Impact of Medicare Managed Care Market Withdrawal on ...

The 2001 Survey of Involuntary Disenrolleeswas conducted to investigate the impact ofMedicare+Choice (M+C) plan withdrawalson Medicare beneficiaries. Eighty-four per-cent of a total of 4,732 beneficiaries whoseMedicare managed care (MMC) planstopped serving them at the end of 2000responded to the survey. Their responsesindicated that the withdrawal of plans fromMedicare af fected beneficiaries in terms ofconcerns about getting and paying for care,increased payments for premiums and out-of-pocket costs, and changes in health carearrangements. Of particular concern werethe impacts on those in vulnerable sub-groups such as the disabled, less educated,and minorities.

INTRODUCTION

The Balanced Budget Act (BBA) of 1997expanded the health care options potentiallyavailable to Medicare beneficiaries throughthe establishment of M+C, allowing benefi-ciaries to enroll in a variety of private healthcare options beyond the original Medicarefee-for-service (FFS) program. However,although BBA 1997 increased the healthcare provider options available to seniors,the payment rates to plans were lower thanexpected. These rates, along with newadministrative requirements, may have initi-

ated the withdrawal of many managed careorganizations from the Medicare market(U.S. General Accounting Office, 2001).

The effect of withdrawals by MMCplans, at the individual beneficiary level,can be disruptive, particularly for any ben-eficiaries who have to change providers.Statistics compiled from CMS’ 1998reports indicate that Medicare health careplan withdrawals resulted in 407,000 bene-ficiaries (6.5 percent of M+C enrollees)making a plan change in January 1999. In2000, 327,000 beneficiaries (5 percent ofM+C enrollees) were affected by plans’withdrawals or reduction in service areas.Some plans withdraw completely fromMMC while others continue participationin Medicare but reduce their service areasby no longer serving beneficiaries in somecounties or ZIP codes. In 2001, 934,000Medicare beneficiaries (15 percent of totalenrollment in M+C) were forced to makenew choices about their health plan cover-age when their Medicare health care planwithdrew from the program or reducedtheir service areas. Most recently, 536,000(10 percent of M+C enrollees) were affect-ed in 2002 (Gold and McCoy, 2002).

Few studies have documented the experi-ences these involuntary disenrollees facewhen they are forced to select new coverageupon withdrawal of their plan from the M+CProgram (Kaiser Family Foundation, 1999;U.S. General Accounting Office, 1999; Goldand Justh, 2000). For that reason, CMSdecided to conduct a survey to assess theongoing impact of MMC market withdrawalon beneficiary coverage.

HEALTH CARE FINANCING REVIEW/Fall 2002/Volume 24, Number 1 95

Bridget C. Booske is with the University of Wisconsin atMadison. Judith Lynch is with RTI International. Gerald Riley iswith the Centers for Medicare & Medicaid Services (CMS). Theresearch in this article was funded under HCFA ContractNumber 500-95-0061 (TO#10). The views expressed in this arti-cle are those of the authors and do not necessarily reflect theviews of the University of Wisconsin at Madison, RTIInternational, or CMS.

Impact of Medicare Managed Care Market Withdrawal onBeneficiaries

Bridget C. Booske, Ph.D., Judith Lynch, and Gerald Riley, M.S.P.H.

Most health maintenance organizations(HMOs) that participate in Medicare offeradditional benefits outside the regularMedicare benefit package (Achman andGold, 2002). Extra benefits commonlyinclude low copayments, prescriptiondrugs, unlimited hospitalization, and pre-ventive services. Many beneficiaries havecome to rely on the extra benefits theyreceive from their HMO, particularly pre-scription drug coverage. Replacing thesebenefits through Medigap insurance isusually very expensive, and may be unaf-fordable for some (Gold and Mittler, 2001).Joining another HMO or going to FFS mayalso force many beneficiaries to change doc-tors, creating dissatisfaction and disruptingexisting patterns of care. Therefore, therehas been concern among policymakersabout the impact of the recent HMO with-drawals on the beneficiary population(Barry and Kline, 2002).

There were two previous national effortsspecifically designed to assess the impact ofthe plan withdrawals and service areareductions on beneficiaries.1 The first con-sisted of a survey sponsored by the HenryJ. Kaiser Family Foundation after theJanuary 1999 withdrawals. A report basedon the survey results indicated thatalthough most disenrollees fared relativelywell after their HMO withdrew fromMedicare, many experienced a reduction insupplemental benefits, an increase in pre-miums, and/or disruptions in their carearrangements (Kaiser Family Foundation,1999). Problems were disproportionatelyexperienced by disabled beneficiaries,racial and ethnic minorities, the poor andnear poor, and those reporting fair or poorheath. The second effort consisted of a tele-phone survey of several hundred beneficia-

ries conducted by the Department ofHealth and Human Services’ Office of theInspector General. The survey coveredenrollee notification; information and assis-tance in exploring new insurance options;what option beneficiaries selected; changesin benefits and costs; problems encoun-tered; and satisfaction. The Office of theInspector General survey was conductedtwice, following the January 1999 andJanuary 2000 withdrawals. These surveysdid not find severe problems, but no analy-ses were done for vulnerable populations.

A third study, by Gold and Justh (2000),involved a national sample of over 6,000Medicare beneficiaries of whom 425 were inM+C plans that stopped serving enrolleesat the end of 1999. (This survey was con-ducted as part of the larger monitoringM+C project of Mathematica PolicyResearch, funded by the Robert WoodJohnson Foundation.) A recent publicationfrom this study summarizes the differencesbetween plans that withdraw from M+C andthose that remain (Achman and Gold,2002). Withdrawing plans tend to havelower enrollments, offer less-generous ben-efit packages, had less stable benefits from1999 to 2000, and faced competition prob-lems within their markets.

The purpose of the 2001 Survey ofInvoluntary Disenrollees (Center for HealthSystems Research and Analysis and RTIInternational, 2001) was to understandhow Medicare beneficiaries are affected byMedicare health plan withdrawals andreductions in service areas. This articledescribes the methods and the results ofthat survey. Also, we discuss: • What types of beneficiaries are affected

by plan withdrawals. • What information beneficiaries receive

about plan withdrawals.• The extent to which beneficiaries under-

stand the implications of their plan’s with-drawal and their new coverage options.

96 HEALTH CARE FINANCING REVIEW/Fall 2002/Volume 24, Number 1

1 Another study looked at the impact of HMO withdrawalsspecifically on rural beneficiaries. University of Minnesotaresearchers surveyed 1,093 rural beneficiaries who lost HMOcoverage in January 1999 (Casey, Astrid, and Moscovice, 2002).The survey was conducted from February to May 2000.

• The impact on beneficiaries in terms ofnew coverage, concerns, costs, providerarrangements, and access to care.

METHODOLOGY

Design

The survey was conducted by mail withtelephone followup of non-respondents.Data were collected between March andJune 2001. (A copy of the survey is avail-able on request from the authors.) Wedesigned two versions: the first targeted tosample members who did not have endstage renal disease (ESRD) and the secondtargeted to those who did. ESRD patientsmay be adversely impacted by plan with-drawals because they tend to have highhealth care expenses, and at the time thatthe survey was designed, they were notpermitted to enroll in other MMC plans.Until the Medicare, Medicaid, and StateChildren’s Health Insurance Program(SCHIP) Benefits Improvement andProtection Act of 2000 (BIPA) was enactedin December 2000, ESRD beneficiaries innon-renewing plans could not join a newM+C plan. (The main goal of BIPA was toincrease payments to M+C organizationsto maintain and expand beneficiary accessto M+C plans. However, on implementationof BIPA in March 2001, only seven M+Corganizations re-entered counties fromwhich they had previously withdrawn orexpanded into new counties [U.S. GeneralAccounting Office, 2001]). Since the imple-mentation of BIPA, ESRD beneficiaries areallowed to enroll in another M+C plan iftheir plan terminates its contract withCMS. This provision applies to termina-tions occurring on or after the date ofBIPA’s enactment and retroactively to ter-minations on or after December 31, 1998.

Both questionnaires contained the samequestions, with the ESRD version contain-ing an additional three questions specifi-cally related to dialysis treatment. The top-ics included:• Questions about the sample member’s

former health insurance. • Choosing new health insurance.• Questions about the sample member’s

current health insurance.• Getting needed care since the sample

member left their former plan.• Impact on dialysis treatment (for ESRD

sample members only).• Respondent health status and demo-

graphic characteristics.

Sample Selection and Weighting

For the non-ESRD population, a sam-pling frame was constructed which includ-ed all enrollees, as of October 1, 2000, inplans that terminated or reduced their ser-vice areas effective January 1, 2001. Thereason for using a 3-month window was tocapture people who stayed in the plan untilthe end of the year, as well as those whomay have left earlier, in the event that therewere differences between these types ofenrollees. All beneficiaries who lived out-side the United States, deceased, and insti-tutionalized sample members were exclud-ed from the sampling frame. Once theframe for the non-ESRD population wasconstructed, beneficiaries were assignedto one of two strata—those who lived inareas where another MMC plan was avail-able after December 31, 2000, and thosewho lived in areas in which no otherMedicare plan was available. A separatesample consisting of all Medicare benefi-ciaries who had ESRD who were affectedby plans’ withdrawals and reduction in ser-vice areas was selected. No stratum forthese sample members was defined since,


at the time that the study was designed,ESRD patients would not be able to enrollin another MMC plan even if otherMedicare health plans were available intheir area on January 1, 2001.

The sample sizes for the choice and nochoice strata were based on a goal ofobtaining 3,000 completed interviews withnon-ESRD Medicare beneficiaries, and 385completed interviews with Medicare bene-ficiaries with ESRD. The choice strataincluded 2,772 Medicare beneficiaries wholived in areas in which another Medicarehealth plan was available after December 31,2000. The no choice strata included 1,422Medicare beneficiaries who lived in areasin which no other Medicare health planwas available after December 31, 2000. TheESRD strata included 538 beneficiaries.Data collection activities resulted in anoverall response rate of 83.7 percent. (Theresponse rate was calculated using the fol-lowing formula: Numerator—the numberof completed interviews. Denominator—allsample members in the sample minus thosewho were institutionalized or deceased,and those who reported that they were stillenrolled in the sample plan or left the planbecause they moved out of the plan’s ser-vice area. Questionnaires were consideredcomplete if beneficiaries responded to atleast one of the items concerning theimpact of plan withdrawal. [Additionalinformation about the completeness crite-ria are available on request from theauthors.])

To adjust for the potential of differentialnon-response bias, we used logistic regres-sion to model the functional relationshipbetween a set of predictors and a dichoto-mous response outcome and then usedthat model to construct response propensi-ty weights. The potential set of predictorsavailable for both respondents and non-respondents included age, race, and geo-graphic area. Black persons and other

minority races had less than one-half theodds of a response than did white persons.This was the most significant effect in themodel. Age was also a significant factor.The odds of a response steadily decreasedwith the age of the sample member. Nochange in odds was noted in the under age65 population. Address fields that con-tained rural routes or post office boxes(along with variant abbreviations) had alower odds of response. Address fieldsthat were abnormally short or long had ahigher odds of response. Abnormally longaddresses often suggest that an individualis in the care of someone else which, in thiscase, may have increased the likelihood ofresponse. Responses varied only slightlyamong the U.S. census divisions.

Data Analysis

Descriptive statistics and chi-squaretests of independence were used to assessstatistical associations between a numberof potential outcomes of the plan with-drawals and beneficiary characteristics.These outcomes include new coveragearrangements and the financial, psycholog-ical, and care-related impacts of the planwithdrawals from the Medicare Program.The results presented in this article arebased on weighted data. A weightingmodel was developed incorporating bothsample design and response propensity.Consequently, the data reported for thetotal sample reflects the total population ofinvoluntary disenrollees. (Additional detailsare available within the final report avail-able on request from the authors [Booskeet al., 2002]). Analyses were conductedusing SUDAAN® software that appropri-ately accounts for the sample weightingapproach in calculating standard errors.Findings of significance at the 99 percentprobability level and differences of at least10 percentage points are reported. In


addition, where appropriate, results of mul-tivariate analysis (using logistic regres-sion) are reported to further examine therelationships between beneficiary charac-teristics and the impact of plan with-drawals. Results are reported for signifi-cant logit models with a minimum Cox &Snell R-square of at least 0.10 or where themodel increases the likelihood of predic-tion from the logistic model by at least 10percent (versus a model that simply assignsall responses to the most frequent responsecategory). Additional variables used in thelogistic regression analyses, such as theMMC penetration rate groups and the pay-ment rates that M+C organizations receiveper enrollee per month, were derived fromCMS files, Internet site: http://www.cms.hhs.gov/healthplans/reportfilesdata/

RESULTS

Who are Affected by PlanWithdrawals?

Table 1 shows the characteristics of thethree sample groups of beneficiaries whoresponded to the survey: those who lived ina county with a choice of another MedicareHMO, those in a county without anotherMedicare HMO alternative, and those withESRD. The total column representsweighted data from all three strata. Thus,the total columns in selected tables reflectthe weighted mix of those in counties withand without a Medicare HMO (as ofJanuary 1, 2001) and of those with ESRD,i.e., in proportion to the composition of theentire population of involuntary disen-rollees. Overall, 92 percent of the benefi-ciaries responding to the survey reside in ametropolitan county compared with 76 per-cent of Medicare beneficiaries nationally(Achman and Gold, 2002). However, therewas a significant difference between thegeographic location of the choice and no-

choice strata: only 3 percent of those with achoice of another HMO lived in a non-met-ropolitan area while 34 percent of the ben-eficiaries without another HMO availablelive in non-metropolitan counties.

Compared with CMS data on voluntarydisenrollees (those who leave MedicareHMOs of their own accord), enrollees (thosewho stay in an HMO), and involuntary dis-enrollees (those in plans that leave theMedicare Program) are similar in age, sex,and education, but more likely to reporttheir health to be fair or poor and less like-ly to be Hispanic (Figure 1).

Information about Plan Withdrawals

Sixty-six percent of beneficiaries firstfound out that their plan was going to stopcovering them from the plan itself. Thenext most common source of informationabout the plan withdrawal came from themedia, 18 percent of beneficiaries firstfound out that their plan was leaving theMedicare Program from newspapers,radio, or television.2 Ninety-six percent ofbeneficiaries recalled receiving a letter atsome point from the plan about its impend-ing withdrawal.3

Only about 6 out of 10 beneficiaries indicat-ed they had enough information about theiroptions when their plan stopped coveringthem. Disabled beneficiaries (under age 65)and the oldest-old beneficiaries were less like-ly to indicate that they received enough infor-mation about their coverage options whenthey heard that their plan would stop cover-ing them. Compared with other racial/ethnicgroups, black beneficiaries were less likely toindicate that they had received enough


2 These figures represent additional data from the 2001 Surveyof Involuntary Disenrollees that are not included in any of thetables.3 This rate is significantly higher than the number of involuntarydisenrollees in the case studies who recalled receiving a letterfrom their plan (Grad and Hassol, 2002). They found that about75 percent of beneficiaries recalled receiving a letter from theirplan.

information. Those in poor health were alsoless likely than healthier beneficiaries to indi-cate that they received enough informationabout their coverage options (Table 2).

Knowledge of the availability of supple-mental insurance was somewhat common:68 percent of beneficiaries were aware ofthe availability. Responses to the question

on availability of supplemental options mayreflect beneficiary experiences with healthscreening, i.e., some may have been turneddown, and so the insurance is unavailablealthough it is likely that some of theseresponses reflect misunderstandings onthe part of respondents about what optionsare available to them. Beneficiaries who


Table 1

Survey of Involuntary Disenrollee Sample Strata, by Beneficiary Characteristics: 2001

Total Medicare HMO Available No Medicare HMO Available ESRDCharacteristic (n=3,780) (n=2,215) (n=1,195) (n=370)

PercentAgeUnder 65 Years 7 7 7 1665-74 Years 53 52 55 4675-84 Years 33 33 31 3485 Years or Over 7 7 7 5

SexFemale 57 58 54 44Male 43 42 46 56

Race/EthnicityWhite Non-Hispanic 84 84 84 61Black 9 1 7 25Hispanic 5 4 8 11Other 2 2 1 2

EducationLess than 9th Grade 13 12 18 19Some High School 18 18 17 27High School Graduate 36 36 37 29Beyond High School 33 34 28 25

Self-Reported Health StatusExcellent 6 6 5 2Very Good 22 22 19 6Good 37 37 38 21Fair 28 28 29 41Poor 8 7 1 3

Recent HospitalizationsAt Least 1 in Past 12 Months 21 21 21 69None in Past 12 Months 79 79 79 31

LocationMetropolitan County 92 97 66 95Non-Metropolitan County 8 3 34 5

Dual Eligibility StatusNot Medicaid Eligible 97 97 96 9Medicaid Eligible 3 3 4 1

NOTES: HMO is health maintenance organization. ESRD is end stage renal disease. Percentages are based on weighted data. The weighting incorporatesboth the sample design and response propensity. Consequently, the percentage in the total column cannot be calculated based on the weighted aver-age of the three component columns. Numbers may not sum to 100 percent due to rounding. Metropolitan/non-metropolitan county designation basedon the 1993 Office of Management and Budget definition.

SOURCE: Booske, B.C., University of Wisconsin at Madison, Lynch, J., RTI International, and Riley, G., Centers for Medicare & Medicaid Services, 2001.

were disabled (under age 65), older, in theall other racial group (including Hispanic),and less educated beneficiaries were theleast likely to report availability of supple-mental insurance options (Table 2).

Beneficiaries were asked what theythought would happen to them when theirplan stopped covering them. Those whothought they would be covered by the originalMedicare plan, covered through their currentor former employer, would be able to select anew plan, or would have to purchase supple-mental insurance were considered to under-stand what would happen. (Those whothought they would have to purchase supple-mental insurance may believe that prudenceor financial necessity demanded the purchaseof supplemental insurance rather than it beinga legal or program requirement. Conse-quent-

ly, these individuals were also considered tohave understood what would happen.) Lessthan one-half of the beneficiaries thought thatthe original Medicare plan would cover them.Some beneficiaries thought they would beautomatically enrolled in another HMO whileonly 2 percent thought they would be able toselect a new plan. One in 10 beneficiarieseither indicated that they did not know whatwould happen or did not respond to the ques-tion. Conversely, those who thought that theywould end up with no health insurance or thatthey would be automatically enrolled inanother HMO apparently did not understandthe implications of their plan’s decision to stopcovering them.4 Over one-quarter thoughtthey would end up with no health insurance.


0

Characteristic

Per

cen

t

50 Involuntary Disenrollees with HMO Choice

Involuntary Disenrollees without HMO Choice

Voluntary Disenrollees

Enrollees

Under 65 Years

75 Yearsor Over

Black Hispanic Less thanHigh School

Graduate

Fair orPoor Health

30

15

20

10

5

45

40

35

25

NOTE: HMO is health maintenance organization.

SOURCE: Booske, B.C., University of Wisconsin at Madison, Lynch, J., RTI International, and Riley, G., Centersfor Medicare & Medicaid Services, 2001.

Figure 1

Involuntary and Voluntary Disenrollees and Medicare Beneficiaries that Remain with theirManaged Care Plan: 2001

4 Those who did not respond to this question or whose respons-es could not be coded were classified as not understanding.

Just over one-half of beneficiariesappeared to understand exactly whatwould happen when their plan left theMedicare Program (Table 2). There wasless understanding of what would happenamong beneficiaries who were less educated,in the all other racial group, or Hispanic.However, those who reported that theyhad received enough information aboutthe plan withdrawals were more likely to

indicate they understood what would hap-pen than those who did not indicate thatthey had enough information.

New Coverage Arrangements

Beneficiaries were asked about theirnew coverage arrangements followingtheir plan’s withdrawal from Medicare(Table 3). Respondents could indicate


Table 2

Beneficiary Reports of Adequacy of Information and Availability of Supplemental Insurance andtheir Understanding of Implications of Plan Withdrawal, by Beneficiary Characteristics: 2001

Reporting Having Enough Reporting that Understood What Would Information when their Supplemental Insurance Happen when their Plan

Characteristic Plan Withdrew Option Available Withdrew1

PercentAll Beneficiaries2 63 68 53

Age3

Under 65 Years 46 51 5565-74 Years 64 71 5575-84 Years 65 70 5085 Years or Over 59 61 46

Race/Ethnicity3

White Non-Hispanic 65 72 56Black 48 44 39Hispanic 56 53 34Other 60 65 35

Education3

Less than 9th Grade 56 55 39Some High School 60 62 4High School Graduate 65 71 55Beyond High School 65 75 63

Self-Reported Health Status3

Excellent 63 62 45Very Good 67 73 57Good 66 72 54Fair 59 63 51Poor 49 61 47

End Stage Renal Disease3

No 63 68 53Yes 55 67 47

Information About Plan Withdrawal3Received Enough information — 77 58Did not Receive Enough Information — 54 451 Percentage of beneficiaries who thought they would be covered by the original Medicare plan, covered through their current or former employer,would be able to select a new plan, or would have to purchase supplemental insurance.2 n=3,780.3 Chi-square for each column significant at 0.01 level.

NOTES: Percentages are based on weighted data. The weighting incorporates both the sample design and response propensity.


coverage under more than one arrange-ment so a hierarchical approach was usedto assign them to the types of coverage. Ifrespondents reported enrollment in aMedicare HMO, they were assigned tothis category. For the remaining respon-dents (those who did not report enrollmentin an HMO), if they reported that Medicaid

covered them, they were assigned to thiscategory. This process was repeated foreach category so that the final categoryrepresented all respondents who did notreport that they were enrolled in aMedicare HMO, were covered by Medicaid,were not covered through a current or for-mer employer, and had no supplemental


Table 3

Sample Strata, by Beneficiary Reports of New Coverage Arrangements after Plan Withdrawal: 2001

Medicare HMO No Medicare HMO Total Available Available ESRD

Coverage (n=3,780) (n=2,215) (n=1,195) (n=370)

PercentNew Arrangements1

Enrolled in Medicare HMO 52 56 34 37Covered by Medicaid 3 3 4 11Covered Through Current or Former Employer 8 8 9 7Have Supplemental Insurance 22 19 35 34Covered by Original Medicare Only 15 15 18 11

Satisfaction with New Health Insurance Less Satisfied Now 37 37 4 36About the Same Now 38 39 3 34More Satisfied Now 17 17 19 24Don’t Know or Missing 8 8 11 6

Payments for Monthly Premiums Pay More Now 56 54 63 57Pay Same Amount Now 13 13 10 13Pay Less Now 8 8 6 9Don’t Pay Premiums2 13 14 9 9Don’t Know or Missing 11 11 11 13

Former Plan Paid Cost of MedicinesYes 74 76 65 75No 16 14 23 14Don’t Know or Missing 10 10 12 11

Health Insurance Now Pays Cost of Medicine Yes 53 55 41 42No 38 36 49 49Don’t Know or Missing 10 9 11 10

Paying for Prescription MedicinesPay More Now 51 52 49 55Pay Same Amount Now 25 25 25 18Pay Less Now 10 10 10 17Don’t Know or Missing 14 14 16 101 Respondents could indicate coverage under more than one arrangement so a hierarchical approach was used to assign new coverage arrange-ments. First, if respondents reported enrollment in a Medicare HMO, they were assigned to this category. Next, if applicable, they were assigned toMedicaid. This process was repeated for each category. The final category includes all respondents who did not report enrollment in a MedicareHMO, Medicaid, coverage through an employer, or supplemental insurance.2 Beneficiaries who paid no premiums both before and after plan withdrawal.

NOTES: HMO is health maintenance organization. ESRD is end stage renal disease. Percentages are based on weighted data. The weighting incorporates both the sample design and response propensity. Consequently, the percentage in the total column cannot be calculated based on theweighted average of the three component columns. Numbers may not sum to 100 percent due to rounding.


health insurance. Thus, we designatedthese respondents as covered by originalMedicare only.

Just over one-half of the involuntary dis-enrollees reported enrollment in anotherHMO after their plan withdrew (Table 3).Fifteen percent of the involuntary disen-rollees reported being covered by originalMedicare only. The results of a logisticregression model to predict which benefi-ciaries would end up with only originalMedicare are shown in Table 4. As mightbe expected, a market characteristic (themarket penetration of MMC in an area)was a predictor of whether a beneficiarywould end up with original Medicare only.Beneficiaries in areas of high managedcare penetration were less likely to end upwith only original Medicare only after theirplan withdrew. However, the model alsosuggests that those in vulnerable sub-groups, such as the disabled, and thosewho report their health to be fair or poor,were also more likely than other beneficia-ries to only have original Medicare cover-age. In addition, those who reported thatthey did not have enough informationabout plan withdrawals were also morelikely to end up with Medicare only.However, caution is advised in interpreta-tion of these results since the multivariatemodel using logistic regression to predictwhether a beneficiary reported having tra-ditional Medicare coverage did not have asmuch explanatory power as other modelsderived from this study.

Those living in areas without a MedicareHMO and those with ESRD were far morelikely to report having supplemental insur-ance than beneficiaries in areas with aMedicare HMO (Table 5). Beneficiarieswho indicated that they had enough infor-mation about the plan withdrawals andthose with more education were also morelikely to report that they have supplemen-tal coverage. Beneficiaries who were dis-

abled, under age 65, in the all other racialgroup, Hispanic, and living in CMS region65 were all less likely to report having sup-plemental insurance (Table 4).

These frequencies are based on benefi-ciaries’ reporting of their current healthinsurance coverage—and often do not cor-respond with administrative data. Conse-quently, caution is advised in interpretingthese numbers, e.g., the 34 percent of ben-eficiaries who live in an area without anoth-er Medicare HMO, but reported belongingto an HMO. In general, there was over 90percent agreement between CMS enroll-ment records and beneficiaries’ reports ofnot being in an HMO. However, when ben-eficiaries reported that they do belong toan HMO, CMS records only confirmedHMO membership in about 50 percent ofthese cases.

Beneficiaries in counties without anotherMedicare HMO available who reportedenrollment in an HMO may have been cor-rect at the time of their survey response.The definition of whether a county offered achoice reflects whether there was a choiceof another M+C HMO available for enroll-ment as of January 1, 2001. Subsequently, afew plans did expand their service intosome counties (e.g., Texas, New York, andNew Mexico as of March 1, 2001). Also,some Medicare HMOs imposed capacitylimits. There are two types of capacity lim-its: (1) plans can request capacity limits beestablished prospectively to be appliedwhen their enrollment reaches a certainlevel, or (2) plans can request that theirenrollment level be limited to the number ofbeneficiaries currently enrolled. Conse-quently, at any point in time a plan withcapacity limits may or may not actually beaccepting new enrollees. Other HMOs onlycover parts of counties (particularly groupand staff models).


5 CMS Region 6 consists of the following States: Arizona,Lousiana, New Mexico, Oklahoma, and Texas.


Table 4

Results of Logistic Regressions of New Coverage for All Involuntary Disenrollees: 2001

Having Original Medicare Only Having Supplemental Insurance After Plan Withdrawal1 After Plan Withdrawal2

95 Percent 95 Percent Independent Variable Odds Ratio Confidence Interval Odds Ratio Confidence Interval

Intercept 0.15 0.10-0.22 3.33 2.42-4.59

AgeUnder 65 Years (Disabled) *1.89 1.29-2.77 *0.41 0.27-0.6065 Years or Over 1.00 — 1.00 —

SexFemale 0.81 0.63-1.02 1.07 0.88-1.31Male 1.00 — 1.00 —Race/EthnicityAll Other Racial Groups and Hispanic 1.45 0.07-1.98 *0.54 0.41-0.72White Non-Hispanic 1.00 — 1.00 —

EducationLess than 9th Grade 1.36 0.98-1.89 *0.67 0.50-0.919th Grade or More 1.00 — 1.00 —

Self-Reported HealthFair or Poor *1.58 1.22-2.05 0.83 0.67-1.03Good or Excellent 1.00 — 1.00 —

Hospitalized in Past 12 MonthsYes *0.65 0.48-0.89 1.15 0.89-1.47No 1.00 — 1.00 —

CMS RegionRegion 63 *1.82 1.38-2.41 *0.58 0.45-0.74Other Regions 1.00 — 1.00 —

CountyNon-Metropolitan 1.42 0.96-2.11 0.80 0.57-1.14Metropolitan 1.00 — 1.00 —

Reported HMO EnrollmentYes NA NA *0.70 0.57-0.85No — — 1.00 —

Reported Medicaid EnrollmentYes NA NA 1.01 0.67-1.51No — — 1.00 —

Medicare Managed Care Market PenetrationHigh (35-45 Percent) *0.44 0.22-0.86 *0.38 0.24-0.58Moderate (15-34 Percent) *0.46 0.32-0.67 *0.62 0.47-0.82Limited (6-14 Percent) 0.98 0.68-1.40 0.74 0.54-1.01Minimal (1-5 Percent) 0.78 0.55-1.09 *1.39 1.02-1.89None (No HMOs) 1.00 — 1.00 —

Medicare Monthly Payment Rate to Medicare+Choice Organization

Less than $525 *0.47 0.32-0.69 *2.90 2.08-4.04$525 0.91 0.66-1.25 *1.52 1.19-1.93More than $525 1.00 — 1.00 —

Information about Plan WithdrawalNot Enough Information *1.61 1.26-2.06 0.84 0.69-1.03Enough Information 1.00 — 1.00 —

*p<0.05.1 Model is significant. Cox & Snell R-square for dependent variable having Medicare only = 0.07.2 Model is significant. Cox & Snell R-square for dependent variable having supplemental insurance = 0.12.3 Centers for Medicare & Medicaid Services (CMS) Region 6 consists of the following States: Arizona, Louisiana, New Mexico, Oklahoma, and Texas.

NOTES: Overall sample size for these logistic regressions was 3,780. HMO is health maintenance organization. NA is not available. Metropolitan/non-metropolitan county designation based on the 1993 Office of Management and Budget definition. Medicare managed care market penetration groupsbased on characterization of Medicare markets by the Center for Studying Health System Change.


Also, the definition of choice does notaddress the availability of cost contract orprivate FFS plans or of HMO coverage thatis offered to beneficiaries by a current orformer employer or via participation in

Medicaid. In addition, some beneficiariesmay not understand that if they continue tosee a provider that was affiliated with theirformer HMO, their services may now becovered under the original Medicare. Of


Table 5

Results of Logistic Regressions of Enrollment in Another HMO for Beneficiaries Living inCounties with Choice of Another Medicare HMO: 2001

Enrolled in Medicare+Choice Plan Report Enrolling in Another after Plan Withdrawal, per HMO after Plan Withdrawal1 CMS Administrative Records2


Intercept 0.63 0.47-0.84 0.17 0.12-0.26AgeUnder 65 Years (Disabled) 0.92 0.61-1.40 1.50 0.95-2.3965 Years or Over 1.00 — 1.00 —SexFemale 1.16 0.94-1.42 *1.29 1.02-1.62Male 1.00 — 1.00 —Race/EthnicityAll Other Racial Groups and Hispanic 1.03 0.76-1.39 0.94 0.68-1.30White Non-Hispanic 1.00 — 1.00 —EducationLess than 9th Grade 1.04 0.75-1.45 0.97 0.68-1.399th Grade or More 1.00 — 1.00 —Self-Reported HealthPoor or Fair 0.85 0.67-1.07 0.82 0.64-1.05Good or Excellent 1.00 — 1.00 —Hospitalized in Past 12 MonthsYes 1.07 0.82-1.38 0.95 0.72-1.25No 1.00 — 1.00 —CMS RegionRegion 63 *0.72 0.55-0.94 0.88 0.65-1.17Other Regions 1.00 — 1.00 —CountyNon-Metropolitan 1.59 0.83-3.07 0.69 0.23-2.03Metropolitan 1.00 — 1.00 —Medicare Managed Care Market PenetrationHigh (35-45 Percent) *6.58 4.15-10.43 *22.89 13.59-38.54Moderate (15-34 Percent) *3.91 2.99-5.12 *9.21 6.35-13.37Limited (6-14 Percent) *1.74 1.27-2.38 *3.25 2.12-4.99Minimal (1-5 Percent) 1.00 — 1.00 —Monthly Medicare Payment Rate to

Medicare+Choice OrganizationLess than $525 0.93 0.53-1.64 *0.35 0.14-0.83$525 *0.70 0.54-0.90 0.87 0.66-1.14More than $525 1.00 — 1.00 —Information about Plan WithdrawalNot Enough Information 0.81 0.65-1.01 *0.67 0.52-0.85Enough Information 1.00 — 1.00 —

*p<0.05.1 Model is significant. Cox & Snell R-square for dependent variable having Medicare only = 0.10.2 Model is significant. Cox & Snell R-square for dependent variable HMO enrollment according to administrative records = 0.203 Centers for Medicare & Medicaid Services Region 6 consists of the following States: Arizona, Louisiana, New Mexico, Oklahoma, and Texas.

NOTES: Initial sample size for these logistic regressions was 2,215. HMO is health maintenance organization. Metropolitan/non-metropolitan countydesignation based on the 1993 Office of Management and Budget definition. Medicare managed care market penetration groups based on character-ization of Medicare markets by the Center for Studying Health System Change.


the beneficiaries who live in an area with-out another Medicare HMO, CMS admin-istrative files showed no record of M+Cenrollment for 80 percent of the beneficia-ries who reported HMO membership. Ofthe remaining 80 beneficiaries, administra-tive records showed that as of January 1,2001, 27 were enrolled in cost plans and 24enrolled in a private FFS which were notincluded in the definition of the choice/no-choice counties.

While over one-half of all beneficiariesreported membership in an HMO, this pro-portion was not constant across the threesample strata (Table 3). Beneficiaries incounties without a Medicare HMO andbeneficiaries with ESRD were far less like-ly to report membership in an HMO. Infact, since only 37 percent of ESRD benefi-ciaries reported membership in an HMO(similar to the level for those in countieswithout Medicare HMOs), this suggeststhat few ESRD beneficiaries have benefitedso far from the legislative changes inBIPA) of 2000. (CMS records actually onlyconfirmed reports of HMO enrollment for11 percent of the 38 percent of ESRD ben-eficiaries who reported enrollment in aM+C plan.) By far the most significant pre-dictor of whether a beneficiary reportedenrollment in another HMO in multivariatemodeling was the market penetration ofMedicare managed care in their county(Table 6). None of the beneficiary demo-graphic characteristics were associatedwith a significant increase or decrease inthe likelihood of reporting enrollment inanother HMO. The odds of CMS Region 6beneficiaries reporting enrollment in anotherHMO were 28 percent lower than those forbeneficiaries living elsewhere in theUnited States. A logistic regression modelof administrative records of HMO enroll-ment produced similar results with respectto market penetration; however, Region 6was not significant in this model (Table 5).

Beneficiaries who indicated that they hadreceived enough information about theplan withdrawals were more likely to enrollin an HMO, according to administrativerecords, than those who said they did notget enough information.

Impact of Plan Withdrawals onBeneficiaries

This section of results summarizes theimpact of plan withdrawals on beneficia-ries’ care and provider arrangements.When asked about their level of satisfac-tion with their new coverage following thewithdrawal of their former plan fromMedicare, 37 percent of beneficiaries indi-cated that they were less satisfied withtheir insurance coverage now (Table 3).However, this may be due in part to lack ofexperience with their new coverage.

Another set of questions addressed theconcerns that beneficiaries faced when theyfound out that their plan was withdrawingfrom Medicare. Fifty-one percent of the ben-eficiaries reported being very concernedabout getting care that they needed.Respondents in counties with or withoutanother Medicare HMO appeared not to dif-fer with respect to concerns regarding abili-ty to get care. However, there were signifi-cant differences between various subgroupswith the more vulnerable expressing moreconcerns about their ability to get care aftertheir plan withdrew: the beneficiaries whowere disabled, less educated in the all otherracial group or Hispanic, and reported fair orpoor health were all more likely to reportmore concerns than other beneficiaries(Table 6). In particular, when all other con-ditions are held constant, the odds of thosewho are disabled being very concernedabout getting care after their plan withdrewfrom Medicare are over 100 percent higherthan those for aged beneficiaries. Informationclearly reduced the likelihood of concerns:


beneficiaries who did not have adequateinformation about the plan withdrawals werefar more likely to have concerns about get-ting care than those with enough informa-tion (Table 6). Beneficiaries who live in non-metropolitan areas or areas with low MMCpenetration were also more likely to be con-cerned.

Fifty-six percent of the beneficiaries whoresponded to the survey reported that theyhad to pay more in premiums after theirplan withdrew (Table 3). However, manyHMOs that have remained in Medicarehave increased premiums and reducedbenefits in the last few years. In their con-tinuing analysis of trends in benefits and


Table 6

Results of Logistic Regressions of Concerns about Getting Care and Having to Pay More forPremiums after Plan Withdrawal: 2001

Being Very Concerned about Getting Having to Pay More for Premiums Needed Health Care after Plan Withdrawal1 after Plan Withdrawal2


Intercept 0.75 0.52-1.09 11.73 7.32-18.78

AgeUnder 65 Years (Disabled) *2.30 1.53-3.47 *0.52 0.31-0.8865 Years or Over 1.00 — 1.00 —SexFemale *1.37 1.14-1.64 1.19 0.93-1.52Male 1.00 — 1.00 —Race/EthnicityAll Other Racial Groups and Hispanic *1.66 1.26-2.18 0.77 0.52-1.13White Non-Hispanic 1.00 — 1.00 —EducationLess than 9th Grade *1.55 1.17-2.04 1.27 0.84-1.909th Grade or More 1.00 — 1.00 —Self-Reported HealthPoor or Fair *1.56 1.29-1.89 1.07 0.81-1.41Good or Excellent 1.00 — 1.00 —Hospitalized in Past 12 MonthsYes *1.38 1.10-1.72 1.30 0.95-1.76No 1.00 — 1.00 —CMS RegionRegion 63 1.05 0.83-1.33 2.33 1.63-3.33Other Regions 1.00 — 1.00 —CountyNon-Metropolitan *1.42 1.01-2.00 *0.79 0.47-1.33Metropolitan 1.00 — 1.00 —New Coverage ArrangementMedicare HMO *0.72 0.53-0.96 1.00 —Medicaid 0.75 0.44-1.28 NA 0.70-1.54Employer-Provided *0.31 0.21-0.47 1.04 3.72-7.37 Supplemental *0.43 0.31-0.59 *5.24 —Original Medicare only 1.00 — NA —Medicare Managed Care Market PenetrationHigh (35-45 Percent) 0.69 0.46-1.05 *0.44 0.26-0.73Moderate (15-34 Percent) *0.74 0.57-0.96 *0.55 0.39-0.77Limited (6-14 Percent) 1.05 0.78-1.41 *0.64 0.43-0.96Minimal (1-5 Percent) 0.98 0.74-1.30 1.30 0.85-1.99None (No HMO) 1.00 — 1.00 —

See footnotes at end of table.

premiums funded by the CommonwealthFund, Achman and Gold (2002) noted thatdespite the congressional action toincrease the payment rates that M+C orga-nizations receive, mean premium and cost-sharing levels in M+C plans continued toincrease in 2001. For example, averagemonthly premiums went from $14.43 in2000 to $22.94 in 2001. (These averagemonthly premiums are for all M+C plansincluding those with zero premiums. In2001, 46 percent of M+C plans offered zeropremium packages.)

The odds of the disabled having to paymore for premiums were lower than thoseof aged beneficiaries. Fewer disabled ben-eficiaries may have experienced increasesin premiums because many are dually eli-gible for Medicare and Medicaid and payno premium for Medicaid. However, thelogistic regression excluded those who aredually eligible and those who report hav-ing Medicare only and controlled for other characteristics (Table 6). The odds ofbeneficiaries with supplemental insurance

reporting that they had to pay more forpremiums after their plan withdrew werefar greater when compared with beneficia-ries reporting enrollment in anotherMedicare HMO (Table 6). Compared withbeneficiaries in other regions, the odds ofbeneficiaries from CMS Region 6 having topay more for premiums were also higher.

Seventy-four percent of beneficiariesreported that their former plan paid all orsome of the cost of their prescription med-icines whereas, after the withdrawing plansstopped covering them, this percentagedropped to 53 percent (Table 3). This com-pares with a decrease in the proportion ofall M+C enrollees with prescription drugcoverage that went from 78 percent in 2000to 70 percent in 2001 (Achman and Gold,2002).

The loss of prescription drug coveragemeant that 51 percent of involuntary disen-rollees found themselves paying more forprescription medicines after their formerplan withdrew from Medicare (Table 3).However, about 1 in 10 beneficiaries did


Table 6—Continued

Results of Logistic Regressions of Concerns about Getting Care and Having to Pay More forPremiums after Plan Withdrawal: 2001

Being Very Concerned about Getting Having to Pay More for Premiums Needed Health Care after Plan Withdrawal1 after Plan Withdrawal2


Monthly Medicare Payment Rate to Medicare+Choice Organization

Less than $525 *0.67 0.47-0.95 1.36 0.84-2.18$525 0.89 0.72-1.12 0.88 0.65-1.19Greater than $525 1.00 — 1.00 —

Information about Plan WithdrawalNot Enough Information *2.55 2.11-3.07 NA —Enough Information 1.00 — NA —

*p<0.05.1 Initial sample size for this logistic regression was 3,780. Model is significant. Cox & Snell R-square for dependent variable having Medicare only = 0.40.2 Initial sample size for this logistic regression was 2,604 (excludes those with Medicare only, those with Medicaid coverage, and those who paid nopremiums before and after plan withdrawal and those who did not know whether they paid higher or lower premiums). Model is significant. Cox &Snell R-square for dependent variable having to pay = 0.133 Centers for Medicare & Medicaid Services Region 6 consists of the following States: Arizona, Louisiana, New Mexico, Oklahoma, and Texas.

NOTES: Metropolitan/non-metropolitan county designation based on the 1993 Office of Management and Budget definition. Medicare managed caremarket penetration groups based on characterization of Medicare markets by the Center for Studying Health System Change. HMO is health mainte-nance organization. NA is not available.


not know or did not answer questionsabout their prescription drug coveragebefore or after the plan withdrawals.

About one in five beneficiaries in areaswith another Medicare HMO indicatedthat they had to change their personal doc-tor or nurse after their former plan stoppedcovering them compared with 12 percentin areas without another HMO. Twenty-four percent of the beneficiaries in areaswith another HMO who had to changetheir personal provider indicated that itwas a big problem to find a new provider.Beneficiaries in areas with a choice ofanother Medicare HMO reported a higherincidence of changing providers comparedwith those in areas without a choice ofanother HMO (Table 7).

Four out of 10 beneficiaries reportedthat they had been seeing a specialist whentheir plan stopped covering them (Table7). One in four of these beneficiaries (9percent) reported that they had to stopseeing their specialist.

Twelve percent of the involuntary disen-rollees indicated that they had trouble get-ting care they wanted or needed since theirplan stopped covering them (Figure 2). Thiscompares with 4 percent of all Medicare

beneficiaries in HMOs (Centers forMedicare & Medicaid Services, 2002).Twenty-four percent, however, indicatedthat they had delayed seeking medical carebecause they were worried about the cost,compared with 3 percent of MedicareHMO beneficiaries. Fifteen percent report-ed that there were medicines prescribedfor them that they did not get. Cost and thelack of insurance coverage were the mostcommon reasons for not getting pre-scribed medicines.

There were clear and significant differ-ences in problems with access to care(Figure 2) between the beneficiaries whowere disabled, age 65 or over, white non-Hispanic and those in the all other racialgroup or Hispanic, and between those whoreported good or excellent health andthose who reported themselves to be infair or poor health.

DISCUSSION AND IMPLICATIONS

The characteristics of involuntary disen-rollees living in areas with and without achoice of another HMO were quite similarwith one major exception: only 3 percent ofbeneficiaries affected by the 2001 plan


Table 7

Beneficiary Reports of Impact on Provider Arrangements: 2001

Medicare HMO Available No Medicare HMO AvailableReport (n=2,215) (n=1,195)

PercentHad to Change Personal Doctor or Nurse 22 12

(n=478) (n=144)A Big Problem to Get a Personal Doctor or Nurse 20 13A Small Problem to Get a Personal Doctor or Nurse 26 17Not a Problem to Get a Personal Doctor or Nurse 45 58Don’t Know or Missing 5 1Have Not Found a New Doctor Yet 6 11

Seeing a Specialist in Former Plan 41 38(n=895) (n=448)

Had to Stop Seeing Specialist 24 16Did Not Have to Stop Seeing Specialist 66 73Don’t Know or Missing 8 10Did Not Need to See a Specialist 2 2

NOTES: HMO is health maintenance organization. Percentages are based on weighted data. The weighting incorporates both the sample design andresponse propensity. For example, 22 percent of beneficiaries in areas with another Medicare HMO available had to change providers after planwithdrawal. Twenty percent of this 22 percent, or 20 percent of 478, reported a big problem getting a different personal provider.


withdrawals with a choice of another HMOlived in non-metropolitan counties, while34 percent of beneficiaries without anotherHMO option lived outside of metropolitanareas. This difference highlights the con-tinued disparity noted by Achman andGold (2002) in choices available to metro-politan and non-metropolitan beneficiaries.In contrast to the similarities between ben-eficiaries in the other two sample strata,beneficiaries in the ESRD sample werequite different from those without ESRD:beneficiaries with ESRD were more likelyto be female, black, in fair or poor health,and had to have been hospitalized duringthe past year.

The findings show that there are someclear information and understanding gapsamong beneficiaries, particularly those inmore vulnerable subgroups, regarding theoptions available to them and the implica-tions of plans withdrawing from the Medi-care Program. Letters from the non-renew-ing plans were by far the most frequentfirst source of information about the planwithdrawal for the majority of disenrollees.The media (television, radio, or newspa-per) was the next most frequent source ofinformation except for those in the morevulnerable subgroups such as the less edu-cated and all racial groups other than whitenon-Hispanic. Disabled beneficiaries who


0

Percent

Characteristic

60Needed care, but had trouble getting it.

Needed care, but delayed getting it due to cost.

Needed prescription drugs, but did not get them.

Under65Years

65YearsorOver

Lessthan9thGrade

9thGradeorHigher

WhiteNon-Hispanic

All Other

HospitalizedinPast12Months

NotHospitalizedinPast12Months

FairorPoor

GoodorExcellentHealth

Total

20

50

40

30

10

NOTES: Percentages are based on weighted data. The weighting incorporates both the sample design and responsepropensity. Consequently, the percentage in the total bars cannot be calculated based on the average of the pair ofbars for each characteristic.

SOURCE: Booske, B.C., University of Wisconsin at Madison, Lynch, J., RTI International, and Riley, G., Centers forMedicare & Medicaid Services, 2001.

Figure 2

Beneficiary Reports of Impact on Access to Care, by Beneficiary Characteristics: 2001

were in the all other racial group and in fairto poor health were less likely to indicatethey received enough information aboutthe plan withdrawals. These same groupswere less likely to be aware of the avail-ability of supplemental health insurance.Other groups who were less aware of sup-plemental insurance options includedthose beneficiaries less-educated whowere in the all other racial group (i.e.,other than white) and less-educated benefi-ciaries were also less likely than other ben-eficiaries to understand what would hap-pen to them with respect to health carecoverage when their plan left the MedicareProgram. About 4 in 10 beneficiaries in theall other racial group or beneficiaries withless than a high school education under-stood what would happen compared toone-half of all beneficiaries. Beneficiaries’reports of having enough information wereclearly associated with their understandingof what would happen when their planwithdrew and awareness of supplementalhealth insurance options.

Compared with results from the KaiserFamily Foundation (1999) study of benefi-ciaries affected by plan withdrawals in January1999, those affected by the January 2001withdrawals were less likely to reportenrollment in another HMO (52 percent in2001 compared with 77 percent in 1999)even though similar proportions of benefi-ciaries (four out of five) still had an HMOoption available to them. As was true in1999, reports of enrollment in anotherHMO were strongly related to the numberof Medicare plans and their market pene-tration in an area. Furthermore, it shouldbe noted that when compared with CMSenrollment records, beneficiaries apparent-ly overestimate membership in HMOs.Consequently, the true percentage thatswitches to another M+C is less than thatderived from survey responses. Whilethere are legitimate reasons why many

beneficiaries believe that they are enrolledin a Medicare HMO even though this is notconfirmed by CMS data, the high level ofdiscordance with CMS enrollment recordsmay cause some to wonder about the valid-ity of other responses that beneficiariesgave. Although such discrepancies high-light the lack of understanding that manybeneficiaries have about their coverage,beneficiaries’ reports of coverage arrange-ments do reflect their perceptions of whatcoverage they do or do not have. SinceCMS records do not include informationon supplemental or employer coverage, ask-ing beneficiaries about the coverage doesprovide some incremental informationabout levels of these types of coverage.

After MMC’s role in a local market, hav-ing enough information was the next mostlikely predictor of beneficiaries choosing toenroll in another HMO. As opposed toreverting to original Medicare coverage,joining another HMO requires a consciousaction on behalf of the beneficiary. Thosewho felt they did not have enough informa-tion may not have known that there wasanother HMO that they could join.

The other significant predictor of benefi-ciaries reporting HMO enrollment was livingin a region other than CMS Region 6. Evenafter accounting for the lower availability ofplans to beneficiaries in Region 6 (the aver-age number of plans for a Region 6 benefi-ciary was one versus two for beneficiaries inother parts of the country), beneficiaries inRegion 6 were less likely to enroll in anotherHMO and they were also less likely to havesupplemental insurance coverage. A moredetailed examination in particular marketswithin Region 6 might shed some light onthe particular environment that beneficiariesin this region face, e.g., case studies by Gradand Hassol (2002) of involuntary disenrolleesidentified a unique situation in Houston,Texas in 2000 when the only remainingHMO in the area reached its capacity limit in


September of that year and no longer accept-ed any new enrollees. Stuber et al. (2002)also examined the Houston market in theircase studies of seven markets from whichM+C plans withdrew. They noted the disrup-tion of provider networks that often precedesa plan’s decision to withdraw from a M+Cmarket: in Houston, 27 percent of primarycare physicians left a plan’s network after oneyear. Eighty-six percent of Region 6’s benefi-ciaries and 22 percent of all beneficiaries whoresponded to the 2001 survey were fromTexas.

Reports of supplemental health insurancewere strongly related to low MMC penetra-tion, i.e., beneficiaries with fewer managedcare options did turn to supplemental insur-ance. However, beneficiaries in vulnerablesubgroups, such as the disabled, less edu-cated, and those in the all other racialgroups or Hispanic beneficiaries, were lesslikely to report having supplemental insur-ance than other beneficiaries. Sixty-six per-cent of those with original Medicare cover-age only, i.e., without supplemental insur-ance, reported that they did not have sup-plemental insurance because it cost toomuch. This percentage was even higher (79percent) for those in areas without a choiceof another HMO. However, ESRD beneficia-ries without supplemental insurance werealmost as likely to cite not applying for it orthinking that they would be turned down asthey were to cite its cost as the main reasonfor not having supplemental insurance. Inaddition to the financial barriers to acquir-ing supplemental coverage, lack of informa-tion appears to be a barrier to having sup-plemental insurance. Those who indicatedthat they had enough information about theplan withdrawals were more likely to havesupplemental coverage than those who didnot have enough information.

In examining the impact of plan with-drawals on beneficiaries, we looked at ben-eficiaries’ concerns when hearing about

the plan withdrawals, the impact on benefi-ciaries’ costs, provider arrangements, andaccess to care. Approximately three out ofevery four beneficiaries reported that theywere somewhat or very concerned aboutbeing able to pay for health care when theirplan withdrew from the MedicareProgram. A similar, but not entirely over-lapping, proportion of beneficiaries werealso concerned about getting care whileconcerns about having to change providerswere not quite as widespread, but still con-siderable. Again, the more vulnerable sub-groups, including the disabled, less-educat-ed, people in the all other racial group(other than white non-Hispanic), and in fairor poor health, were disproportionatelyaffected. Beneficiaries in non-metropolitanareas with low MMC penetration were alsomore concerned by the plan withdrawals.Having enough information reduced, butdid not eliminate concerns about theimpact of plan withdrawals.

Plan withdrawals affected just over one-half of beneficiaries with higher premiums.However, as previously mentioned, theaverage monthly premiums increased forall M+C enrollees along with increases incopayments and reductions in benefits. Forinvoluntary disenrollees, higher premiumswere more likely to come from acquiringsupplemental insurance than from higherpremiums due to enrollment in anotherHMO. Consequently, those in the morevulnerable subgroups were less likely toreport paying more since they were lesslikely to have supplemental coverage. It istherefore likely that in the tradeoffbetween higher premiums versus lowerbenefits, the vulnerable were forced by cir-cumstance to go with lower benefits. In aneffort to keep our survey to a manageablelength for respondents and due to theknown unreliability of reports of specificbenefit details, the only specific benefitabout which beneficiaries reported was


prescription drug coverage. About one infive beneficiaries lost prescription drug cov-erage as a result of their plan withdrawal.The disabled and those in fair or poorhealth were again impacted more than oth-ers by having to pay more for prescriptionmedicines. Those in areas without a choiceof another HMO were less likely to reporthaving prescription drug coverage thanthose in areas with at least one HMO option.

Disruptions in provider arrangementswere less widespread than some of theother outcomes of the plan withdrawals:only one in five beneficiaries indicated thatthey had had to change their personal doc-tor or nurse after their plan withdrew fromMedicare. This was similar to the ratefound in the Kaiser Family Foundation(1999) study. One out of 10 total beneficia-ries had to stop seeing a specialist (some-what lower than the rate found in theKaiser study). Although the rate of disrup-tion was not as high as might have beenpredicted, there were significant differ-ences within the involuntary disenrolleepopulation. Those in areas with a choice ofanother HMO were more likely to reporthaving to change their personal doctor ornurse than those without another HMOoption. Among those who were seeing aspecialist when their plan withdrew, thedisabled and those in fair or poor healthwere more likely to have to stop seeingtheir specialist. As would be expected, thepotential for disruption in providerarrangements was a tradeoff that benefi-ciaries had to deal with in exchange forHMO coverage and the potential for morecomprehensive benefits. Those who didnot enroll in a different HMO were lesslikely to have to change providers.

In terms of access to care, one in 10 ben-eficiaries who needed care after their planwithdrew indicated that they had troublegetting care. However, one in five benefi-ciaries delayed seeking care because of

being worried about the cost and 15 per-cent did not get medications that had beenprescribed for them. This rate was threetimes higher than the rate found in theKaiser study reflecting, perhaps, theincreasing concerns about the costs of pre-scription drugs. Again, it was the benefi-ciaries who were disabled, in the all otherracial group or Hispanic, and in fair or poorhealth whose access to care was mostaffected.

Clearly, the withdrawal of plans from theMedicare Program affected large numbersof beneficiaries causing at least short-termanxiety, higher costs, and/or disruptionsin health care. However, what is of contin-uing concern is the effect of these with-drawals on the most vulnerable. WhileHMOs that have managed to stay in theMedicare market may continue to offerbeneficiaries an attractive alternative tooriginal Medicare, if and when the plansleave Medicare, thousands of disabled,minority, less educated, and sick beneficia-ries are left uncertain about their options,concerned about costs (particularly of pre-scription drugs), and less likely to get neededcare. One strategy that appears to amelio-rate some, but not all of these problems isthe provision of information about whatwill happen when a plan leaves and what, ifany, options are available. However, thebeneficiaries most likely to report they didnot have enough information were thosewith the least education, the very old andthe disabled, i.e., the beneficiaries who areprobably least able to process additionalinformation. Consequently, the problemmay not have been a lack of information,but beneficiaries may have found it to betoo much or too confusing. Providing moreinformation in the future to them withoutsimplifying it may actually be counterpro-ductive. Furthermore, as an anonymousreviewer pointed out, it seems that, no mat-ter how hard agencies such as CMS try to


provide simplified material, there is stillgoing to be a population that does notunderstand their choices or the implica-tions of these choices while others are sim-ply not motivated enough to try to under-stand. Consequently, rather than simplyproviding more information, descriptionsof the specific implications of plan with-drawals on beneficiaries and their optionsfor coverage should be tailored to meet thespecific needs and interests of these vul-nerable subgroups.

REFERENCES

Achman, L., and Gold, M.: Medicare+Choice 1999-2001: An Analysis of Managed Care PlanWithdrawals and Trends in Benefits and Premiums.The Commonwealth Fund, Publication Number497. New York, New York. February 2002. Internetaddress: http://www.cmwf.orgBarry, C.L., and Kline, J.: Medicare Managed Care:Medicare+Choice at Five Years. Issue Brief Number537. The Commonwealth Fund. New York, New York.April 2002. Internet address: http://www.cmwf.orgBooske, B.C., Lynch, J., Kenyon, A., and Scheffler,S.: Medicare Beneficiaries Who InvoluntarilyDisenroll from Their Health Plans. Final Report tothe Centers for Medicare & Medicaid Services,Contract Number 500-95-0061, (TO#10). March2002. Casey, M., Astrid, K., and Moscovice, I.: MedicareMinus Choice: The Impact of HMO Withdrawals onRural Medicine Beneficiaries. Health Af fairs21(3):192-199, May/June 2002. Center for Health Systems & Analysis and BTIInternational: 2001 Survey of Involuntary Disenrollees.Madison, WI. Winter/Spring 2001.Centers for Medicare & Medicare Services:Unpublished data from the 1998 Medicare CurrentBeneficiary Survey. Baltimore, MD. 2002.Gold, M., and Justh, N.: Forced Exit: Beneficiariesin Plans Terminating in 2000. MonitoringMedicare+Choice, Fast Facts Number 3, September2000. Internet address: http://www.mathemati-ca.mpr.com

Gold, M., and McCoy, J.: Choice Continues toErode in 2002. Monitoring Medicare+Choice, FastFacts Number 7, January 2002. Internet address:http://www.mathematica-mpr.comGold, M., and Mittler, J.: The Structure ofSupplemental Insurance for Medicare Beneficiaries.Monitoring Medicare+Choice, Operational InsightsFast Facts Number 3, June 2001.Internet address:http://www.mathematica-mpr.comGrad, O., and Hassol, A.: Involuntary Disenrollmentfrom Medicare+Choice Medicare Managed CarePlans: Experiences of Beneficiaries in SixCommunities. Final Report to Centers for Medicare& Medicaid Services. Abt Associates Inc.Cambridge, MA. September 2001.Kaiser Family Foundation: How Medicare HMOWithdrawals Affect Beneficiary Benefits, Costs andContinuity of Care. The Henry J. Kaiser Foundation,Menlo Park, CA. November 1999.Stuber, J., Dallek, G., Edwards, C., et al.: Instabilityand Inequity in Medicare+Choice: The Impact onMedicare Beneficiaries. Executive SummaryNumber 496. The Commonwealth Fund. New York,New York. January 2002. Internet address: http://www.cmwf.orgU.S. General Accounting Office: Medicare+Choice:Recent Payment Increases Had Little Ef fect onBenefits or Plan Availability in 2001. GAO-02-202.U.S. General Accounting Office. Washington, DC.,2001.U.S. General Accounting Office: Medicare ManagedCare Plans: Many Factors Contribute to RecentWithdrawals; Plan Interest Continues. GAO/HEHS-99-91. U.S. General Accounting Office. Washington,DC. April 1999.

Reprint Requests: Bridget Booske, Ph.D., University ofWisconsin-Madison, 610 Walnut Street, Number 1167, Madison,WI 53726. E-mail: [email protected]


Impact of Medicare Managed Care Market Withdrawal on ...

Documents

Transcript of Impact of Medicare Managed Care Market Withdrawal on ...