A,L L N A L L Hawaii.,. , N U U U N A A,L A Idaho", .. N L L L L A,L N A Illinois, , , A A A A N A L A Indiana, , , , , , , L A A,L L N A A A Iowa"., .. "., , N U L U N L L L Kansas, , , , , , . A,L L L L A,L A,L A A,L Kentucky N L L L N A A,L A,L Louisiana N L L L N A,L A A,L

Maine""" , N U U U N U U L Maryland """,. N L L L A,L L L A,L Massachusetts " , , , .. , .. A,L L L L L L L A Michigan ... ,"',. N U L U U L A,L L Minnesota, , , .... L U L U L A,L A,L A Mississippi N L L L N L N L Missouri, .. , , , , , , N L L L L L A,L L Montana, , . , ..... , . , L L L L L A A,L A,L Nebraska , , , , , , , . A U U L L A,L A U Nevada, , .. A A A A A A A A New Hampshire .. A,L L L L L A A L New Jersey, , , N A A A N A A,L A New Mexico N A,L A,L A,L N A A A New York ... , A A A A A A A,L A North Carolina N L L L N L N L North Dakota, U U U U N U U U Ohio.", .... , L U A,L U N L A A,L Oklahoma, , , N L N L A L L L Oregon, " , A A A A A A A A Pennsylvania .. N L L L N A A,L A,L Rhode Island .. N A,L A,L A,L N A A A,L South Carolina N L L L N A,L A A,L South Dakota , , , , N A A,L A L L L L Tennessee .. , , N L L L N L A L Texas, " N A,L N A,L A,L A,L A,L L

Utah"", , A U A,L U N A A,L A Vermont , ... , , N L L L N L L L Virginia, , , , , . N U U U N A,L N A Washington, , , , A,L L A,L L N A A,L A West Virginia, , A U U U N A A A Wisconsin ... A U A,L U A,L L A A,L Wyoming, . N L L U N L L L

Total: iN. 30 0 4 0 30 0 4 1 #U. 2 16 8 16 2 5 2 3 #A .. , 13 14 17 13 9 29 31 31 Jil 10 30 39 32 14 27 29 28

KEY: N = not provided; U = no limits; A = prior authorization required; L = other limits aphysical, occupational, and speech and hearing. Respiratory therapy has recently been added as an optional Medicaid home health service, but it was not allowable in 1984.

SOURCE: U,S. Department of Health and Human Services, Health Care Financing Administration, Health Care Financing Program Statistics: Analysis of State Medicaid Program Characteristics, 1984 (Baltimore, MD: U.S. DHHS, August 1985),

I

Table 20.—Comparison of State Medicaid Options for Extended Home- and Community-Based Care— . — — —

Number of —

individuals able

-—

AllowableCategoricaleligibility

Disabled Individuals who,because of relatives income,would otherwise be eliglblefor Medicaid only ifinstitutionalized

State may target to aged ordisabled, mentally retarded ordevelopmentally disabled, ormentally ill. Individuals mustrequire level of care providedin ICF, ICF/MR, SNF, orhospital

States can define specificcategories of disabledindividuals, Individuals mustrequire level of care providedin ICF, in ICFIMR, SNF, orhospital

Disabled individuals under age19 who, because of relatives’income, would otherwise beeligible for Medicaid only ifinstitutionalized; individualmust require level of careprovided in a hospital, ICF,ICF/MRi or SNF

Incomeeligibility— —— to participate Geographic areas services Time periodOption— — — —

Individual waivers(no longer newlyawarded) . . . . . One person per Not applicable

waiverRegular State Medicaidservices only

Individuals eligibleuntil waiver nolonger needed

Deeming rules arewa i ve

Regular 2176w a i v e r . States may waive

deeming rules; mayincrease incomeeligibility to 3000/0of SSI standard

All persons meeting May be less thaneligibility criteria statewide

Can offer certain servicesotherwise not authorizedunder Medicaid law; canprovide more extensivecoverage of regularservices

3-year waiver;5-year renewal

Model 2176waivers . States must waive

deeming rules50 or fewer slots per May be less thanwaiver program statewide

Similar to regular 2176waivers; must offer at leastone service in addition tothose provided by regularMedicaid

3-year waiver;5-year renewal

State planamendment . . Deeming rules are

waivedAll persons meeting statewideeligibility criteria

Regular State Medicaidservices only

State option

. — . —SOURCE H B Fox and R Yoshpe, “Technology-Dependent Children’s Accross to Medlcald Home Care Financing, ” prepared for the Office of Technology Assessment, U S. Congress, August 1966

—

I

62

determined whether it would be inequitable un-der the circumstances to consider family incomeand resources in assessing the child’s eligibility forSSI payments (and thus Medicaid). For the boardto waive the deeming requirements, it had to de-cide that:

1. enabling the individual to be eligible for home-based care would result in reduced Medic-aid expenditures, and

2. the quality of the home-based care would beas good as or better than that provided inan institution.

The board also could impose additional stand-ards in particular cases, depending on the factspresented.

Once a waiver was approved, it remained ineffect—and the individual retained Medicaid eligi-bility at home—until the waiver was no longerappropriate. This would be the case if the indi-vidual could no longer meet the SSI disability cri-teria; if the countable income and resources of theparent (or spouse) fell below the SSI or State sup-plement standard; or if a waivered child reachedthe age of 19, at which time he or she could qual-ify for SSI and Medicaid as an adult without con-sideration of parental income and resources.

The obvious attraction of the individual waiveroption for States was the ability to provide moreappropriate Medicaid coverage for selected indi-viduals. States pursued the option as a short-termresponse to a small number of extraordinary cases,usually in the face of significant public pressure.But since the waiver only entitled individuals toregular Medicaid services, States without manyMedicaid home care benefits may have found itdifficult to use this option unless other sources ofhome care financing were available to the childas well.

States used this option not only to cover institu-tionalized children who needed the deeming ruleswaived in order to return home, but also to coverdisabled children already at home. In some in-stances, these children needed Medicaid benefitsas a backup for private insurance; in others, theyalready were SSI- and Medicaid-eligible and neededthe deeming rules waived so that their parentswould be permitted to earn higher incomes.

Regular 2176 Waivers

These waivers, authorized by Congress in Sec-tion 2176 of the Omnibus Budget ReconciliationAct of 1981 (Public Law 97-35), enable States tofinance a wide array of home- and community-based services for Medicaid recipients who other-wise would require institutionalization. Under thewaivers, States can designate specific target pop-ulations who will be subject to broader incomeeligibility policies and receive a wider range ofhome- and community-based services than nor-mally covered under the State plan.

Eligibility for regular 2176 waiver programs islimited to Medicaid recipients who, in the absenceof home and community services, would requirelong-term care in a skilled nursing facility (SNF),intermediate care facility (ICF), or hospital. Spe-cial reference to hospital-level care for the venti-lator population (Public Law 99-272) and for allother individuals (Public Law 99-509) was addedto the statute in 1986. Even before the addition,States could have included hospitalized individ-uals in these waiver programs, but this policy wasunwritten and not clearly communicated to theStates (59).

States must specify a projected number of peo-ple to be served under the waiver. In defining thepopulation to be served, States must select a tar-get group from one of the following three cate-gories or subcategories of Medicaid recipients:aged or disabled, or both; mentally retarded ordevelopmentally disabled, or both; or the men-tally ill. (States can have more than one waiverif they wish to serve more than one group. ) Inaddition, they may restrict eligibility for partici-pation in the waiver to:

individuals residing within a certain geo-graphic area of the State,individuals being discharged from a long-term care institution, orthose particular individuals for whom theMedicaid cost of providing home- and com-munity-based services is less than the cost ofproviding institutional care.

A State can also expand income eligibility forthe target waiver population beyond that of theregular Medicaid program in two ways. One is

63

to increase Medicaid income eligibility limits toa level equal to three times the maximum pay-ment made to an individual under the SSI pro-gram. (Individuals becoming eligible under thishigher income standard, however, would be re-quired to contribute to the cost of their care. ) Theother option is not to deem a certain portion ofthe family’s income to be available to an individ-ual who receives care at home. Once a State hasset its eligibility criteria for a 2176 waiver pro-gram, all individuals who apply to the programand meet the specified criteria must be accepteduntil the projected limit is reached.

States may provide services under the waiverthat are otherwise not allowed by Medicaid, suchas respite care and habilitation services. 12 Theycan also expand the amount, duration, or scopeof coverage of regular Medicaid services offeredin that State, Among the various regular Medic-aid services that have been offered more exten-sively under the waiver are case management,hourly nursing care, home health aides, personalcare services, medical supplies, and durable med-ical equipment. States may also offer other serv-ices approved by the Secretary, such as minorhome modifications and utility expenses.

To receive waiver approval, it is essential fora State to show that its proposed program of aug-mented services will be no more costly to the Med-icaid program than institutional care. Estimatedper capita expenditures for all Medicaid servicesprovided to all long-term care recipients, includ-ing both home and hospitalization, cannot begreater than they would have been in the absenceof the waiver (42 CFR 441). States that want toserve technology-dependent children are able tocompare the cost of their home care to the costof hospitalization using the prescribed formula.Given that the waiver naturally increases the num-ber of Medicaid recipients receiving long-term careservices in the home, a waiver application usu-ally is expected to demonstrate cost savings in twoways: by showing that the total cost of home- andcommunity-based services is less than the totalcost of institutional care, and by documenting thatthe waiver will afford a reduction in the number

.‘2 Habilitation services now include prevocational, educational,

and supported employment services for discharged nursing homepatients (effective Apr. 6, 19861 (Publlc Law 99-272).

of Medicaid beneficiaries receiving institutionalcare.

The advantage of the regular 2176 waiver is itsflexibility. Eligible groups can be defined narrowlyor broadly, and the waiver can be applied to theentire State or only to a small area. States can,if they wish, use a regular 2176 waiver to servea relatively small group of disabled children whootherwise would be hospitalized. The often pro-longed process of completing the very detailedwaiver application, however, may have discour-aged States from targeting these waivers to thissmall population. Regular 2176 waivers generallyare perceived by the States as being for larger andmore inclusive populations, such as the elderlyand disabled, and incorporating technology-depen-dent children into such a waiver is not attractiveto all States. For one thing, States appear reluc-tant to waive the SSI deeming rules for the largenumber of recipients who would be eligible un-der the waiver. For another, they seem to preferto control the number of very high-cost individ-uals who come into the program for fear of ex-ceeding their original cost estimates and havingtheir renewal request denied. 13

Model 2176 Waivers for the Disabled

Using its statutory authority for regular 2176waivers, the Health Care Financing Administra-tion (HCFA) developed a “model” 2176 waiverin December 1982 to encourage States to providehome- and community-based services to certaindisabled individuals who otherwise would loseMedicaid eligibility outside of an institution. Amodel waiver is similar to a regular waiver ex-cept in two essential respects:

1. it may serve no more than 50 blind or disa-bled children and adults at any one time, *5

andI JA]though the States perceive this as a real obstacle, HCFA staff

report that they routinely grant approval for higher cost ceilingswhere States have incurred unanticipated expenses for medically nec-essary services. (Between March 1985 and April 1986, before Pub-lic Law 99-272 prohibited this practice, the Health Care FinancingAdministration had been denying Federal matching payments forMedicaid expenditures that exceeded a State(s original cost ceilings, )

l~see the State Medicaid Manual, Part S—Eligibility, transmittal

no. 1, February 1983 and final regulations at 42 CFR 441.300.*’Prior to the enactment of Public Law 99-272 in 1985, States oper-

ating under a model waiver ~’ere able to serve only .s0 unduplicatedrecipients, which meant that participants who died or left the pro-gram for any reason could not be replaced.

64

2. it must provide that the SSI income deem-ing rules are waived to permit Medicaid eligi-bility for noninstitutional services.

HCFA’s intention was to assist States in mov-ing quickly through the waiver application proc-ess and to eliminate the need for individual waivers(47 FR 24274). States applying for the model waivermust meet all of the basic statutory and regula-tory requirements for regular 2176 waivers butare required to offer only one home- or commu-nity-based waivered service. As under the regu-lar 2176 waiver, States may target their programsto particular subgroups of the disabled population.

Once a model waiver is approved, States canadmit only those eligible individuals whose esti-mated home care costs are below the estimatedcosts for institutionalization. By contrast, undera regular waiver, States need only show that Medi-caid’s average per capita costs with the waiverwould be less than they would be without thewaiver.

The advantage of the model waiver is that itgives States a built-in cap on costs and a chanceto gain experience with home care for the disa-bled on a small scale. In addition, States inter-ested in serving children who otherwise would behospitalized generally find that a model waiverrequest is more likely to be approved by HCFAthan a regular waiver request. The standardizedapplication form makes it possible to isolate asmall, closely defined group of these children andshow, on a case-by-case basis, the often dramaticprogram cost savings of caring for them at home.If the model waiver is targeted exclusively to tech-nology-dependent children, the State also canavoid the requirement of the regular waiver todocument a reduction in the number of nursinghome residents. The 50-person limit, however,may mean that some States must apply for morethan one waiver to serve this population ade-quately.

State Plan Amendment

In addition to the waiver options, States havethe option of amending their State plans to ex-pand Medicaid eligibility to disabled children un-der age 19 living at home who, because of the SSIdeeming rules, otherwise would be eligible for

Medicaid only if institutionalized (Public Law 97-248). Only the normal range of covered Medic-aid services in that State are available under thisoption; special services cannot be added solely forthis particular group. States must ascertain foreach child that home care is appropriate, and thatthe cost of this care is less than it would be in aninstitution of the appropriate care level. Once aState amends its plan, all children meeting theeligibility criteria, whether or not institutional-ized, must be allowed to participate. A State canelect to discontinue coverage for this group of chil-dren at any time.

The State plan option does not require a Stateto prepare cost documentation or to await alengthy approval process. States are free to de-velop their own implementing regulations. Yet,some States have viewed the option as being toobroad and having the potential of extending Med-icaid eligibility to large numbers of children whoare currently being cared for in the community.From the perspective of families and providers,however, the option’s major drawback is that inStates with meager Medicaid home care benefitsa technology-dependent child’s requirements forservices may not be adequately met.

State Use of Medicaid Options toServe Technology= Dependent Children

An overview of State experience with the fourMedicaid options is presented in table 21. Thistable summarizes waivers serving physically dis-abled, but not mentally or developmentally dis-abled, children. Although at least one State (NewMexico) uses a waiver for the developmentallydisabled population as its major vehicle for pro-viding extended Medicaid home care services totechnology-dependent children, and other Statesmay serve a few such children under such waivers,most States thus far include technology-dependentchildren under waivers for the physically disabled.

Thirty-three States were serving technology-dependent and other physically disabled childrenthrough a waiver as of April 1986.16 Eight Stateswere providing these children with special home

16A number of states Were renewing waivers in 1986, so thesenumbers may have changed.

65

Table 21.–State Activity in Medicaid Home” and Community”Based Service Options as of Apr. 15, 1986

Regular waivers serving Model waivers serving State plandisabled children disabled children amendment

Number of Number of Number of Indiwdual waiversApproved children Approved children Approved

Statechildren for children

waivers served waivers sewed plan change served (number of waivers)Alabama 1 9AlaskaArizonaArkansas Yes 47

California 1 60—

Colorado Yes (7)

Connecticut 1 6Delaware

District of Columbta ‘-Yes (8)

Florida—— —Georgia 3 7 Yes 0 Yes (18)

Hawaii

Idaho 1 31 1 3 Yes 100

Illinois 1 50 Yes (30

Indiana Yes (1)

Iowa 1 14 Yes (10)— . .

Kansas

Kentucky 1 36

Louisiana Yes (2)

Maine Yes 66

Maryland——

1 13—Massachusetts Yes a Yes (28)

Michigan 2 38

Minnesota 1 14 Pending b

—Mississippi 1 0Missouri 1 2 T— -Montana l – 30

Nebraska —Nevada — –

— .Yes 5 Yes (2)—

New Hampshire——.New Jersey 1 3 3 40 Yes (6)

New Mexico— .

1 25 Yes (1)

New York 1 3 Yes (6)

North Carolina 1 7

North Dakota

Ohio 1 24

Oklahoma——

--Oregon —Pennsylvania Yes (13)

Rhode Island Yes 12

South Carolina

South Dakota

Tennessee—Texas - ‘ * - =

Utah

Vermont— —

Virginia

Washington 1 1 5

West Virginia Yes (2)—Wisconsin

-.Yes 100—

Wyoming—

Total “yes” answers 8 207 19 244 9 331 14 134—aMassachusetts began admitting children under this eligibility provision in fiScal Year 1987.bMlnnesota’5 State Plan Amendment has been approved by the State and Is pending in HCFA.

SOURCE: H.B Fox and R Yoshpe, ‘( Technology .Dependent Ch!ldren’s Access to Medicaid Home Care Financing, ” prepared for the Office of Tech-nology Assessment, U.S Congress, August 1986.

66

care services under a regular 2176 waiver and 14States were broadening their eligibility as well astheir benefits through 19 separate model waivers.In addition, nine States had amended their plans(a tenth has now been added) and 14 had re-quested and received individual waivers.

Together these options have been serving 938physically disabled children. Technology-depen-dent children who require device-based respira-tory or nutritional support (equivalent to GroupsI through III in this Technical Memorandum) ap-pear to comprise over 60 percent of the under 21population receiving home care under one of thethree waiver programs (47).*7 Among the otherphysically disabled children covered by the waivers,about 25 percent have central nervous system dis-orders (e.g., cerebral palsy, quadriplegic, or spinabifida), and about 10 to 15 percent are charac-terized by congenital, metabolic, or immune dis-orders (e.g., cystic fibrosis or congenital heart dis-ease, ) or by injury-induced trauma.

In the 17 States without either waivers or a Stateplan amendment, technology-dependent childrenare subject to the same Medicaid eligibility re-quirements and home care coverage that otherMedicaid recipients are. In these remaining States,children who cannot qualify for Medicaid as cate-gorically eligible or medically needy may rely onMaternal and Child Health program funds forsome home services; or they may remain in aninstitution in order to retain Medicaid eligibility.Children who can qualify for Medicaid as pooror medically needy individuals may receive regu-lar Medicaid home services; or, in one or twoStates, they may be able to receive certain addi-tional home services under the Medicaid Early andPeriodic Screening, Diagnosis and Treatment(EPSDT) program. 18

17 Comparable diagnostic information was not available for chil-dren receiving home care services under a State plan amendment.

18 EPSDT is a separately authorized program under Medicaidoffered to all low-income, Medicaid-eligible children. Its funct ionis to detect and treat correctable abnormalities in children, such asvision and dental problems. EPSDT services are not subject to thesame Federal limits and requirements as other Medicaid services,and under EPSDT States can offer services to poor children thatare not available to other Medicaid beneficiaries.

Practices Limiting Use ofthe Medicaid Options

The special Medicaid options that can be usedto extend services to technology-dependent chil-dren are limited by Federal statute and regulations,State implementation, and insufficient knowledgeand understanding of the options. While somelimits are unintended, others are the result of con-scious efforts to control costs, or the result of serv-ing technology-dependent children under waiversthat were tailored primarily for the elderly pop-ulation. For example, the requirement that a Statemust prove that a 2176 waiver will not increaseMedicaid costs is one that for many States entailsexpensive and difficult documentation. In a sec-ond example, the fact that children with eligibil-ity under the State plan amendment option canreceive only regular Medicaid home health cov-erage means that this option may be only mini-mally useful in some States unless general cover-age is expanded. And, expanding coverage wouldmean extending the home services available to allMedicaid recipients, including the elderly, whichmany States fear will be very costly.

A number of States have argued that HCFAprocedures for waiver approval are unduly con-fusing and time-consuming. HCFA, on the otherhand, argues that the process is relatively straight-forward if States are adequately prepared and thatHCFA itself offers assistance in preparing the ap-plications. Both of these perspectives are prob-ably valid. A lack of communication and under-standing between HCFA and the States seems tohave contributed to a reluctance on the part ofsome States to apply for waivers (or to implementState plan amendments), to follow through on theapplications, or to tailor the waivers to the needsof technology-dependent children.

Although a substantial number of 2176 waiversand State plan amendments are in effect, in manyStates not as many technology-dependent childrenas might be expected are receiving the benefits ofthese options. Variation in the use of waivers isfrequently a function of the way a program isstructured with regard to income eligibility, cate-gorical eligibility, cost-saving determinations, andservice coverage. Specific State restrictions that

. -

can limit access of technology-dependent childrento Medicaid services include:

not waiving SSI deeming rules (possible onlyunder the regular 2176 waivers, since theserules must be waived under a model waiver),restricting eligibility for a waiver only to cer-tain disease categories (possible only undermodel waivers),allowing waivered services only to individ-uals actually discharged from an institution,not allowing home care costs to be comparedagainst the costs of hospitalization (as op-posed to SNFs or ICFs),limiting reimbursable hospital days (whichmay make it difficult to show program costsavings from home care to Medicaid),not covering skilled shift nursing (i. e., pri-vate duty nursing) as a regular or a waiveredservice, andnot expanding in other ways the range ofregula Medicaid home services availablewhen relying on an individual waiver orState plan amendment to serve the needs ofthe technology-dependent population.

Tables 22, 23, and 24 summarize the restrictionsof the various home care options in specific States.

In some instances, the State’s “attitude” towardfinancing the care of these high-cost children, amore subtle program feature to capture, is the realdeterminant of how many technology-dependent(and other physically impaired) children receiveMedicaid home care benefits. For example, only

about one-half of the States operating Section2176 waiver programs that include children rou-tinely inform the families of children who facelong-term hospitalization of their right to beevaluated for waiver program participation (59).Moreover, only one-fourth of the States withregular waivers and 15 percent of those withmodel waivers report that they publicize the avail-ability of their programs. In Georgia, whichsought (and received) three model waivers in re-sponse to great political pressure to help a fewparticular children, not even hospital dischargeplanners have been told about the waivers. In-deed, even among the Medicaid agency staff, thereis much confusion and misinformation aboutwhether additional children may be covered. Asimilar situation exists in Mississippi.

Many of the waiver and amendment programshave had long initial delays, often due to a short-age of case managers and home health agency per-sonnel. In such instances, disabled children, likeother potential participants, have been unable toobtain the intended home care benefits.

For the most part, the 2176 waiver programs—particularly the regular waivers—have been de-signed and used to serve populations other thanphysically impaired children. Accordingly, thesewaiver programs often have State restrictions in-tended as gatekeeping mechanisms to reduce pro-gram costs, but in practice the restrictions act tolimit the usefulness of these programs for tech-nology-dependent children.

STATE-PROVIDED SERVICES FOR CHILDRENWITH SPECIAL HEALTH CARE NEEDS

Prior to 1981, States provided a number of spe-cific health services to women and children undera series of categorical grants, authorized under Ti-tle V of the Social Security Act and jointly fundedby the States and the Federal Government. Theseservices included maternal and child health serv-ices; crippled children’s services; supplementalsecurity income services for disabled children; he-

mophilia treatment centers; and other programsaimed at specific groups or health problems. The1981 Omnibus Budget Reconciliation Act (Pub-lic Law 97-35) replaced these categorical grantswith a single block grant to each State, eliminat-ing most of the requirement for specific servicesand allowing greater State discretion, A specifiedportion of the total funding continued to be set

State(year

Table 22.—State Policies and Practices That Limit Participation Under the Regular Waiver ProgramsThat Theoretically Could Serve Physically Disabled Children, April 1986

Service limitations

Number and percent of

participants implemented) under 21

Alabama (1984)

California (1985)

Florida (1985)

Georgia (1984)

Idaho (1984)

Kentucky (1983)

Missouri (1925)

Montana (1983)

New Jersey (1983)

Rhode Island (1982) (renewed 1985)

South Carolina (1984)

Virginia (1982) (renewed 1985)

Washington (1982)

9 (0%)

60 (48%)

o (0%)

o (0%)

31 (12%)

36 (18%)

23 (96%)

3 (6%)

3 (0%)

o (0%)

o (0%)

o (0%)

15 (2%)

ncome eligibility restrictions

x x N/A

x x

x x N/A

x x N/A

x x N/A

x x x

x x N/A

N/A

x x

x x

x

NA = Not applicable in that State.

Categorical eligibility restrictions

x

x

x

x

x

x

x

x

x

x

x

x

x x

x x

x

x

x

x

x x x

x

x x

x x

x

x

x

x x x

x x x

x x x

SOURCE: H.B. Fox and R. Yoshpe, "Technology·Dependent Children's Access to Medicaid Home Care Financing," prepared for the Office of Technology Assessment, U.S. Congress, August 1986.

x

x

Table 23.—State Policies and Practices That Limit Participation Under the Model Waiver Programs, April 1986

(1985) - (loo%)New Mexico 25 x x(1984) (loo%)— —New York I 3 x x(1984) (loo%)New York II o x x

(1983) (100%)

\,aMlchigan compares the cost Of home care to 60 percent of the DRG howmal remtnmement

—

bThe Mlsslsslppl model waiver has not Served any cllents and ther@fore has m w09rarn exmrleflcecKJeW Jersey’s Medically Ne@ Program began in July 1966, but the waiver programs do not cover the medlcall Y needy.

SOURCE H B. FOX and R Yoshpe, “Technology-Dependent Children’s Access to Medicaid Home Care Flnanclng, ” prepared for the Office of Technology Assessment, U S Congress, August 966

70

Table 24.—State Policies and Practices That Limit Participation Under the State Plan Amendment, April 1986

Service limitations Service cost limitations

State relies heavilyNumber of State plan on private insurance, Does State plan limits Does State plan does

State children covered does not voluntary services, not use hospital coverage not use not cover SNFs(year under this cover skilled and/or family hospital cost and allows no SNF cost for individualsimplemented) provision shift nursing delivered services comparison exemptions comparison under age 21—Arkansas 47 x – — x(1985) ——Georgia

—-0 x x x

(1982)

Idaho 100 x x x x(1984) —————Maine 49 Xa

(1983) — .Massachusetts

—Ob

(1986)

Nevada 5(1982)

Rhode Island 12 ‘– x(1985) ——South Dakota

—.l – x x

(1985) —W i s c o n s i n ‘– 100

—x

(1 983) —aMalne ~ amending Its State plan to Include skilled nUrsin9.

———

bMasSaChu~ett~ began admlttln~ ~hlldren under this eligibility provision In fiscal year 1987

SOURCE H B Fox and R Yoshpe, “Technology-Dependent Children’s Access to Medicaid Home Care Financing, ” prepared for the Off Ice of Technology Assessment,U S. Congress, August 1986

aside, however, for special demonstration proj-ects, training, and genetic disease and hemophiliaprograms (158).

Under the present Maternal and Child Health(MCH) block grant program, States must matchevery 4 Federal dollars with 3 State dollars. 19 Anevaluation of the implementation of the blockgrant program by the General Accounting Office(GAO) found that States tended to spend theirallotments in ways substantially similar to priorpatterns (158), In all 13 States studied by GAOin 1984, States were offering extensive services tocrippled children (recently redesignated “childrenwith special health care needs” (CSHCN)). Serv-ices offered by the States were extensive. MostStates had actually increased their funding forthese services, and four of the 13 States had addednew services. The programs themselves showedgreat diversity, however, maintaining differencesthat existed before the block grant was established.Most program funds are now spent on screening

— . —l~The Federa] Government Spent $67 mi]]ion on handicapped chil-

dren’s services in 1983, most of it from the block grant (amountingto 23 percent of the total MCH grant) (8). States spent an additional$247.6 million, some of which was matching MCH funds.

and treatment of handicapping conditions. How-ever, they also fund a variety of ongoing supportservices such as counseling and case management.A few States operate State-owned hospitals forhandicapped children.

The population served by the CSHCN programhas changed considerably since 1935, when theprogram was first enacted. Originally, Title Vspecified that the program was to provide diag-nostic, corrective, and rehabilitation services tochildren with crippling conditions, such as polioand cerebral palsy. Over time, however, the origi-nal program has expanded in many States to servechildren with a wide range of chronic health con-ditions, and the ventilator-dependent child is a re-cent example of the new population (107).

The States interviewed by GAO typically pro-vided CSHCN services through State health agen-cies and physicians on a fee-for-service basis. Serv-ices include “screening, diagnosis, surgical andother corrective procedures, hospitalization andafter care, and speech, hearing, vision, and psy-chological care” (158). The Federal legislationestablishing the MCH block grant prohibits thecharging of fees to low-income mothers and chil-

71

dren and requires that when fees are charged theyreflect the income, resources, and family size ofthe beneficiary. A number of States have slidingfee schedules for services (158).

CSHCN is a strongly clinic-based program inmost States, actually providing some or all of thecovered services (rather than simply reimbursingfor them, as Medicaid does) (36). It is often co-ordinated with Medicaid; in many clinics, theCSHCN program provides the services and Med-icaid reimburses the clinic for services providedto Medicaid-eligible individuals (135).

Despite the traditional emphasis of clinic-basedcare, most CSHCN programs fund or provide atleast a limited amount of home care services, andsome provide a fairly wide array of such services.The CSHCN program in Los Angeles County,California, for example, will provide or pay forhome nursing services, physical and occupationaltherapy, respite care, and other home services.The program also provides case management forchildren receiving home health services, and train-ing for families of technology-dependent children(103).

The CSHCN programs area particularly signifi-cant source of funded care for technology-depen-dent children in Illinois, Louisiana, and Maryland.Between 1983 and 1986, these three States wererecipients of MCH demonstration project fundsfor Special Projects of Regional and National Sig-nificance (SPRANS). They developed programs,extentions of their CSHCN programs, aimed atappropriate long-term care for ventilator-depen-

dent children. The programs had two principaldesign objectives:

1. to develop a regionalized system of care forsuch children; and

2. to develop a comprehensive, coordinatedmodel of care.

The SPRANS programs of the three States dif-fer in a number of ways. Illinois and Louisianaoriginally targeted only ventilator-dependent chil-dren under age 22, while Maryland targeted allchildren requiring some specialized respiratorysupport under age 18. Louisiana and Illinois runtheir programs out of single hospital-based cen-ters, while Maryland’s program is based in a con-sortium of several hospitals with a coordinatingboard. All programs, however, emphasize train-ing of parents and professionals, case managementand coordination of care, and care in nonhospi-tal settings.

Through the SPRANS projects, the CSHCN pro-gram has focused attention on ventilator-depen-dent children and their problems in acquiring com-munity care, These projects, and programs incertain other States, have centered on the CSHCNprogram as a coordinator of care. Observers andprogram administrators have noted that children’shealth services have tended to be fragmented anddisease-specific, and that State CSHCN programsshould seek a role in the coordination, not justthe provision, of services required by technology-dependent children (106,177). Some of the StateCSHCN agencies appear eager to take on this re-sponsibility.

A COMPARISON OF HOME CARE BENEFITS IN FIVE STATES:THREE HYPOTHETICAL CASES

Because eligibility criteria, covered services, andpayment mechanisms for Medicaid and CSHCNservices vary so dramatically across the 50 States,it is difficult to describe generally a technology-dependent child’s access to publicly financed homehealth services. A child may have access to ex-cellent services in one jurisdiction but be able toreceive little or no financial assistance for nonin-stitutional care in another. Furthermore, a Statethat offers little access to Medicaid services to onechild may offer substantial services through its

CSHCN program, and the reverse might be truein another State. Some States that seem, on pa-per, to offer few services in reality have innova-tive ways of extending certain vital services to atleast a limited population, On the other hand,States with apparently generous benefits may bevery strict in actually authorizing them.

In order to portray the diversity among States,this section examines the opportunities for homecare covera~e that three hypoothetical children

could expect in five different States: California,Georgia, Kansas, Maryland, and Missouri. TheStates were chosen to reflect diversity in size,geography, and urban-rural composition, andalso to represent a wide range of Medicaid andCSHCN program designs. The services availableto the three hypothetical children in each Stateare described below.

C a s e 1 : “ K ”

“K” is a hospitalized 8-month-old infantwith severe bronchopulmonary dysplasia.She is medically stable but still requires full-time ventilator and tracheotomy care (suc-tioning up to 50 times a day) and special for-mula feeding through a nasogastric tube fivetimes a day. She also receives medicationtreatments, chest physical therapy, and rangeof motion exercises four times daily. Themonthly cost of her home care is projectedto be about $7,620 ($6,650 for paid nursing,$300 for equipment, $600 for supplies, $50for medication, and $20 for a physicianvisit). There would be additional initial costsof approximately $900 to cover equipmentand supplies. Costs could decline as she isweaned from the ventilator.

“K” 's mother is single and unemployed;she will rely on AFDC and food stamps tosupport “K” and two other children. “K”grandmother also lives with the family.

Medicaid Services

“K,” as an AFDC recipient, would be automat-ically eligible for regular Medicaid home care serv-ices in all five States, In Maryland, California,and Missouri, “K” would be eligible to receiveaugmented home services through a 2176 waiverprogram. (Maryland has a model waiver programto serve severely disabled children. California andMissouri both operate regular waiver programsthat can include disabled children and allow homecare costs to be considered against the cost of hos-vitalization. ) In these three States, the waivers areroutinely used to serve ventilator-dependent chil-dren and could provide all of the home care ben-efits that “K” requires.

Although Georgia operates a model waiver pro-gram specifically for ventilator-dependent childrenand “K” would meet the established eligibility cri-teria, it is not clear whether she could participate.Thus far, the State has elected to serve only threeventilatordependent children, each of whom iscomatose. Medicaid staff report that, for finan-cial reasons, the agency is not interested in increas-ing the number of children receiving intensivehome care services under the model waiver.20

“K” would be dependent on nonwaivered Med-icaid services in Kansas and probably also inGeorgia. 21 In Kansas, most of her home care needscould be reimbursed through EPSDT, because theState allows home care benefits up to $240 perday through this special Medicaid program. How-ever, “K” ‘s home care needs most likely couldnot be met in Georgia, where she would receiveonly those Medicaid services regularly availableunder the State plan—physician services, medi-cation, a limited number of intermittent nursingvisits, and the ventilator equipment itself.

CSHCN Support

Extensive case management assistance for “K” ‘sfamily could be provided in California, Kansas,Maryland, and Missouri, because the CSHCNprogram in each of these States has agreed to man-age the care of Medicaid children with complexmedical needs. In Georgia, though, CSHCN casemanagement expertise would not be available.

Basic differences in home care benefits amongCSHCN programs could affect “K” ‘s potential forhospital discharge. In California and Maryland,two States in which Medicaid benefits availableto “K” are already substantial, CSHCN programswould be willing to provide certain gap-fillingservices that may not be fully financed throughMedicaid. Missouri CSHCN also could provide

20Many chi]&en m=t the model waiver criteria but have not beenbrought into the program. At present, in one hospital alone, thereare more than 20 ventilator-dependent children unable to obtainhome care financing.

“Georgia uses a maximum monthly home care service limit of$1,200 in determining an individual child’s eligibility. Kansas usesa standard of $240 per day. However, since total parenteral nutri-tion would be an additional inpatient hospital cost, it is calculatedas an additional home care cost above the maximum day rate.

. .

73

equipment, supplies, and some therapeutic serv-ices. Yet in Georgia, neither Medicaid nor theCSHCN program would cover “K” ‘s skilled nurs-ing care.

Case 2: “M”

“M” is an 18-month-old toddler who hasbeen hospitalized since birth due to multi-ple metabolic and developmental problems,including insulin-dependent diabetes, con-genital heart disease, seizures, liver dysfunc-tion, and failure to thrive. “M” ‘s mother,who is divorced and has no other children,is eager to bring him home and would pro-vide much of his care herself. Once home,“M” will continue to require an evaluationand adjustment of his diet on a daily basis,monitoring of his blood glucose level fourtimes each day, and 12 different medications,some of which must be taken two or threetimes daily. The estimated monthly cost ofhis home care is $4,320 ($3,700 for nursing,$70 for equipment, $280 for supplies, $250for medication, and $20 for a physicianvisit).

“M” ‘s mother earns $14,5OOper year andhas saved $2,800. Her employee health ben-efits were meager and ran out quickly dur-ing “M” 's prolonged hospital stay.

Medicaid Services

“M’”s disability and low family income makehim eligible to receive SSI cash benefits. In fourof the five States—California, Georgia, Kansas,and Maryland—he, therefore, would be eligiblefor Medicaid (and regular Medicaid home healthbenefits) as well. Missouri, however, has electednot to provide Medicaid benefits to SSI recipientsunder age 21 unless they are residents of an ICF.“M,” in fact, would have no opportunity to becovered by Medicaid in Missouri. His mother’sincome places them far above the AFDC paymentlevel and Missouri does not provide benefits tothe medically needy.

Under the Maryland and California waiver pro-grams, “M” could be covered by Medicaid for hiscomplete home care package. He would also becovered in Kansas under its EPSDT program,

which in that State is used to fund extensive treat-ment services for certain chronically ill children.In Georgia, the most expensive part of his care—the skilled shift nursing service—could not be re-imbursed, although medication, equipment, sup-plies, and physician visits could. Although Geor-gia has three model waivers for disabled children,“M” would not be able to participate in any ofthese.

CSHCN Support

In Maryland and Kansas, two of the threeStates where “M’”s home care needs could befinanced adequately by Medicaid, case manage-ment and family training would be providedthrough an arrangement with the CSHCN pro-gram. The Maryland and also the California pro-gram could contribute certain services, supplies,and equipment in the event that these were notcovered by Medicaid. The California CSHCNprogram does not charge Medicaid recipients, butthe Maryland program would require “M’”s fam-ily to pay a small co-payment charge. In Geor-gia, where the Medicaid home care benefits avail-able to “M” would be minimal, the CSHCNprogram could provide his family no additionalassistance. “M” would be financially eligible forCSHCN services at no charge, but the agency nei-ther finances nor arranges for skilled shift nurs-ing care at home.

In Missouri, where “M” would not have accessto Medicaid benefits at all, he would be financially

eligible for all CSHCN services, although his fam-ily would have to pay a small fee. The program,however, does not provide skilled shift nursingor other extended home care services. It wouldcover only “M’”s physician visits, equipment, sup-plies, and medication.

Case 3: “T”

“T” was diagnosed at birth as having shortgut syndrome and malabsorption. Now age6, he has been hospitalized approximately

20 times for varying periods. When at home,“T” attends school regularly with a nurse.His daily home care requirements include 20hours of intravenous nutrients, care of thecentral line, and frequent monitoring of his

74

glucose level. The monthly cost of his careis $17,035 ($10,000 for nutritional supplies,$7,000 for nursing, and $35 for a physicianvisit).

“T” ‘s parents, who have three children,both are employed full time and have a com-bined annual salary of $.52,000. They have$6,500 in the bank, two cars, and are pay-ing off the mortgage on their home. “T” ‘shospital and home care both had been cov-ered under his father’s company insuranceplan, but the family recently reached the$1,000,000 lifetime benefit maximum.

Medicaid Services

“T” cannot become eligible for Medicaid in Mis-souri, where the medically needy are not covered,but in the other four States he could become eligi-ble as a medically needy recipient. His parents,however, would have to reduce their savings andincur very substantial medical expenses in orderfor him to qualify. The actual amounts wouldvary from State to State:

●

●

●

in California, the family first would have toreduce its savings to $3,000 and then spend$2,405 for medical care each month to be-come eligible;in Georgia, the family first would have to re-duce its savings to $2,700 and then spend$16,332 in each 6-month period before be-coming eligible (an average of $2,722 permonth); andin Kansas, the family first would have to re-duce its savings to $1,700 and then spend$15,438 in each 6-month period (an averageof $2,573 per month) .22— .

ZZThese spend-down figures are based on the SS1 eligibility de-termination methodology and each State’s medically needy incomelevel (MNIL) for one person. (The family’s monthly income and anestimated $30 in bank interest were added together, $336 was sub-tracted as a living allowance for the other two children, $1,008 was

subtracted as a living allowance for the parents, and $65 plus $2owas subtracted as an exclusion from the parent’s earned income.The remainder, minus a $2o exclusion, was deemed available to thedisabled child, Then each State’s MNIL for one person was applied,as appropriate, on a l-month or 6-month basis. )

None of the State Medicaid agency staff contacted by Fox andYoshpe (s9) would have followed the SS1 methodology as it isprescribed by the Social Security Administration. In fact, one ofthe States would have used its AFDC methodology. Of the threethat would have used the SS1 methodology, two would have useda different methodology appropriate when no other children are in

Spend-down requirements of this magnitude(ranging from $28,860 to $32,664 annually) ob-viously would place an enormous financial bur-den on the family. Only in Maryland could “T”be brought into the Medicaid program withouthis family first having to meet the spend-downrequirement, because under Maryland’s modelwaiver “T’”s family income would not be deemedavailable to him.

Georgia operates three model waiver programs,but “T’”s condition is not covered by any of them.Having amended its State plan, Georgia also pro-vides Medicaid to certain children who, becauseof the deeming rules, otherwise would be eligibleonly in an institution. Under this provision,though, the State restricts eligibility to childrenwhose home care costs would be less than the costin an SNF or ICF, and “T” would not qualify .23In California, “T” could participate in the regu-lar waiver program, but since the SS1 deemingrules would not be waived, “T” would be eligi-ble for the program only after his family met themedically needy standard. Both California andMaryland offer parenteral nutrition as a regularMedicaid benefit and provide skilled shift nurs-ing care as a waivered service.

In the two States where “T” could receive onlynon waivered services, his chances for adequatebenefit coverage would differ dramatically. Kansascovers all necessary treatment services for chil-dren through its EPSDT program, which is notsubject to the service limits of its regular Medic-aid plan. Thus, after an initial screening, “T’”sparenteral nutrition and skilled nursing care bothcould be authorized under EPSDT.24 Georgia, incontrast, does not use EPSDT to expand cover-age for treatment services beyond what is regu-larly available under the State plan. “T” wouldbe covered for the intravenous equipment, phy-sician visits, and rehospitalizations, but not for

the family, two would have used the MNIL for either a three- orfive-person family, and none would have included the unearned bankinterest income.

IJGeorgia uses a maximum monthly home care service limit of$1,200 in determining an individual child’s eligibility.

ZdKansa5 u5es a standard of $240 per day. However, since total

parenteral nutrition would be an additional inpatient hospital cost,it would be calculated as an additional home care cost above themaximum day rate.

his nutritional products or skilled shift nursingcare.

CSHCN Support

CSHCN programs in three of the States—Kan-sas, Maryland, and Missouri—provide case man-agement and family training services to technol-ogy-dependent children enrolled in Medicaid.“T’”s family, therefore, could receive these serv-ices in Kansas and Maryland, where he wouldqualify for Medicaid coverage.

In California and Kansas, where SSI deemingrules could not be waived, “T” would need casemanagement and any other available CSHCNservices prior to meeting the Medicaid spend-down requirement. In California, the CSHCNprogram could purchase “T’”s equipment andparenteral nutrition, deliver 3 months of skilledshift nursing care, and provide continuous casemanagement support; his family would be chargedan amount equal to two times their State incometax. In Kansas, though, CSHCN covered services

DEPARTMENT OF DEFENSE

The Department of Defense (DOD) providesmedical care, or payment for medical care, to thedependents of active and retired military person-nel. It does so in two ways: through its own hos-pitals, operated independently through each of thefour Armed Services branches; and throughCHAMPUS, which pays for care that cannot beobtained in the military hospitals. Armed serv-ices hospitals and CHAMPUS are operated inde-pendently of one another, but they provide ac-cess to the same general categories of services.

DOD pays for nonhospital long-term care intwo ways. First are the regular home health ben-efits available under CHAMPUS. These benefitsinclude:

durable medical equipment, including ven-tilators;oxygen;parenteral and enteral nutrition therapies;physical therapy;skilled nursing care;

75

—which include many of the services “T” needs—are available only to children whose family in-come falls below the poverty line or who are eligi-ble for Medicaid. “T” could not receive these serv-ices during the spend-down period.

CSHCN services in Georgia, not unlike manyother States, do not include any high-cost homecare services. If “T” and his family lived in Geor-gia, therefore, they could not depend on eitherMedicaid or the CSHCN program to finance, evenpartially, the skilled shift nursing care that “T”requires.

Unfortunately, in Missouri, where “T” has noopportunity at all for Medicaid coverage, he alsowould have no way of obtaining home care serv-ices through CSHCN. The program, like Geor-gia’s, emphasizes treatment of crippling conditionsand, while it has purchased sophisticated equip-ment on occasion, “T’”s family would be finan-cially ineligibleincome cut-off

even for this benefit. The annualfor a family of four is $19,000.

. medications and medical supplies; and● physician visits.

Many technology-dependent children, however,may not be judged eligible for the full extent ofthese home benefits, however, because neithermilitary hospitals nor CHAMPUS may provide,or pay for, “custodial care” (164). CHAMPUS’spolicy manual defines “custodial care” as care ren-dered to a patient:

1.

2.

3.

4.

who has a mental or physical disability thatis expected to be prolonged;who requires a protected, monitored, or con-trolled environment, whether in an institu-tion or in the home;who requires assistance to support the essen-tials of daily living; andwho is not under active treatment that willreduce the disability to the extent necessaryto enable the patient to function outside theprotected environment (164).

76

If a military hospital (for a hospitalized child) ora CHAMPUS intermediary should determine thata technology-dependent child fits this definition,that child is then eligible only for a subset of theusual home benefits. These limited benefits includemedications and medical supplies and up to 1 hourper day of nursing care.

CHAMPUS began a home care demonstrationprogram on July 1, 1986, under which it providesextensive home care benefits (including 100 per-cent coverage of most costs and coverage of home-maker services) to patients who would otherwisebe receiving hospital care. These patients couldinclude children receiving intravenous drug ther-apy and many infants who can gradually be weanedfrom their dependence on respiratory or nutri-tional support. However, a child “must be receiv-ing inpatient hospital care that is an otherwiseauthorized CHAMPUS benefit” in order to beeligible for the program (51 FR 23809). This pro-

vision still could exclude many technology-depen-dent children under current policy.

A second source of long-term care coverage isthe Program for the Handicapped (PFTH), a spe-cial CHAMPUS benefit for handicapped depen-dents of military personnel. In order to receivebenefits under this program, an individual mustshow that he or she cannot get services from pub-lic programs or institutions. Prior approval is re-quired for coverage of all supplies and servicesunder the program (164). PFTH covers institu-tional, outpatient, and home care but pays a max-imum of only $1,000 per month in benefits (164).In addition to those supplies and equipment cov-ered under the basic program, it covers physical,occupational, and speech therapy and specialeducational services. Skilled shift nursing andhomemaker services are not covered. PFTH, likethe regular benefits program, does not cover cus-todial care,

OTHER PUBLIC PROGRAMS AND

A wide variety of programs and services fi-nanced by Federal or State governments can affectthe resources and services available to technol-ogy-dependent children. For example, the FederalGovernment provides SSI maintenance paymentsto disabled individuals, and it provides certainadoption and foster care incentive payments toassist in finding homes for needy children (159).States can, and often do, supplement these pay-ments with their own. For example, most Statesprovide supplemental payments to foster parentswho provide care for handicapped or other chil-dren with special needs (159).

Certain in-home services, funded jointly by theStates and by Federal Title XX social service blockgrants (Public Law 97-35), may be provided tolow-income disabled individuals. Title XX fundsare provided to States in order to prevent orremedy abuse of children and other family mem-bers; reduce inappropriate institutional care; se-cure admissions to and services in institutionswhen such a setting is appropriate; and prevent

SERVICES

or moderate the dependence of individuals onother persons (159). Services may include home-maker, home health aide, and other basic homeservices (e.g., transportation) that can supplementthe home-based medical services available throughMedicaid.

States may also have their own special pro-grams, funded entirely through State and localtaxes, that provide special benefits to targetedgroups. Wisconsin, for example, has a State pro-gram that provides “gap-filling” funds to individ-uals, including children, who are at risk of institu-tionalization. Wisconsin also has a family supportprogram that provides, separately from SSI, upto $3,000 per year cash assistance to families withseverely disabled children living at home (37).This example demonstrates that the resourcesavailable to a child can be enormously varied, de-pending on where the child lives—and on the ac-cess of that child’s family to appropriate infor-mation and coordination of services.

77—

CHARITABLE ORGANIZATIONS

Charitable organizations have long been visi-ble sources of research and services to aid the dis-abled. About 20 national children’s health chari-ties operate in this field, ranging in size from verylarge organizations such as the National EasterSeal Society to small organizations such as theRetinitis Pigmentosa Foundation (117). The mis-sions, disease orientations, and structures of thevarious charities are similarly diverse. The majorityof national charitable organizations focus their ef-forts on one disease or closely associated set ofdiseases. However, an organization may concen-trate on research, public education and politicallobbying, direct provision of services, family edu-cation and support, or any of a number of otheractivities.

Charitable organizations have functioned aslast-resort providers for many families with tech-nology-dependent children. One of their most im-portant functions in this regard is as a providerof family support and education. Table 25 lists

the expenses of selected foundations for variousservices, including medical services and patienteducation, Spending for these services range from15 percent of expenditures (March of Dimes) to92 percent of expenditures (Easter Seal Society)(27). “There is no strong relationship betweenprevalence of a chronic condition and relativemagnitude of foundation support. . . . Conse-quently, children with certain disabilities havemore resource available to them than others” (27),Researchers who interviewed a number of na-tional charitable organizations concluded:

Although foundations expend a significantamount on direct services, they tend to provideassistance to cover only those services that arenot otherwise reimbursable and that place an un-reasonable financial strain on families with dis-abled children. These services included transpor-tation, educational and recreational activities,physicial and occupational therapy, special med-ical equipment, and to a lesser extent, medicalcare (27).

Table 25.—Total Amount of Expenses Allocated for Programs of Selected Foundations,1979 and 1980 (millions of dollars)

Total Medical services Public andprogram and patient professional Community

Private foundation services Research education education services/advocacy

Muscular Dystrophy Association, 1979 . $56.6 $18.0 $33.3 $5.3March of Dimes, 1980 . . . . . . . . . . . . . . . . . . 49,9 10.2 7.6 18.4 $13.6Cystic Fibrosis Foundation, 1980 . . . . . . . . 11.1 1.7 4.2 3.6 1.5American Diabetes Association, 1980 . . . . 9.7 1.7 2.7 3.6 1.7Arthritis Foundation, 1980 . . . . . . . . . . . . . . 6.0 2.9 a 2.3 0.8a

Leukemia Society of America, 1980 . . . . . . 3.9 2.2 1.0 0.5 0.2American Kidney Fund, 1979 . . . . . . . . . . . . 1.5 0.04 0.9 0.2 0.4Easter Seal Society, 1979b . . . . . . . . . . . . . . 85.7 0.5 79.1 6.1 —aThe AflhrltlS FOUf_tdatlOn combines patient and COmm Unity SeNICeS Into One cate90rYbThe Easter Seal Society Includes the comb!ned expenditures for the national and all State and tf3VltOri$31 Easter Seal SOCietles

SOURCE J A Butler, P Budettl, M A McManus, et al , “Health Care Expenditures for Children With Chron!c Illnesses “ /n N Hobbs and J M Perrln (eds ), /ssuesIn the Care of Ch//dren W(h Chron/c ///ness (San Franc!sco, CA Jossey.Bass, 1985)

CONCLUSIONS

It is impossible to provide any accurate estimate surance is a major source of third-party paymentof the proportion of technology-dependent chil- for children, as presently structured it is inade-dren with private insurance whose insurance cov- quate to provide for the needs of technology-de-erage includes intensive home care benefits, but pendent children. It fails in several ways. First,it is possible to get a sense of how likely compre- many children are left uninsured as a result of theirhensive coverage is, Although private health in- families’ economic positions. Second, some tech-

78

nology-dependent children may find it difficultor impossible to obtain private insurance. Third,even those children who are privately insuredhave coverage that is likely to be inadequate tocover the expenses associated with these medicalconditions. Finally, the structure of benefits un-der many policies is too rigid to deal with theneeds of technology-dependent children whenthey are cared for in the home. Parents whose chil-dren require full-time monitoring and medical caredare not give up employment to provide some ofthis care, and yet in the majority of cases theirinsurance will not pay for a qualified professionalcaretaker.

Notwithstanding the seriousness of the currentinsurance situation for these families, the privateinsurance industry has made strides over the past5 years towards accommodating payment systemsto complex medical care in the home setting. Ithas done so primarily through case-by-case ex-ception to normal home coverage limits.

The willingness of private insurers to provideat least case-by-case exception is vital to both ben-eficiaries and to public payers, because manytechnology-dependent children quickly lose theirprivate benefits by reaching the maximum allow-able benefit amount. The longer these children canstretch out private insurance through home care,the longer they have before they become depen-dent on Medicaid for health insurance.

Medicaid has likewise made some progress inthe past few years towards accommodating thispopulation. Showing cost savings to Medicaid bycaring for a technology-dependent child at homeis by no means impossible, and the current waiverprograms have shown considerable success atserving at least a few children in this setting atless cost to the program. It is, however, gener-ally much more difficult to show cost savings toMedicaid than cost savings to a private insurer,because Medicaid pays much less in the hospital.States have attempted to limit home and commu-nity costs by restricting eligibility or services insome cases. Unfortunately, the exclusion of cer-tain expensive services-particularly skilled shiftnursing—can absolutely prevent many technol-ogy-dependent children from coming home.

DOD has found it more difficult than Medic-aid to adapt its payment system and benefits to

technology-dependent children. While the usualhome benefits under CHAMPUS can be substan-tial, those benefits are not available to a technol-ogy-dependent child who is judged to need veryprolonged, supportive care. Unless the regulationsdefining custodial care are changed, or the mili-tary hospitals and CHAMPUS undertake a muchmore liberal interpretation of the regulations whenthe prolonged care is very complex, long-termhome care benefits for many technology-depen-dent children are unlikely to be forthcoming.

To the CSHCN programs in many States, thecomplex needs of nonhospitalized technology-dependent children offer a new opportunity to bea primary player in a significant health care is-sue. These programs have often acted as advo-cates for their clientele in the past, and they nowhave a significant new role to play as coordina-tors of payment and community services to thisgroup of disabled children. The degree to whichthe programs are prepared to play this role, andtheir proficiency at it, undoubtedly varies fromState to State. But the role seems an appropriateone, and it may give many of the programs newpurpose and direction.

It is very possible that the extension of privateand public insurance benefits into the home caresetting will replace charity care to some extent.Charitable organizations, including local commu-nity and religious organizations, have helped many

children obtain certain equipment and facility ren-ovation. However, care coordination and skilledshift nursing have never been the province ofcharitable organizations, and these are the areasin which improved health insurance benefits aremost likely to have an impact.

None of the solutions being implemented at themoment regarding technology-dependent childrenare applicable to children who will, due to somemedical or home characteristic, be more expen-sive to care for at home than in the hospital. Forsome such children, home care may be the mosteffective and desirable even if it is not the leastexpensive. For other children, however, particu-larly those without a supportive family, other carealternatives will be necessary. Unfortunately, atpresent, appropriate and effective long-term careoptions other than the family home and the hos-pital are extremely rare.

Appendixes

‘

Appendix A

Acknowledgments

The OTA staff would like to express its appreciation to the Advisory Panel for Technology and Child Healthand to the following

Elizabeth AhmannBaltimore, MD

Marlene AitkenCenter for Hospita

people for providing information and assistance:

Administration StudiesUniversity of ChicagoChicago, IL

Duane AlexanderNational Institute of Child Health and Human

DevelopmentNational Institutes of HealthBethesda, MD

Jonathan BatesChildren’s HospitalSan Diego, CA

Marilyn BergnerSchool of Hygiene and Public HealthThe Johns Hopkins UniversityBaltimore, MD

Kelly BurchLeBonheur Children’s Medical CenterMemphis, TN

John ButlerChildren’s HospitalBoston, MA

Georgia CleverlyNew Mexico Human Services DepartmentSanta Fe, NM

Harvey ColtonChildren’s HospitalSt. Louis, MO

Robin CooperDepartment of Health and Social ServicesMadison, WI

Carl CranorUniversity of CaliforniaRiverside, CA

Thomas CulleyAetna Life & CasualtyHartford, CT

Susan EpsteinProject SERVEBoston, MA

Joseph FitzgeraldJohn Hancock Mutual Life Insurance Co.Boston, MA

Juanita FlemingCollege of NursingUniversity of KentuckyLexington, KY

Harriette FoxFox Health Policy ConsultantsWashington, DC

Dale GarellCalifornia Children’s ServicesLos Angeles, CA

Allen GoldbergThe Children’s Memorial HospitalChicago, IL

David GreenbergSchool of EducationI. U. P.U. I.Indianapolis, IN

Dorothy GuyotDepartment of PediatricsAlbert Einstein College of MedicineBronx, NY

Lenore HeapheyThe Oley FoundationAlbany, NY

Janet HeinrichAmerican Nurse’s AssociationWashington, DC

Jo Anne KaufmanCoordinating Center

CareMillersville, MD

Kathryn KirkhartChildren’s HospitalNew Orleans, LA

Art Kohrmann

for Home and Community

La Rabida Children’s Hospital and Research CenterChicago, IL

81

82 —

Mary Grace KovarNational Center for Health StatisticsHyattsville, MD

Maxine LangeBlue Cross and Blue Shield AssociationChicago, IL

Connie LiermanChildren’s Hospital National Medical CenterWashington, DC

Gregory LiptakDepartment of PediatricsUniversity of Rochester Medical CenterRochester, NY

Gerald LongOffice of Civilian Health and Medical Program of

the Uniformed ServicesAurora, CO

Margaret MikolSKIP of New York, Inc.New York, NY

Naomi MorrisSchool of Public HealthThe University of IllinoisChicago, IL

Nancy MurrayOmaha, NE

Paul NewacheckInstitute for Health Policy StudiesUniversity of CaliforniaSan Francisco, CA

Peter S. OliphantGovernor’s Council on the Prevention of Mental

RetardationTrenton, NJ

Judith PalfreyChildren’s HospitalBoston, MA

James PerrinMassachusetts General HospitalBoston, MA

Patricia M. PierceFamily Health and Habilitative Services, Inc.Gainesville, FL

Jan RoughanThe Equitable Life Assurance Society of the U.S.Brea, CA

Jim ScaleseZoorhees Pediatric FacilityGibbsboro, NJ

Karen ShannonSKIP, Inc.Severna Park, MD

Donald ShapiroDepartment of PediatricsUniversity of Rochester Medical SchoolRochester, NY

George SilverInstitution for Social and Policy StudiesYale UniversityNew Haven, CT

Ruth SteinPediatric Ambulatory CareAlbert Einstein College of MedicineBronx, NY

Margretta StylesAmerican Nurses’ Association, Inc.Kansas City, MO

Eileen ThomasFederal Employee Health Benefits ProgramBlue Cross and Blue Shield AssociationWashington, DC

Phyllis TschumperDepartment of Health and Social ServicesMadison, WI

Robert WardwellHealth Care Financing AdministrationBaltimore, MD

Daniel WhitlockMinneapolis Children’s Medical CenterMinneapolis, MN

Gordon WorleyDepartment of PediatricsDuke University Medical CenterDurham, NC

Penny PollardDepartment of Health and Human ServicesWashington, DC

Appendix B

Workshop on Technology-DependentChildren

Participant List

John Butler, ChairmanResearch Associate, Children’s Hospital, Boston, MA

Lu Ann AdayAssociate Professor of Behavioral SciencesSchool of Public HealthThe University of TexasAustin, TX

Marlene AitkenResearch Project SpecialistCenter for Hospital Administration StudiesUniversity of ChicagoChicago, IL

Jonathan BatesSenior Vice President for Planning and External AffairsChildren’s HospitalSan Diego, CA

Julianne BeckettProgram AssistantChild Health Specialty ClinicUniversity of IowaIowa City, IA

Thomas CulleyMedical DirectorAetna Life & CasualtyHartford, CT

Susan EpsteinProject Co-DirectorProject SERVEBoston, MA

Harriette FoxPresidentFox Health Policy ConsultantsWashington, DC

Dale GarellMedical DirectorCalifornia Children’s ServicesLos Angeles, CA

Dorothy GuyotResearch DirectorHealth Care Financing ProjectDepartment of PediatricsAlbert Einstein College of MedicineBronx, NY

Jo Anne KaufmanExecutive DirectorCoordinating Center for Home & Community CareMillersville, MD

Kathryn KirkhartProgram CoordinatorVentilator Assisted Care ProgramChildren’s HospitalNew Orleans, LA

Mary Grace KovarSpecial Assistant for DataPolicy and AnalysisNational Center for Health StatisticsHyattsville, MD

Edward LisDirectorDivision of Services for Crippled ChildrenUniversity of IllinoisChicago, IL

John MacQueenDirectorChild Health Specialty ClinicUniversity of IowaIowa City, IA

Merle McPhersonActing DirectorDivision of Maternal & Child HealthBureau of Health Care Delivery & AssistanceRockville, MD

83

84

Paul NewacheckAssistant Professor of Health PolicyInstitute for Health Policy StudiesUniversity of CaliforniaSan Francisco, CA

Barbara StarfieldProfessor& HeadDivision of Health PolicyThe Johns Hopkins UniversitySchool of Hygiene and Public HealthBaltimore, MD

Ruth SteinProfessor of PediatricsPediatric Ambulatory CareAlbert Einstein College of MedicineBronx, NY

Robert WardwellChief, Medicaid Special Issues BranchHealth Care Financing AdministrationBaltimore, MD

Daniel WhitlockMedical DirectorChronic Care ProgramMinneapolis Children’s Medical CenterMinneapolis, MN

AgendaWorkshop on Technology-Dependent Children

Office of Technology Assessment600 Pennsylvania Avenue, S.E.

Washington, D.C.April, 11, 1986

9:00 Refreshments9:15 Opening remarks

Clyde Behney, OTA Health Program ManagerJudith Wagne' Project Director Technologies and Child HealthElaine Power, Study Director, Technology Dependent Children

9:30 Defining the population of technology-dependent children● discussion of economic implications of alternative definitions for children, providers, payers● current research on definitions● potentially useful combinations of candidate definitions● definitions not included in candidate list● the place of case-by-case review in defining for Medicaid eligibility purposes

11:00 Break11:15 Defining the population (continued)

● discussion of criteria for judging definitions● assessment of candidate definitions according to criteria

12:30 Lunch1:30 Data sources

● approaches to getting data● programs, offices, states, etc. collecting data on this group

2:00 Changes in the technology-dependent population● current evidence on trends● changes due t. extension of current technology (e.g., more NICUS, trauma centers, extended access

to treatments)● changes due to new technologies (e.g., artificial surfactants)● apparent changes due to the new availability of home services

4:00 Summary and closing remarks

Appendix c

Implications of the Population Definition

Introduction

The way the population of technology-dependentchildren is defined and enumerated has clear implica-tions for the costs to third-party payers of paying forcare, and the access of these children to different carealternatives. The broader the definition, the larger thenumber of children who may become eligible for spe-cial benefits. Providing enhanced insurance coveragefor technology-dependent children may itself lead toan increase in the size of the population, through en-couragement of more aggressive medical practices.

The definition of technology dependence presentedin Chapter 2 was developed for the purpose of enumer-ating the population, not for describing it for insur-ance or program eligibility purposes. These two defini-tional purposes overlap to some extent, but they canalso conflict. The pragmatic, data-based definition ap-plied in this technical memorandum would be inappro-priate if applied in a program context without otherconsiderations. To be applied appropriately to eligi-bility, a definition of technology dependence must takeinto consideration the following questions:

●

●

●

●

●

Does the definition include all children who wouldreasonably be considered to be technology de-pendent?Is the definition flexible, or would it need to berevised frequently to accommodate new groupsof deserving children?Can the definition identify children with similarneeds for health care, so that they can receive thesame level of benefits (horizontal equity); and canit distinguish those with greater need from thosewith lesser need (vertical equity)?Can the definition distinguish between childrenfor whom home care is less expensive than institu-tional care from those for whom it is more expen-sive (possibly because the child would not be in-stitutionalized even in the absence of home carebenefits)?Is the definition compatible with distinguishingchildren for whom home or community-basedcare is feasible and desirable, and can it providea basis for estimating the cost of services providedin these environments?

Three potential specific approaches to identifyingthe population are to use: 1 ) diagnosis, 2) functionallimitation, or 3) medical services needed. These ap-proaches are not necessarily mutually exclusive, buttheir benefits and drawbacks can be discussed sepa-rately from one another.

Three Alternative Approaches

Definition Based on Diagnoses

Diagnoses could be used as a basis for identifyingchildren as technology dependent, an approach thathas two attractions. First, in most cases diagnoses pro-vide distinct and verifiable information. Second, diag-nostic data on hospitalized patients are regularly col-lected and analyzed on a national basis. 1 A definitionof technology dependence based on diagnosis couldbe specific (e. g., bronchopulmonary dysplasia) orbroad (e.g., any chronic lung disease).

There are a number of serious problems with usingthis approach. First, there is not a one-to-one cor-respondence between diagnoses and the need for long-term intensive nursing care. Table 26 lists a few of themany diseases (some of them very rare) that can leadto life-sustaining dependence on respiratory or nutri-tional support. Maintaining a comprehensive list mightbe very difficult, preventing some technology-depen-dent children from being included. Also, only a smallproportion of the children with these diseases requireprolonged technology supports. For example, of chil-dren with muscular dystrophy or cystic fibrosis, onlythose in the later stages require ventilators or even lessintensive respiratory support such as frequent suction-ing and oxygen (4,79). Thus, any definition that in-cludes diagnostic criteria must rely heavily on othercriteria as well.

Defining the population based on broader catego-ries of diagnoses or disorders would be considerablyless cumbersome but correspondingly less specific. It,too, would produce categories that are larger, prob-ably many times larger, than the population of chil-dren that is usually institutionalized and is dependenton life-sustaining medical devices.

Definition Based on Functional Limitation

Identifying disabled people, particularly the elderly,according to their functional limitations and their abil-ity to carry out certain activities of daily living hasbeen common for some time. Activity limitation ques-tionnaires have been used in surveys to provide na-

1 Diagnoses are coded onto hospital discharge abstracts, acc(}rd-ing to the conventions of the International Classification of Diseases,Qth Revision, Clinical Modification (I CD-9-CM ) coding system.These codes and other information from discharge abstracts are thenmaintained, summarized, and anal }~zed b}. a number of differentgovernment and private organizati[~ns,

8 5

Table 26.–Some Conditions That Can Lead toDependence on Respiratory or Nutritional Support

Conditions that can lead to dependence on respiratory support:brainstem aneurysmbronchopulmonary dysplasiacentral hypoventilation syndrome (Ondine’s curse)congenital heart diseasecystic fibrosisEllis-van Creveld syndromeencephalitisinterrupted phrenic nervesmultiple sclerosismuscular dystrophymyelodysplasianear-drowningnemaline rod myopathyneonatal asphyxiaPierre-Robin syndromePompe’s diseaseradiation lung damagesevere head injuryspinal muscular atrophysubglottic stenosisupper spinal cord injuryWerdnig-Hoffman disease

Conditions that can lead to dependence on nutritional support:Alagielle’s syndromechronic diarrheacongenital bowel defectcystic fibrosisfailure to thriveinflammatory bowel diseaseischemic bowel diseaseliver diseasemilklsoy protein intolerancemotility disordernecrotizing enterocolitisneoplasmsneurological disorders of swallowingradiation enteritisNOTE These diagnoses constitute only a partial IISt of conditions that can lead

to dependence on respiratory or nutritional support Conditions Iisted hereare actual diagnoses of children using these technologies, as recordedi n a national nutritional support database and a summary of chiIdrenserved by special Title V programs in three States in 1985

SOURCES L Heaphey, The Oley Foundation, Albany, NY, personal communi-cation, Aug. 21, 1986; M J Aitken and L A Aday, “Home Care for theChronically Ill and/or Technology Assisted Child An EvaluationModel, ” unpublished, November 1985

tional estimates of disability prevalence and severityin the population (63) and in studies of resource utili-zation among nursing home residents (133,140,182).Scales to measure activity limitation are relatively welldeveloped and seem to be good predictors of the in-tensity of required nursing and personal care servicesfor many elderly and disabled people.

The main limitations of these scales are that eachperson must be assessed individually and frequently,which is time-consuming and leaves considerable dis-cretion to the assessor;- and the scales are not well

suited to identifying the specific skilled nursing serv-ices an individual may need. z

Another approach could be to identify children bythe limitations of their normal body functions, suchas eating or breathing. This approach (the one usedin this technical memorandum) has intuitive appeal,because it would identify those children who use spe-cific technologies that replace or compensate for nor-mal body functions. The limitation of this approachis the difficulty in distinguishing levels of care neededin conjunction with the various technologies.

Definition Based on Type or Amountof Services Needed

A third approach might be to identify technology-dependent children by the type or amount of medicalservices they require. This might take the form ofdefining the population according to the need for cer-tain nursing services, such as catheterization. Or, itmight take the form of an indirect but explicit indica-tion of level of services needed, such as prior institu-tionalization or time in a neonatal intensive care unit.Finally, the population might be identified by the typeof long-term care plan required by its members. Forexample, the defined population might include chil-dren whose documented care plans specify hospice careand long-term chronic, continuous care, but not chil-dren requiring intermittent monitoring, occasional cri-sis care, or post-acute, recuperative care.

Considerations in Applyingthe Definition

Within the group of children identified as technol-ogy dependent, there will exist considerable variationin health and social needs. Ideally, an appropriate def-inition should be able to be applied in such a way thatdifferences in need among children can be discerned,with appropriate differences in benefits provided tothem. For example, two children might be equally ven-tilator dependent, but one might be able to dress andfeed himself while the other cannot. This example em-phasizes the value of functional assessment in apply-ing a definition equitably.

Home care may be feasible and desirable, but notcheaper than institutional care, for some children. If

‘Although there is considerable experience in applying specificassessments of a person’s ability to function, few of these applica-tions have assessed any limitations in basic body functions that re-quire nursing skills (e.g., the need for colostomy care). One surveythat includes these categories is currently being conducted on chil-dren with six types of disability and chronic illness (73).

8 7

these children are to be included, the definition shouldhave a mechanism for detecting those children forwhom the medical, psychological, and developmen-tal benefits of home care are high in relation to theadditional costs of home care. This criterion again im-plies that the definition should include some indica-tion of relative need and prognosis over time. A childwith a long-term or terminal illness, for example,might benefit more from the psychological and socialaspects of home care than a child recovering rapidlyfrom an acute condition, and consequently it mightbe desirable to be able to distinguish the former childfrom the latter for the purposes of providing benefits.

Meeting a particular definition need not necessarilyimply absolute access to a special program or set ofbenefits. A definition can also be thought of as ascreening mechanism to most easily identify the bulkof children who would benefit from extensive individ-ual assessment and a particular set of services. Onepossibility is that some fairly rigid, easily identifiedcharacteristics be used for rapid screening purposes,but that actual eligibility y for benefits be dependent onthe child’s functional or nursing assessment score,where activity limitations, degree of independence ca-pability, and limitations of body functions are allevaluated.

Appendix D

Alternative Settings of Care

Introduction

Most of the public discussion surrounding technol-ogy-dependent children, and most of the evidence dis-cussed in this technical memorandum contrasts twosettings of care for these children: hospital care, usu-ally in an acute-carel hospital; and home care with thechildren’s natural families. Within acute-care hospi-tals, technology-dependent children have access to thefull spectrum of medical services and equipment, mon-itoring, intensive nursing, professional backup, andemergency services that can be mobilized immediately.The children typically reside in intensive care units orspecialty wards (e. g., burn units), but they may residein general nursing wards (for children not requiringmechanical ventilation) or, sometimes, “step-down”transitional care wards.

In contrast to acute-level hospital care, home careoffers an environment most nearly like those in whichnon-technology-dependent children grow up. From theperspective of third-party payers of health care, tradi-tional home care offers the financial advantage of basicliving expenses that are borne by families. Many tech-nology-dependent children currently living at homehave highly trained and motivated parents and othercaregivers, whose time attending the child is also freeto the payer. At home, unlike in an institution, theneeded quantity of some services—e. g., the numberof paid nursing hours or the amount of respite care—depend as much on the social, psychological, and fi-nancial characteristics of the family as on the physi-cal condition of the child.

As more payers finance and perhaps even empha-size nonhospital care, other settings of care may alsobecome important for technology-dependent children.These alternative settings of care may serve three pos-sible purposes:

1. Transitional care for children who are movingfrom hospital to home or other long-term care.Transitional care is appropriate after the childhas become medically stable, while the home (orother setting) is being prepared for the child andthe myriad of financial and administrative de-tails are being completed. It usually includes anemphasis on training the family and graduallyincreasing the care the family provides. Transi-tional care can be provided in a special hospi-tal unit or in a separate rehabilitative or sub-acute care facility.

“’Acute-care hospital” as used here means a hospital that pro-vides complex medical care to patients and has an average lengthof patient stay of less than 30 days.

88

2.

3

Respite care for technology-dependent childrenwho are living at home. Institutional or fosterhome respite care may be an important optionin situations where qualified professional nursesare not available for home respite care, or wherea family vacation or emergency might make thehome an inappropriate setting of care for a shortperiod of time.Long-term care for children whose parents are un-—willing to have them home, negligent, abusive,or simply unable to cope with them. Extensivesupportive home services and counseling mayhelp parents cope with having a technology-dependent child at home. Even so, there will bea small group of children for whom care settingsother than a natural home must be explored asa long-term option,

A Michigan task force on home care guidelines forventilator-dependent children summarized the need forthese three types of options as follows:

As the child’s condition stabilizes, there should beprogression from the intensive care setting to one ofhabilitation/rehabilitation and eventually to a home-like environment. If the home is not a short or longterm option for care, alternative, home-like situationssuch as foster homes and small group homes must beexplored. Such institutional alternatives must alwaysremain an option to avoid crisis when home care ceasesto be feasible or is not longer the best option for theresponaut [ventilator-dependent child], family, com-munity and fiscal agencies (114).

Foster and Adoptive Care

For technology-dependent children who cannot re-turn to their natural family home (temporarily or long-term), a foster or adoptive home may become the set-ting of choice. If home care services and financing be-come more accessible to children whose families areable and willing to accept them and help care for them,children needing foster or adoptive homes are likelyto become a growing proportion of the residual institu-tionalized population.

Foster home need is likely to be greater among thispopulation than the child population in general, be-cause in addition to the need to find homes for chil-dren with incompetent or abusive parents, there is aneed to find homes for technology-dependent childrenwhose parents simply cannot accept their extensive dis-abilities. Furthermore, technology-dependent childrenare considered to be harder to place in foster homesthan other children. A concerted drive to serve all

89—

technology-dependent children at home would soonrun up against a shortage of available foster homes.

A lack of foster and adoptive homes may becomean equal or greater barrier to home care than a lackof sufficient home medical care benefits. The totalnumber of foster care homes in the United Statesdropped from 594,000 in 1977 to 187,680 in 1984, at-tributed in part to greater efforts to keep children withtheir natural parents (the number of foster children hasdropped from roughly 500,000 in the late 1970s toroughly 250,000 in 1984) but also to a greater dropin families willing to take in foster children (77,9 o).

The Federal Government provides matching subsi-dies with the States to families who adopt children withspecial needs, as well as to those families who providethem with foster homes (Public Law 96-272). Childrenin both categories for whom Federal subsidies are pro-vided are automatically eligible for Medicaid. Ironi-cally, those same children may not be eligible for Med-icaid if they remain with their natural families.

Community Group Homes2

The group home provides a community-basedoption, midway between institutionalization and afamily home, that could be attractive for sometechnology-dependent children if it were available.Group homes for adults who are ventilator-dependentdue to polio have existed in England and France fora number of years (67), and a few similar group homeshave recently opened in California (115), thoughapparently none are accepting young children atpresent. Louisiana is considering the establishment ofa group home that could accommodate ventilator-dependent children as well as other developmentallydisabled children (97).

For some children, the costs of group home caremight actually be lower than either hospital or familyhome care because a single trained nurse might be ableto care for more than one technology-dependent child.However, OTA knows of no present examples ofgroup homes that accept, or were designed for, tech-nology-dependent children. The relative rarity of suchchildren in the population suggests that group homesorganized for this purpose would probably be practicalsolutions only in densely populated areas.

Institutional Settings of Care

who cannot, for whatever reason, be placed in homecare. None of these are likely to be appropriate for allsuch children, nor are they likely to be preferred overhospital care (e.g., in a special long-term care unit) inall cases. But they may well be appropriate optionsfor a proportion of the population. Unfortunately,even when they might be appropriate, they are likelyto be unavailable.

Hospital Settings

Some acute-care hospitals have “step-down” unitswith the capacity for intensive care but an emphasison transition to a less intensive setting. A few hospi-tals have experimented with special wards in whichthe parent cares for the child during part or all of theday (51,119).

A fairly recent phenomenon is the development ofspecial pediatric respiratory centers, focused specifi-cally on the long-term care needs of medically stable,ventilator-dependent children. Such centers may be inacute-care tertiary hospitals, or in chronic care and re-habilitation hospitals. In both cases, the centers havegenerally been developed as “step-down” units thatserve the needs of ventilator-dependent children (andtheir families) in the transition to long-term commu-nity-based care. However, in practice many childrenlive on such wards indefinitely.

Children’s Hospital of Philadelphia has one of thebest-known pediatric respiratory units in an acute-carehospital. Similar units exist at a few other acute andlong-term care hospitals, though not all are exclusivelypediatric. 3 Ranchos Los Amigos Hospital, for exam-ple, a rehabilitation hospital that serves some childrenas well as adults, first established a special respiratoryunit in 1952 to better serve its long-term polio patientson respirators (2). Other pediatric respiratory units andintensive care units in extended-care hospitals exist(e.g., in Chicago, IL; Pittsburgh, PA; Washington,DC; and Baltimore, MD) or are being contemplated,but they are still rare.

Skilled Nursing and IntermediateCare Facilities

Skilled nursing facilities (SNFS) are an importantsource of care for many elderly, chronically ill peo-ple, but they do not generally have sufficient staff toprovide intensive nursing services and usually do not

With appropriate enhancement of facilities and staff,a multitude of subacute institutional settings could be ‘The distinction between rehabilitation, chronic care, and otherappropriate for many technology-dependent children types of long-term care hospitals is largely one of self-definition,

associated with how a hospital sees its mission. It is not clear thatrehabilitate ion hospitals are more likely than chronic care hospitals

‘Nlany group homes are apparentl y regulated as foster home~ (or vice \’ersa I to establish respiratory units.

90

provide an environment conducive to pediatric careand child development. The children most likely to befound in SNFs, where they are accepted at all, are thosewho are comatose or have low mobility but few con-stant skilled medical needs—perhaps daily medicationsor, at most, the need for multiple daily tube feedings(97). Intermediate care facilities (ICFs) are more likelyto care for children, but they are even less likely tobe able to provide intensive medical care than SNFs.

SNFs do sometimes accept technology-dependentadults. For instance, a 1985 survey of ventilator-dependent patients in long-term care facilities in Penn-sylvania documented 55 such patients in 4 nursinghomes and 1 skilled/intermediate care facility, all ofwhom were adults (94). Likewise, a few SNFs in Cali-fornia accept ventilator-dependent patients, but noneare known to accept such patients under age 16 (115).

At least two SNFs in the United States (one in NewJersey and one in Ohio) are equipped to serve childrenexclusively and can provide the complex care neededby technology-dependent children (139). In manyways, these SNFs are more similar to pediatric long-term care hospitals than they are to geriatric SNFs. Forexample, the pediatric SNF in New Jersey is staffed toprovide 6.5 nursing hours per patient per day, almostthree times the nursing intensity provided in geriatricSNFs in that State (139).

A trend towards making SNFs a more common siteof care for ventilator-dependent individuals and other

individuals (not necessarily children) needing post-acute complex care seems to be taking place. ThreeStates have recently proposed or established regula-tions for “super-SNF” subacute care, and at least 13others have instituted some reforms that can allow forextra payments to nursing homes for complex care pa-tients (88). California, for example, has proposed reg-ulations that will enable its Medicaid program to payfor care in specially certified SNF units that have ahigher level of nursing intensity and skill than normalSNF care (30). These subacute units will receive ahigher per diem rate than the usual SNF rate. A de-scription of California’s subacute care regulations ispresented in box E.

ICFs are less oriented toward complex medical carethan SNFs, and they are thus even less likely to ac-cept technology-dependent patients or to be able toprovide them with comprehensive care. ICFs are typi-cally institutions in which most residents require rela-tively little skilled nursing but considerable custodialcare (e. g., dressing, feeding, bathing, or just frequentattention). Homes for the mentally retarded are prob-ably the most familiar form of ICFs. There may be afew technology-dependent children who are alert butneed a highly protected environment and for whoman ICF with enhanced services and staff might be anappropriate setting.

91

Box E.—Medicaid Coverage of S@mcute Care in California

California has recently confronted the problem of appropriate institutional placement (and payment)for technologydependent persons when home care is not feasible. On August 19,1986, the State held pub-lic hearings on proposed Medicaid regulations establish~ a category of subacute care in skilled nursingfacilities (SNFS). (As of March 1987, it appeared that the Health Care Financing Administration will allowCalifornia to implement these regulations, but Federal approval was not yet final.) The revised text of theproposed regulations adopts additions to State Medicaid regulations, as follows.

Definition.-”Subacute level of care means a level of care needed by a patient who does not requireacute care but who requires more intensive licensed skilled nursing care than is provided to the majorityof patients in a skilled nursing facility.” A subacute care unit is “an identifiable unit of a skilled nursingfacility accommodating beds including contiguous rooms, a wing, a floor, or a building that is approvedby the Department for such purpose” (30). Subac@e C- U@J - subject tO all Of the State ce~ificationand licensing requirements appli~able to skilled nursing facilities. They may be in hospital-based or freestand-ing SNFs.

x.–’’suba~te - tits ddl ernploy sufficient licensed staff to provide a minimum daily averageof 4.8 actual licensed nursing hours per patient day for non ventilator dependent patients, and a minimumdaily average of 6,2 actual licensed nursing hours per patient day for ventilator dependent patients” (30).At least one registered nurse (RN) and one licensed vocational nur= (LVW must be on each sh~~ ~dthe ratio of LVNS to RNs cannot exceed 4 to 1. Both RNs and LVNS must have prior acute care experience.The unit must be able to provide, within the institution orthrcmgh contract, laboratory, X-ray, respiratory -therapy, and pharmacy services.

Services.-The proposed regulations define subacute care services as “a type of skilled nursing faciiityservice which is provided by a subacute care unit” (30). Patients must be under the care of a physicianwho makes frequent visits and must have 24-hour access to services in an acute-care hospital. They mustrequire special supplies or equipment, 24-hour nursing, and administration of three or more of the follow-ing treatment procedures:

1. traction and pin care for fractures;2. total parenteral nutrition;3. inpatient physical, occupational, and/or speech th=apy, at least 2 how per daY# 5 daYS per w*k;4* tube feeding;5. tracheotomy care with suctioning;6. oxygen therapy and/or inhalation therapy treatments at least four times per day;7. continuous or frequent intravenous therapy via a peripheral and/or central line;8. medically necessary isolation;9. debridement, packing, and medicated irrigation with or without whirlpool treatment; and

10. continuous mechanical ventilation for at least 50 percent of each day (30).Medicaid Payment.–The State calculated payment amounts for these new subacute facilities based

on hourly costs of nursing care and facility costs reported by SP?FS, adjusted by the more intense nursingrequirements of the subacute care units and predicted higher use of supplies and electricity (29). The resul-tant recommended maximum daily rates for SNF subacute level of care were:

● $221.93 for ventilator-dependent patients in hospital-based units,● $187.71 for other eligible patients in hospitid-b~sed units,. $140.62 for ventilatordependent patients in freestanding units, and● $109.62 for other eligible patients in freestanding units.

Appendix E

The Educational System as a Sourceof Health Care Services and Funding

Introduction

An important aspect of the cost of care for technol-ogy-dependent children in the home setting is that sub-stantial portions of this cost may be borne by publicschools. Public schools are mandated by Federal lawto provide educational and necessary related suppor-tive services to handicapped children (Public Law 94-142). Schools, through special education programs,regularly provide medical services such as physical andspeech therapy, medication administration, and evenurinary catheterization to children (179). Since schoolattendance may account for more than one-fourth ofa child’s time and care needs, one consequence fortechnology-dependent children of this Federal mandateis to shift substantial portions of the cost of a child’smedical care services from Federal to State and localgovernments (i.e., from Medicaid to public schools),and from private health insurers to the public.

The issue of who will pay for the medical care ofthese children in the schools is a growing one. Publicschools, pressed for funds, may often be reluctant topay for additional full-time nurses and special trans-portation vehicles and to assume legal liability formedical care during school hours. At the same time,private insurers—and Medicaid—will seek to minimizetheir costs of serving technology-dependent childrenat home by shifting financial responsibility to theschools. School districts may respond by serving mostof these children with occasional home visits in orderto avoid the extraordinary nursing costs and poten-tial lawsuits. Clear Federal and State policies on thisissue could greatly aid in minimizing total costs, en-couraging education in the environment most appro-priate to the individual child, allocating public dollarsappropriately (e. g., to Medicaid or to public schoolassistance), and preventing the emotional and finan-cial stress of legal battles.

Local Options for Complex MedicalCare in Schools

The issue of complex medical care for children at-tending public schools can be summarized in threequestions:

1. Where is this care provided?2. If it is provided in the school, who provides it?3. If it is provided in the school, who pays for it?

For some children, such as those with frequent anduncontrollable seizures, home education may be theonly feasible choice. In these cases, school districts mayprovide an individual teacher for a few hours a weekin the child’s own home. In such cases, the child’s nurs-ing needs are usually met by the normal home care-giver (a parent or home nurse), and reimbursement forthat care is indistinguishable from reimbursement forthe child’s usual home care. The school system paysfor the teacher’s time and transportation.

Many technology-dependent children receive theireducation in special classes or schools. In some of theseschools, nursing care is provided by full-time profes-sional nurses. In others, the teachers themselves, ora classroom aide, may be trained to provide these serv-ices. In either case, the school system generally paysfor the medical care, since the nurses or teachers areproviding care to a number of children.

The third setting of care and education for a tech-nology-dependent child is in a normal classroom. Thissetting is particularly appropriate for a child who isintellectually normal and has no mental or emotionalconstraints to maintaining a normal class schedule.However, the dilemmas regarding who shall provide,and pay for, the nursing care needed by a technology-dependent child are particularly acute in this setting.

Three options exist for providing nursing care in anormal school classroom. First, care may be providedby a school nurse. In most schools, a nurse providesservices to all children, and the nurse may even servemore than one school. The school district is responsi-ble for the salary of the nurse and any other costs asso-ciated with nursing services. Technology-dependentchildren, however, are characterized by their need forthe uninterrupted availability of nursing services. Fora school to provide such services, the school districtmust hire an additional full-time nurse or aide for eachtechnology-dependent child in the district, as well asthe regular nurse. Under this option, the insurer avoidsall nursing costs during school hours.

A second option for providing care in a normalclassroom is through a home nurse, whose salary andexpenses are covered through Medicaid or anotherthird-party payer, who accompanies the child whileat school. Although the effect of this option is the sameas the first—a full-time nurse for every technology-dependent child—it is clearly less desirable to the third-party payer, which must now pay the costs, and more

92

93—

desirable to the school district, which need not. If Med-icaid is paying for home care, the nurse would be paidfor through public funds in any case, but the sourceof the funds is administratively distinct.

A third option is to train teachers and other regularschool personnel to provide the necessary nursing care.Louisiana, for example, has chosen to train bus driv-ers, teachers, school nurses, and principals to performboth routine and emergency procedures that might beneeded by ventilator-dependent children (97). In thiscase the costs incurred are training costs, which maybe paid by the district, the health insurer, or someother source, and possibly the costs of a smaller stu-dent-to-teacher ratio in the classes that include thesechildren so that the teachers are not overburdened.

There are few Federal or State legal or administra-tive guidelines regarding who should pay for thesenursing services in the schools, or how they shouldbe provided, A survey of education and public health

departments in all 50 States (but not the District of Co-lumbia) regarding the provision of a specified list ofnursing practices’ found that 13 States (26 percent) hadno written State guidelines regarding the provision ofany of these services in the schools (184). An additional13 States had guidelines only for medication adminis-tration. Only six States (12 percent) had guidelinescovering all listed procedures. The remaining 18 States(36 percent) had written guidelines covering some, butnot all, of the specified procedures. The lack of com-prehensive guidelines in most States may reflect thefact that serving medically complex students is an is-sue that is usually addressed on the local rather thanthe State level (184).

‘The nursing practices included i n the su r~’ey ~~.ere catheteriza-tion, seizure management, medication administration, respirator }

care, tube feeding, positioning, colostomy ileostom~r care, and other(including allergy shots),

Appendix F

Glossary of Terms and Acronyms

Glossary of TermsAcute-care hospital: A hospital in which the average

length of stay is less than 30 days.Apnea monitor: A medical device that detects the ces-

sation of breathing.Asphyxia: Lack of oxygen resulting in suffocation or

near-suffocation.Augmented care: Care of a greater level, scope, or du-

ration than that normally provided under a particu-lar program or protocol.

Bronchopulmonary dysplasia: A chronic lung diseasein newborns, often defined by a characteristicappearance of the lungs on X-ray and the need formechanical ventilation for more than 4 weeks.

Case management: Coordination and oversight of thepackage of services provided to an individual. Casemanagement may be provided by an insurer, a pe-diatrician, a parent, a social worker, or some otherhealth care professional. The comprehensiveness ofcase management, and its goals, depend on themanager.

Cerebral palsy: A paralysis of varying severity thatresults from nonprogressiveor around birth.

Copayment: In insurance, awhereby the insured pays apoint of service or use (e.g.,Coinsurance.

Coinsurance: That percentage

damage to the brain at

form of cost-sharingspecific amount at the$10 per visit). See also

of covered medical ex-penses, after subtraction of any deductible, forwhich an insured person is responsible. Under Medi-care Part B, after the annual deductible has beenmet, Medicare will generally pay 80 percent of ap-proved charges for covered services and supplies;the remaining 20 percent is the coinsurance, forwhich the beneficiary is liable. Also see Copaymentand Deductible.

Colostomy: A surgical opening between the colon(part of the large intestine) and the surface of thebody. A colostomy is perfomed when normal defe-cation is difficult (e. g., because of lack of controlof the necessary muscles) or harmful.

Congenital: Present at birth. Congenital anomaliesusually refer to birth defects that result from imper-fect development during pregnancy.

Cystic Fibrosis: An inherited disorder caused by theproduction of a unique glycoprotein that results inabnormal mucous secretions. It is usually fatal be-fore age 20. Death is due to excess mucus in the lungsand to pancreatic insufficiency.

Deductible: The amount of health care charges thatan insured person must pay each year before he orshe is eligible for coverage.

End stage renal disease: Chronic renal failure that oc-curs when an individual irreversibly loses a suffi-cient amount of kidney function so that life cannotbe sustained without treatment. Chronic renal di-alysis, kidney transplant surgery, and continuousambulatory peritoneal dialysis are forms of therapy.

Gastrostomy: A surgical opening into the stomach. Agastrostomy tube allows food to be introduced di-rectly to the stomach, bypassing the mouth andthroat. A jejunostomy tube (which connects withthe top of the large intestine) may also perform thisfunction.

Hemophilia: A hereditary bleeding disorder distin-guished by a deficiency of one or more blood coagu-lation factors—e.g., Factor VIII (hemophilia A) orFactor IX (hemophilia B).

Home health care: Medical and related services pro-vided in the home.

Hospice care: Medical care rendered to terminally illpatients that is intended to be palliative rather thancurative.

Ileostomy: A surgical opening between the ileum (theend portion of the small intestine) and the surfaceof the body. See also colostomy.

Incidence: The frequency of new occurrences of a con-dition within a defined time period, usually 1 year.Compare prevalence.

International Classification of Diseases, 9th Revision,Clinical Modification (ICD-9-CM): A two-part sys-tem of coding patient medical information used inabstracting systems and for classifying patients intoDRGs for Medicare. The first part is a comprehen-sive list of diseases with corresponding codes com-patible with the World Health Organization’s list ofdisease codes. The second part contains procedurecodes, independent of the disease codes.

Intravenous therapies: Nutrients, medications, or othertreatments administered directly into the blood-stream (specifically, into a vein).

Long-term care: Health care of prolonged or indefi-nite duration. Long-term care hospitals usually havean average length of stay of 30 days or more.

Meconium aspiration: The existence of meconium, adark substance normally found in the intestine ofa full-term fetus, in the airway. Meconium aspira-tion can cause difficulties in breathing after birth.

Medical device: Any instrument, apparatus, or simi-

94

.

95

lar or related article that is intended to prevent, di-agnose, mitigate, or treat disease or to affect thestructure or function of the body.

Medical technology: The drugs, devices, and medicaland surgical procedures used in medical care, andthe organizational and support systems withinwhich such care is provided.

Multiple sclerosis: A progressive, crippling disease ofunknown cause that destroys the myelin sheath thatinsulates nerve cell axons. This results in slowednerve conduction. Symptoms commonly includeweakness, lack of coordination, and speech andvisual disturbances.

Muscular dystrophy: A group of inherited neuromus-cular diseases that result in the progressive deterio-ration of muscle function.

Neonatology: The medical specialty of newborn care.Parenteral nutrition: The intake of nutrients directly

into the bloodstream (intravenously), circumvent-ing the digestive tract. Strictly speaking, intramus-cular administration of nutrients is also parenteralnutrition, but the term as normally used in healthcare implies bloodstream administration.

Patent ductus arteriosus: Incomplete closing of a fetalblood vessel that allows blood that is low in oxy-gen to be returned to the body rather than to thelungs.

Phototherapy: The treatment of diseases, such as jaun-dice in newborns, with light.

Prevalence: In epidemiology, the number of cases ofdisease, infected persons, or persons with disabili-ties or some other condition, present at a particulartime and in relation to the size of the population.Also called “prevalence rate. ” Compare incidence.

Private duty nursing: Services provided by a profes-sional nurse to a patient who needs individual andcontinuous care beyond the level normally providedby a visiting nurse (in the home) or the nursing staff(of a hospital or skilled nursing facility).

Pulmonary surfactant: A substance present in the lungsthat aids in oxygen absorption.

Quadriplegia: Paralysis of all four limbs.Respite care: Care provided in order to give family

caregivers some relief. Respite care is a broad cate-gory that can include occasional home nursing orcustodial care or institutional care.

Sequelae: Aftereffects or secondary consequences.Shift nursing: Nursing provided in the home in hourly

shifts (usually 8-hour shifts), as distinguished fromnursing provided in visits (usually of an hour orless). See also private duty nursing.

Skilled nursing care: In this technical memorandum,any care that requires highly technical nursing skills,including care provided by nonprofessionals suchas parents trained in such skills.

Spina bifida: A birth defect of unknown cause that re-sults in incomplete or improper development of thespine, usually associated with the protrusion of thespinal cord through the bony spine.

Suctioning: As it applies to children with breathingdifficulties, suctioning is the removal of secretionsfrom the airway and is particularly important whenthe child has a tracheotomy tube (artificial airway)that could be blocked by these secretions.

Surfactant: See pulmonary surfactant.Technology-dependent children: Those children who

use a medical technology (embodied in a medicaldevice) that compensates for the loss of normal useof a vital body function, and who require substan-tial daily skilled nursing care to avert death or fur-ther disability.

Third-party payment: Payment by a private insureror government program to a medical provider forcare given to a patient.

Trachea: The airway extending from the back of themouth and nose to the bronchial tubes (which leadto the lungs).

Tracheotomy: A surgical opening into the trachea.A tracheotomy tube is an artificial airway (a tubein the trachea) that opens to the outside at thetracheotomy, where it can be connected to a me-chanical ventilator.

Urinary catheterization: The introduction of a tubeinto the urinary tract to withdraw urine.

Ventilator: A mechanical device used to assist in orcontrol respiration by delivering an appropriate vol-ume of gas to the airways or by promoting inspira-tion. In this report, it refers to both positive- andnegative-pressure devices that cause or help a per-son to breathe.

Glossary of Acronyms

AFDC —Aid to Families With DependentChildren

AIDS —acquired immunodeficiency syndromeBPD —bronchopulmonar y dysplasiaCHAMPUS—Civilian Health and Medical Program

of the Uniformed Services(Department of Defense)

CNS —central nervous systemCSHCN —Children with Special Health Care

NeedsDOD —US Department of DefenseEPSDT —Early and Periodic Screening,

Diagnosis and Treatment (program)ERISA —Employee Retirement Income

Security ActGAO –General Accounting Office (U.S.

Congress)

96

HCFA —Health Care FinancingAdministration (DHHS)

ICD-9-CM —International Classification ofDiseases, 9th Revision, ClinicalModification (ICD-9-CM)

ICF —intermediate care facilityICU —intensive care unitLPN —licensed practical nurseLVN —licensed vocational nurseMCH –Maternal and Child Health

(program)MNIL —medically needy income levelNIH –National Institutes of Health (PHS)NHIS —National Health Interview Survev

O T A

PFTH

PHCRDSREACH

RNSNFSPRANS

SSI

—Office of Technology Assessment(U.S. Congress)

—Program for the Handicapped(CHAMPUS)

—Pediatric Home Care—respiratory distress syndrome—Rural Efforts to Assist Children at

Home—registered nurse—skilled nursing facility—Special Projects of Regional and

National Significance—Supplemental Security Income

(program) (SSA)

References

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Ad Hoc Committee, Respiratory Care Section,American College of Chest Physicians, “Long-Term Mechanical Ventilation: Guidelines forManagement in the Home and at Alternate Com-munity Sites, ” Chest 90(l,supp.):lS-37S, July1986.Affeldt, J. E., Bower, A. G., Dail, C. W., et al.,“Prognosis for Respiratory Recovery in SeverePoliomyelitis, “ Arch. Phys. Med. Rehab. 38:290-295, January-June 1957.Aitken, M. J., and Aday, L. A., Home Care forthe Chronically Ill and /or Disabled TechnologyAssisted Child; An Evaluation Model, unpub-lished, November 1985.Alexander, M. A., Johnson, E. W., Petty, J., etal., “Mechanical Ventilation of Patients WithLate Stage Duchenne Muscular Dystrophy: Man-agement in the Ho me,” Arch. Phys. Med. Re-hab. 60:289-292, July 1979.American Medical Association, Department ofState Legislation, Public Affairs Group, StateHealth Legislation Report, vol. 14, No. 1 (Chi-cago, IL: AMA, February 1986).Antoniskis, A., Anderson, B. C., Van Volkin-burg, E. J., “Feasibility of Outpatient Self-Admin-istration of Parenteral Antibiotics, ” West. J.Med. 128(3):203-206, March 1978.Arnet tr R. H., and Trapnell, G. R., “PrivateHealth Insurance: New Measures of a Complexand Changing Industry, ” Health Care Financ-ing Review 6(2) :31-42, winter 1984.Association of State and Territorial Health Offi-cials Foundation, Public Health Agencies 1983,Vol. 1: Expenditures and Sources of Funds (Wash-ington, DC: ASTHO, August 1985).Avery, M. E., “Antepartum Glucocorticoids andPrevention of Respiratory Distress Syndrome, ”paper presented at the XVI1l International Con-gress of Pediatrics, Honolulu, HI, July 7, 1986.Avery, M. E., Boston Children’s Hospital, Bos-ton, MA, personal communication, July 1986.Avery, M. E., Tooley, W. H., Keller, J. B., et al.,“Is Chronic Lung Disease in Low Birth WeightInfants Preventable? A Survey of Eight Centers, ”Pediatrics 79(1):26-30, January 1987.Bach, J., Albe, A., Pilkington, L. A., et al., “Long-Term Rehabilitation in Advanced Stage of Child-hood Onset, Rapidly Progressive Muscular Dys-trophy, ” Arch. Phys. Med. Rehab, 62:328-331,1981.Banaszak, E. F., Travers, H., Frazier, M., et al.,

14.

15.

16.

17.

18

20!

21.

22.

23.

24

25.

26.

27.

28.

29.

“Home Ventilator Care, ” Respiratory Care 26(12):1262-1268, December 1981.Barker, E., Fairfax County School District, Fair-fax, VA, personal communication, July 1986.Bates, J., San Diego Children’s Hospital, SanDiego, CA, personal communication, July 1986.Beckett, J., Child Health Specialty Clinics, IowaCity, IA, personal communication, April 1986.Bennett, F. C., Robinson, N. M., and Sells, C. J.,“Growth and Development of Infants WeighingLess Than 800 Grams at Birth, ” Pediatrics 71(3):319-323, 1983.Berriman, A. E., “CHAMPUS Homecare Cov-erage, ” Homecare Rental/Sales 6(9):74-78, Sep-tember 1984.B1ue Cross/Blue Shield Association, Home HealthCare Survey Report (Chicago, IL: Blue Cross/Blue Shield Association, April 1985).Blue Cross/Blue Shield Association, InfusionTherapies in Home Health Care (Chicago, IL:Blue Cross/Blue Shield Association, January1986).Boodman, S. G., “Born Dying: AIDS’ 2nd Gen-eration, ” Washington Post, p. Al, A8, Mar. 23,1987.Brakebill, J. I., Robb, R. A., Ivey, M. F., et al.,“Pharmacy Department Costs and Patient ChargesAssociated With Home Parenteral NutritionProgram, ” Am. J. Hosp. Pharm. 40(2):260-263,February 1983.Brootern, D., Kumar, S., Brown, L. P., et al.,“A Randomized Clinical Trial of Early Hospi-tal Discharge and Home FolIow-Up of Very-Low-Birth-Weight Infants, ’f IVew Eng. J. Med.315(15):934-939, Oct. 9, 1986.Buckwald, S., Zorn, W. A., and Egan, E. A.,“Mortality and Follow-Up Data for NeonatesWeighing 500 to 800 G at Birth, ” Am. J. Dis.Child. 138:779-782, 1984.Burr, B. H., Guyer, B., Todres, I. D., et al., “HomeCare for Children on Respirators, ” N. Eng. J.Med. 309(21):1319-1323, Nov. 24, 1983.Butler, J. A., Budetti, P., McManus, M. A., etal., “Health Care Expenditures for ChildrenWith Chronic Illnesses, ” Issues in the Care ofChildren With Chronic Illness, N. Hobbs andJ.M. Perrin (eds. ) (San Francisco: Jossey-Bass,1985) .Cabin, B., “Cost-Effectiveness of Pediatric HomeCare, ” Caring IV(5) :48-51, May 1985.California Department of Health Services, Medi-

99

100

Cal Policy Division, “Reimbursement Study:Subacute Level of Skilled Nursing Facility Care, ”Report No. 01-86-3, August 1986.

30. California Department of Health Services, “No-tice of Public Availability Pursuant to Govern-ment Code Section 11346.8 of Changes to Emer-gency Regulations and Supporting Documentsand Information Regarding Subacute Level ofCare (R-54 -85),” Aug. 27, 1986.

31. Cane, R. D., and Shapiro, B. A., “MechanicalVentilator Support, ” J.A.M.A. 254(1):87-92,July 5, 1985.

32. Care for Life, “Life Sustaining Technologies andthe Elderly: Prolonged Mechanical Ventilation,Data on Utilization and Cost, ” paper preparedfor the Office of Technology Assessment, U.S.Congress, Washington, DC, October 1985.

33. Chollet, D. J., Employer-Provided Health Ben-efits: Coverage, Provisions and Policy Issues(Washington, DC: Employee Benefit ResearchInstitute, 1984).

34. Cleverly, G., Human Services Department,Santa Fe, NM, personal communication, August1986.

35. Colbert, A. P., and Schock, N. C., “RespiratorUse in Progressive Neuromuscular Diseases, ”Arch. Phys. Med. Rehab, 66:760-762, Novem-ber 1985.

36. Colton, M., and Gittler, J., The Title V StatePrograms and the Provision of Case Manage-ment Services for Children With Specia! HealthCare Needs (Iowa City, IA: National Maternaland Child Health Resource Center, 1986).

37. Cooper, R., Department of Health and SocialServices, Madison, WI, personal communica-tion, February 1986.

38. Culley, T., Aetna Life and Casualty, Hartford,CT, personal communication, April 1986.

39. Curran, F. J., “Night Ventilation by Body Respi-rators for Patients in Chronic Respiratory Fail-ure Due to Late Stage Duchenne Muscular Dys-trophy, Arch. Phys. Med. Rehab. 62:270-274,1981.

40. Curran r J. W., Morgan, W. M., Hardy, A. M.,et al., “The Epidemiology of AIDS: Current Sta-tus and Future Prospects, ” Science 229:1352-1357, Sept. 27, 1985.

42. Dorm, S., “Cost Effectiveness of Home Manage-ment of Bronchopulmonary Dysplasia” (letter),Pediatrics 70(2):330-331, August 1982.

43. Doran, T. A., Surfer, P., MacMurray, B., et al.,“Results of a Double-Blind Controlled Study onthe Use of Betamethasone in the Prevention of

Respiratory Distress Syndrome, ” Am. J. Obstet.Gyn. 136:313-320, 1980.

44. Doyle, L. W., Kitchen, W. H., Ford, G. W., etal., “Effects of Antenatal Steroid Therapy onMortality in Very Low Birth Weigh Infants, ” J.Pediatrics 108:287-292, 1986.

45. Driscoll, J.M, Driscoll, Y. T., Steir, M. E., et al.,“Mortality and Morbidity in Infants Less Than1,001 Grams Birth Weight, ” Pediatrics 69(1):21-26, 1982.

46, Dudrick, S. J., Winmore, D. W., Vars, H. M,, etal., “Long-Term Total Parenteral Nutrition WithGrowth and Development and Positive Nitro-gen Balance, ” Surgery 64:134-142, 1968.

47. Duggar, B., Pollatsek, J., Han, C., et al., “Fi-nancing Care of Chronically 111 and DisabledChildren in Home and Other Ambulatory CareSettings, ” LaJolla Management Corp., Rock-ville, MD, April 1986.

48. Dzierba, S. H., Mirtallo, J. M., Graner, D. W.,et al,, “Fiscal and Clinical Evaluation of HomeParenteral Nutrition, ” Am. J. Hosp. Pharm.41:285-291, February 1984.

49. Ellman, F. M., National Association of State Di-rectors of Special Education, Inc., WashingtonDC, personal communication, January 1986.

50. Eron, L. J., “Intravenous Antibiotic Administra-tion in Outpatient Settings, ” Infectious Dis. 14(1):4-11, January 1984.

51. Evans, R. G., and Robinson, G. C., “An EconomicStudy of Cost Savings on a Care-By-ParentWard, ” Medical Care 21(8):768-782, August1983.

52. Farley, P. J., “Who Are the Underinsured?” Health& Society 63(3):476-503, 1985.

53. Farley, P., and Walden, D., “The Privately In-sured Under Age 65: Cost-Sharing, Depth ofBenefits and Other Aspects of their Health In-surance, ” presented to the American PublicHealth Association, 1983.

54. Feinberg, E. A., “Family Stress in Pediatric HomeCare,” Caring IV(5) :38-41, May 1985.

55. Feldman, J., and Tuteur, P. G., “MechanicalVentilation: From Hospital Intensive Care toHome, ” Heart & Lung 11(2):162-165, March/April 1982.

56. Fitzgerald, J. M., John Hancock Mutual Life In-surance Co., Boston, MA, personal communi-cation, March 1986.

57, Foster S., and Hoskins, D., “Home Care of theChild With a Tracheotomy Tube, ” Peal. Clin.No. Am. 28(4):855-857, November 1981.

58. Fox, H. B., and Yoshpe, R., “Private Health In-

59.

60.

61.

62.

63

64

65.

6 6

67.

68.

69

71.

72.

—

surance Coverage of Chronically Ill Children, ‘r

prepared for the National Center for Network-ing Community-Based Services, GeorgetownUniversity Child Development Center, Wash-ington, DC, March 1986.Fox, H. B., and Yoshpe, R., “Technology-Depen-dent Children’s Access to Medicaid Home CareFinancing, ” prepared for the Office of Technol-ogy Assessment, U.S. Congress, August 1986.Frame, C., “Parent of Ventilator-DependentChild Welcomes Waiver, ” AARTimes 8(4):55,April 1984.Frates, R. C., Jr., and Splaingard, M. L., “HomeCare for Children on Respirators” (letter), N.Eng. J. Med., 310(17):1126-1127, Apr. 26, 1984.Frates, R. C., Jr., Splaingard, M. L., Smith, E. O.,et al., “Outcome of Home Mechanical Ventila-tion in Children, ” J. Pediatrics 106(5):850-856,May 1985.Gallin, R. S., and Given, C. W., “The Conceptand Classification of Disability in Health Inter-view Survey s,” Inquiry XIII(4) :395 -4O7, Decem-ber 1976.Gaumer, G. L., Birnbaum, H., Pratter, F., et al.,“Impact of the New York Long-Term HomeHealth Care Program, “ Medical Care 24(7):641-653, July 1986.Giovannoni, R., “Chronic Ventilator Care: FromHospital to Home, ” Respiratory Therapy 29(7):29-33, July/August 1984.Gitlin, J. D., Soil, R. F., Parad, R. B., et al., “ARandomized Controlled Trial of Exogenous Sur-factant for the Treatment of Hyaline MembraneDisease, ” Pediatric Research 20:429A, 1986.Goldberg, A. I., “Home Care for the Ventilator-Dependent Person in England and France, ” Car-ing IV(5) :34-36, May 1985.Goldberg, A. I., “The Illinois Plan,” Report ofthe Surgeon General’s Workshop on ChildrenWith Handicaps and Their Families, DHHSPub. No. PHS-83-50194 (Washington, DC: U.S.Department of Health and Human Services,Public Health Service, 1983).Goldberg, A. I., Faure, E. A., Vaughn, C. J., etal., “Home Care for Life-Supported Persons: AnApproach to Program Development, ” J. Pedi-atrics 104(5):785-795, May 1984.Goldberg, R. N., and Bancalari, E., “Bronchopul-monary Dysplasia: Clinical Presentation and theRole of Mechanical Ventilation, ” RespiratoryCare 31(7):591-598, July 1986.Gortmaker, S. L., and Sappenfield, W., “ChronicChildhood Disorders: Prevalence and Impact, ”Peal. C/in. No. Am. 31(1):3-18, February 1984.

73.

74.

75.

76

77,

77a,

78,

79.

80.

81.

82.

83.

84.

85.

86

101

Guyot, D., Albert Einstein College of Medicine,Bronx, NY, personal communication, June 1986.Hack, M., and Fanaroffr A. A., “Changes in theDelivery Room Care of the Extremely Small In-fant ( <750 g) —Effects on Morbidity and Out-come,” N. Eng. J. Med. 314(10):660-664, 1986.Hadley, J., and Feder, J., “Hospital Caring forChildren: Financial Status and Care to the Poor,1980 and 1982, ” Georgetown University Cen-ter for Health Policy Studies, working paper No.655-1, May 1985.Halliday, H. L., McClure, G., Reid, M. M., etal., “Controlled Trial of Artificial Surfactant toPrevent Respiratory Distress Syndrome, ” Lancet1:476-478, 1984.Hardy r D. R., “Adoption of Children With Spe-cial Needs: A National Perspective, ” AmericanPsychologist 39(8):901-904, August 1984.Harris, J., “Price Rules for Hospitals, ” Bel/Jour-nal of Economics 10:224, 1979.Harris, L. F., Buckle, T. F., and Coffey, F. L., “In-travenous Antibiotics at Home, ” Southern Med.J. 79(2):193-196, February 1986.Harwood, I., Hernried, L., Lemen, R., et al.,“Therapeutic Profiles of Cystic Fibrosis PatientsObtained From a Multi-Center Clinical Encoun-ter, ” Chest 86(6, supp):4814, abstract, June1986.Health Insurance Association of America, NewGroup Health Insurance (Washington, DC:HIAA, 1984).Heaphey, L. L., The Oley Foundation, Albany,NY, personal communication, August 1986.Hedrick, S. C., and Inui, T. S., “The Effective-ness and Cost of Home Care: An InformationSynthesis, ” Health Services Research 20(6, part11):851-880, February 1986.Hewitt Associates, Salaried Employee BenefitsProvided by Major U.S. Employers: A Compar-ison Study, 1979 Through 1984 (Lincolnshire,IL: Hewitt Associates, 1985).Horbar, J. D., “A Multicenter Survey of 28 DaySurvival Administration in Infants 701-1500Grams, ” paper presented at the Ross Labora-tories Conference on Topics on Neonatality,Washington, DC, Dec. 7-9, 1986.Hoskins, E. M., Elliott, E., Sherman, A. T., et al.,“Outcome of Very Low-Birth Weight InfantsBorn at a Perinatal Center, ” Am. J. Obstet.Gyn. 145:135-140, 1983.Houts, P. S., Lipton, A., Harvey, H. A., et al.,“Nonmedical Costs to Patients and Their Fam-ilies Associated With Outpatient Chemotherapy, ”Cancer 53(11):2388-2392, June 1, 1984.

102

87.

88.

89.

90,

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

101

Hunter, G., and Johnson, S. H., “Physical Sup-port Systems for the Homebound Oncology Pa-tient, ” as cited in Joint Commission on Accred-itation of Hospitals, “Guidelines for EvaluatingHome Infusion Therapy Services, ” prepared forBlue Cross and Blue Shield Association, Chi-cago, IL, 1986.Intergovernmental Health Policy Project, “StateRegulation of Subacute Care: A 50-State Sur-vey, ” Focus On . . ., No. 12, October 1986.Jobe, A., “Surfactant Treatment for RespiratoryDistress Syndrome, ” Respiratory Care 31(6):467-479, June 1986.Jordan, M., “Foster Parent Scarcity Causing Cri-sis in Care, ” Washington Post, p, Fl, F8, July20, 1986.Kacmarek, R. M., comments following Kacmarek,R. M., and Thompson, J. E., “Respiratory Careof the Ventilator-Assisted Infant in the Home, ”Respiratory Care 31(7):613, July 1986.Kahn, L., “Ventilator-Dependent Children Head-ing Home,” Hospitals 58(5):54-55, Mar. 1, 1984.Kaufman, J., Coordinating Center for Homeand Community Care, Inc., Millersville, MD,personal communication, May 1986.Keister, H., Department of Public Welfare, Har-risburg, PA, personal communication, August1986.Kelly, D. A., Armitstead, J, A., and Walker-Smith, J, A., “Home Parenteral Nutrition forChildren” (letter), Lancet 2(8509):746, Sept. 27,1986.Kind, A. C., Williams, D. N., Persons, G., et al.,“Intravenous Antibiotic Therapy at Home, ”Arch. Int. Med. 139:413-415, April 1979.Kirkhart, K., Children’s Hospital, New Orleans,LA, personal communications, February 1986and December 1986.Kisker, C. T., The Needs of Children With Can-cer: A Comprehensive View (Iowa City, IA: TheUniversity of Iowa, 1983).Klaus, M. H., and Fanaroff, A. A., Care of theHigh-Risk Neonate, 2d ed. (Philadelphia, PA:W.B. Saunders, 1979).Konishi, M., Fujiwara, T., Naito, T., et al.,“Surfactant Replacement Therapy in NeonatalRespiratory Distress Syndrome (RDS)–A Pro-spective Collaborative Study, ” paper presentedat the XVIII International Congress of Pedi-atrics, Honolulu, HI, July 7, 1986.Laudicina, S. S., A Comparative Survey of Med-icaid Hospital Reimbursement Systems for Inpa-tient Services, State by State, 1980-1985 (Wash-ington, DC: The Urban Institute, 1986).

102.

103,

104,

105.

106.

107.

108

109.

110.

111

112.

113.

114.

115.

Lauri, B., “Polio Survivors Prove the Value ofHome Care Services, ” AARTimes 8(4):47-50,April 1984.Lee, G, B., “Development of a Home HealthCase Management Program in Los AngelesCounty, ” California Children Services, Depart-ment of Health Services, County of Los Angeles,Oct. 31, 1983.Lis, E., Crippled Children’s Services, Chicago,IL, personal communication, April 1986.Lyons, J. M., Falkenbach, K., and Cerra, F, B.,“Home Parenteral Nutrition With Full-TimeHome Care Nurses, ’’J.P.E.N. 5(6):528-530, No-vember/December 1981.MacQueen, J. C., “Implications for Care: TheTitle V Perspective, ” Report of the SurgeonGeneral’s Workshop on Children With Handicapsand Their Families, U.S. Department of Healthand Human Services, Public Health Service,DHHS Pub. No. (PHS) 83-50194 (Washington,DC: U.S. Government Printing Office, 1983).MacQueen, J. C., “Future Directions of Servicesfor Children With Special Health Care Needs, ”paper presented at the Annual Meeting of theAssociation for MCH/CC Programs, Washing-ton, DC, Mar. 19, 1984.MacQueen, J. C., “Alternatives to Hospital Care, ”unpublished, Aug. 5, 1986.Matthews, L, W., and Drotar, D., “Cystic Fibro-sis —A Challenging Long-Term Disease, ” Peal.Clin. No. Am. 31(1):133-152, February 1984.Mayhue, H., Bureau of Data Management andStrategy, Health Care Financing Administra-tion, U.S. Department of Health and HumanServices, Baltimore, MD, personal communica-tion, Feb. 11, 1987.McDonnell, P., Division of National Cost Esti-mates, Health Care Financing Administration,U.S. Department of Health and Human Serv-ices, Baltimore, MD, personal communication,Mar. 5, 1987.McInerny, T., “The Role of the General Pedi-atrician in Coordinating the Care of ChildrenWith Chronic Illness, ” Peal, Clin. No. Am.31(1):199-209, February 1984.McManus, M. A., Melus, S. E., Norton, C. H.,et al., Guide to National Data on Maternal andChild Hea/th (Washington, DC: McManus HealthPolicy, Inc., 1986).Michigan Department of Public Health, Divi-sion of Services to Crippled Children, Guidelinesfor Home Care of Responauts, Task Force Re-port Aug. 14, 1984.Middlemiss, R., Casa Colina Skilled Rehabili-

—

116,

117,

118,

119.

120.

121.

122

123.

124.

126.

127.

128.

129.

130.

tation Center, Los Angeles, CA, personal com-munication, August 1986.Mikol, M., SKIP (Sick Kids [need] InvolvedPeople) of New York, Inc., New York, NY, per-sona] communication, April 1986.Milofsky, C., and Elworth, J. T., “CharitableAssociations, ” Issues in the Care of ChildrenWith Chronic Illness, N. Hobbs and J.M. Per-rin (eds. ) (San Francisco, CA: Jossey-Bass, 1985).Monahan, C. A., “Safety and Monitoring of HomeVentilators, ” paper presented to the InteragencyTechnical Committee Working Group on LungDiseases, Bethesda, MD, Feb. 17, 1987.Monahan, G. H., and Schkade, J. K., “Compar-ing Care by Parent and Traditional NursingUnits, ” Pediatric Nursing 11(6):463-468, Nov-ember-December 1985.Murray, N., Omaha, NE, personal communi-cation, July 1986.National Association of Employers on HealthCare Alternatives, Survey of National Corpo-rations on Health Care Cost Containment, D.J.Chollet, Employer-Provided Health Benefits:Coverage, Provisions and Policy Issues (Wash-ington, DC: Employee Benefit Research Insti-tute, 1984).New York Academy of Medicine, “Registry ofPatients on Home Total Parenteral Nutrition(TPN), ” unpublished data, 1982.Newacheck, P. W., Budetti, P. P., and Half on,N., “Trends in Activity-Limiting Chronic Con-ditions Among Children, ” Am. J. Public Health76(2):178-184, February 1986.Newacheck, P. W., Budetti, P. P., and McMa-nus, P., “Trends in Childhood Disability, ” Am.J. Public Health 74(3):232-236, March 1984.Oley Foundation, Inc., “Nutritional Support andHydration for Critically and Terminally Ill El-derly: Utilization in the Home, ” contract paperprepared for the Office of Technology Assess-ment, U.S. Congress, Washington, DC, Septem-ber 1985.Olson, H. H., “Home Care of the Ventilator-Dependent Child, ” Caring V:48-54, December1984.P e r r yr B . , and Lierman, C., “Hospital BasedPediatric Home Care: Review of the Issue anda Case Example, ” report submitted to the RobertWood Johnson Foundation, June 28, 1985.Pierce, P. M., and Freedman, S. A., “The REACHProject: An Innovative Health Delivery Modelfor Medically Dependent Children, ” Children’sHealth Care 12(2):86-89, Fall 1983.Poretz, D. M., Eron, L. J., Goldenberg, R. I., etal., “Intravenous Antibiotic Therapy in an Out-

131.

132,

133,

134,

135,

136,

137,

138.

139.

140.

141.

142.

143.

144.

145.

103

patient Setting, ” J. A.M.A. 248(3):336-339, July16, 1982.Rehabilitation Institute of Chicago, NationalFoundation—March of Dimes, Care for Life,and Rehabilitation Gazette, “What Ever Hap-pened to the Polio Patient?” Proceedings of aninternational symposium, Chicago, IL, Oct. 14-16, 1981.Rehm, S. J., and Weinstein, A. J., “Home Intra-venous Antibiotic Therapy: A Team Approach, ”Ann. Int. Med. 99(3):388-392, September 1983.Rohrer, J. E., Smith, D.W. and Hogan, A. J.,“Long-Term Care Patient Assessment: Evalua-tion of a Methods Effect, ” Medical Care 24(8):687-693, August 1986.Rosenbaum, S., “Children and Private HealthInsurance, ” Recent Developments in the Financ-ing and Organization of Health Care: Implica-tions for Children (in press).Rosenbaum, S., “The Prevention of Infant Mor-tality: The Unfulfilled Promise of Federal HealthPrograms for the Poor, ” Clearinghouse Reviewpp. 726-729, November 1983.Roughan, J., The Equitable Life Assurance Co.,La Brea, CA, personal communication, June1986.Rucker, B. B., and Holmstedt, K. A., Home ]n-fusion Therapy Industry (San Francisco, CA:Hambrecht & Quist, April 1984).Saigal, S., Rosenbaum, P., Stoskopf, B., et al.,“Outcome in Infants 501 to 1000 Gm BirthWeight Delivered to Residents of the McMasterHealth Region, ”J. Pediatrics 105:969-976, 1984.Scalese, J. P., Voorhees Pediatric Facility, Gibbs-boro, NJ, personal communication, December1986.Schneider, D. P., Fries, B. E., Foley, W. J., et al.,“Development of RUG II Case Mix Measure-ment System for Long Term Care, ” Proceedingsof the Annual Conference of the Health ServiceDivision, IIE (Chicago, IL: American HospitalAssociation, February 1986).Schraeder, B. D., “A Creative Approach to Car-ing for the Ventilator-Dependent Child, ” M. C.N.4:165-170, May/June 1979.Shannon, K., SKIP (Sick Kids [need] InvolvedPeople), Inc., Severna Park, MD, personal com-munication, December 1986.Shapiro, D., Rochester, NY, personal commu-nication, 1986.Shapiro, S., McCormick, M. C., Starfield, B. H.,et al., “Changes in Infant Morbidity AssociatedWith Decreases in Neonatal Mortality, ” Pedi-atrics 72(3):408-415, 1983.Shapiro, D. L., Rosenberg, P., “The Effect of

104—

Federal Regulations Regarding HandicappedNewborns: A Case Report, "J.A.M.A. 252:2031-2033, 1984.

146. Sherman, M., McLean, D., Zorn, E., et al.,“Home Care for Ventilator-Dependent Patients:A Rational Option” (abstract), Respiratory Care27(10):1252, October 1982.

147. Sivak, E. D., Cordasco, E. M., and Gipson,W. T., et al., “Pulmonary Mechanical Ventila-tion at Home: A Reasonable and Less ExpensiveAlternative,” Respiratory Care 28(1):42-49, Jan-uary 1983.

148. Splaingard, M. L., Frates, R, C., Harrison, G. M.,et al., “Home Positive-Pressure Ventilation, ”Chest 84(4):376-382, October 1983.

149. Splaingard, M. L., and Splaingard, D., “HomeCare of the Ventilator-Dependent Child: Reali-ties and Limitations, ” Pride Institute Journal ofLong Term Home Health Care 5(2):15-20, spring1986.

150. Stein, R. E., and Jessop, D, J., “Does PediatricHome Care Make a Difference for ChildrenWith Chronic Illness? Findings From the Pedi-atric Ambulatory Care Treatment Study, ” Pedi-atrics 73(6):845-853, June 1984.

151. Stiver, H. G., Telford, G. O., Mossey, J. M., etal., “Intravenous Antibiotic Therapy at Home, ”Ann. Int. Med. 89(5, part 1):690-693, Novem-ber 1978.

152. Sulvetta, M. B., and Swartz, K., The Uninsuredand Uncompensated Care (Washington, DC:National Health Policy Forum, George Wash-ington University, June 1986).

153, Tarnow-Mordi, W., and Wilkinson, A., “Me-chanical Ventilation of the Newborn, ” Br. Med.J. 292:575-576, 1986.

154. Taussig, L. M., “Bronchopulmonary Dysplasia, ”Follow-Up of the High Risk Newborn—A Prac-tical Approach, E.J. Sell (cd. ) (Springfield, IL:Charles C. Thomas, 1980).

155. Thomas, E., Blue Cross/Blue Shield Federal Em-ployees Benefit Program, Washington, DC, per-sonal communication, 1986.

156. Thomas, R. B., “Ventilator Dependency Conse-quences for Child and Family, ” unpublishedPh.D. dissertation, University of Washington,1986.

157. Tschumper, P., Department of Health and So-cial Services, Madison, WI, personal commu-nication, April 1986.

158, U.S. Congress, General Accounting Office, Mater-nal and Child Health Block Grant: ProgramChanges Emerging Under State Administration,GAO/HRD-84-35 (Washington, DC: U.S. Gen-eral Accounting Office, May 7, 1984).

159. U.S. Congress, House of Representatives, Com-mittee on Ways and Means, Background Mate-rial and Data on Programs Within the Jurisdic-tion of the Committee on Ways and Means,1986 WMCP 99-14, Mar. 3, 1986.

160. U.S. Congress, House of Representatives, SelectCommittee on Aging, The “Black Box” of HomeCare Quality, August 1986.

161. U.S. Congress, Office of Technology Assess-ment, Federal Policies and the Medical DevicesIndustry, OTA-H-230 (Washington, DC: U.S.Government Printing office, October 1984).

162. U.S. Congress, Office of Technology Assess-ment, Technology and Child Health, draft,Washington DC, 1987.

163. U.S. Department of Commerce, Bureau of theCensus, 1976 Survey of Institutionalized Per-sons: A Study of Persons Receiving Long-TermCare, Current Population Reports Special Studies,series P-23, No, 69, June 1978.

163a. U.S. Department of Commerce, Bureau of theCensus, Current Population Survey, unpublisheddata, March 1986.

164. U.S. Department of Defense, Office of the As-sistant Secretary of Defense, Civilian Health andMedical Program of the Uniformed Services(CHAMPUS), DOD 6010.8-R, March 1986.

165. U.S. Department of Education, Office of Spe-cial Education, 9th Annual Report to Congresson the Implementation of the Education of theHandicapped Act, 1987.

166. U.S. Department of Health and Human Serv-ices, Health Care Financing Administration, Bu-reau of Data Management and Strategy, “EndStage Renal Disease Patient Profile Tables, ”1985.

167. U.S. Department of Health and Human Serv-ices, Health Care Financing Administration,Health Care Financing Program Statistics: Anal-ysis of State Medicaid Program Characteristics,1984 (Baltimore, MD: DHHS, August 1985).

168. U.S. Department of Health and Human Serv-ices, Health Care Financing Administration,Medicaid Data Book 1983 (Washington DC:DHHS, 1985).

169. U.S. Department of Health and Human Serv-ices, Public Health Service, Centers for DiseaseControl, personal communication, 1986.

170. U.S. Department of Health and Human Serv-ices, Public Health Service, National Center forHealth Statistics, A Study of Infant MortalityFrom Linked Records By Birth Weight, Periodof Gestation, and Other Variables, United States,1960 Live-Birth Cohort, DHEW Pub. No. (PHS)79-1055, (Hyattsville, MD: NCHS, May 1972).

105—

171.

172

173.

174

175,

176,

177,

178.

—

U.S. Department of Health and Human Serv-ices, Public Health Service, National Center forHealth Statistics, “Advance Report of Final Na-tality Statistics, 1984, Table 24, ” Monthly Vi-tal Statistics Report, 35 (4, suppl. ):37, July 18,1986.U.S. Department of Health and Human Serv-ices, Public Health Service, National Center forHealth Statistics, “Detailed Diagnoses and Pro-cedures for Patients Discharged from Short-StayHospitals, United States, 1984, ” Vital andHealth Statistics Series 13, No. 86, DHHS Pub.No. 86-1747, (Washington, DC: U.S. Govern-ment Printing Office, April 1986).U.S. Department of Health and Human Serv-ices, Public Health Service, National Institutesof Health, Infantile Apnea and Home Monitor-ing (draft), Bethesda, MD, 1986.U.S. Department of Health and Human Serv-ices, Public Health Service, National Institutesof Health, Prevention of Respiratory DistressSyndrome, Effect of Antenatal DexamethasoneAdministration, NIH Pub. No. 85-2695 (Wash-ington, DC: NIH, August 1985).U.S. Department of Labor, Bureau of Labor Sta-tistics, Employee Benefits in Medium and LargeFirms, 1984, Bulletin 2237, (Washington, DC:U.S. Government Printing Office, June 1985).Valdez, K., Human Services Department, SantaFe, NM, personal communication, July 1986.Van Dyck, P. C., “The Crippled Children Serv-ices Program, ” Workshop on Financing HealthCare for Handicapped Children (Rockville, MD:U.S. Department of Health and Human Serv-ices, January 1984).Ventilator Assisted Care Program, Children’s

179%

180.

181.

182.

183.

184.

185.

186.

187.

Hospital, New Orleans, LA, “The VentilatorAssisted Population in Louisiana, ” unpublished,Sept. 21, 1985.Walker, D. K., “Care of Chronically I11 Childrenin Schools, ” Peal. Clin. No. Am. 31(1 ):221-233,February 1984.Walker, L. S., Ford, M. B., and Donald, W. D.,“Cystic Fibrosis and Family Stress: Effects ofAge and Severity of Illness, ” Pediatrics 79(2):239-246, February 1987.Wateska, L. P., Sattler, L. L., and Steiger, E.,“Cost of a Home Parenteral Nutrition Program, ”J. A.M.A. 244(20):2303-2304, Nov. 21, 1980.Weissert, W. G., Scanlon, W. J., Wan, T. T. H.,et al., “Care for the Chronically 111: NursingHome Incentive Payment Experiment, ” HealthCare Financing Review 5(2):41-49, Winter 1983.Wilkinson, A., Jenkins, P. A., and Jeffrey, J. A.,“Two Controlled Trials of Dry Artificial Sur-factant: Early Effects and Later Outcome in Ba-bies with Surfactant Deficiency, ” Lancet 2:287-291, 1985.Wood, S. P., Walker, D. K., and Gardner, J.,“School Health Practices for Children With Com-plex Medical Needs,”]. School Health 56(6):215-217, August 1986.Worley, G., Duke University Medical Center,Durham, NC, personal communication, July1986.Wyatt Co., 1984 Group Benefits Survey (Wash-ington, DC: The Wyatt Co., 1984).Zill, N., The School-Aged Handicapped, pre-pared for the National Center for Education Sta-tistics, U.S Department of Education, NCE$85-400 (Washington, DC: U.S. Government Print-ing Office, January 1985).