48

C H A P T E R 5

T H E I L L I N O I S R E P O R T 2 0 1 3

C H A P T E R 5

Institute of Government & Public Affairs • igpa.uillinois.edu 49

• Seventeen states are implementing differentialresponse in full, and another 13 had a programthat included at least some components. Illinoisjust ended a five-year pilot program. The fullevaluation is expected in 2013.

• The core features of differential response can ap-peal to a bipartisan agenda: it shifts responsibil-ity from the government to neighborhoods andcommunities, and also offers a safety net to fam-ilies struggling with structural barriers, likepoverty, rather than criminalizing poor families.

• A look at all differential response tracks in thecountry reveals that children are not at increased

risk when placed in a differential response track.For most programs, rates of recurrence were bet-ter—or no different from—the investigatorytrack. And, families in the differential responsetrack are more satisfied with the interventionthey received than families in the investigationstrack.

• Going forward, Illinois lawmakers will need toconsider: ways to maintain child safety; the im-plications of voluntary service delivery; ap-proaches to building information supports andcommunity capacity; and strategies to increasecollaboration across human service silos.

In 2009, the Illinois Department of Children and Family Services (DCFS) began a pilot program andevaluation of one of the latest approaches to serving lower-risk children referred into the childwelfare system (differential response). This intervention approach connects low-risk households tocommunity services, and is completely voluntary. This chapter provides an overview of the practice,its effectiveness, and future policy needs.

In 2010, the Illinois Department of Children andFamily Services (DCFS) began a pilot program and

evaluation of one of the latest approaches to servinglower-risk children referred into the child welfaresystem. This approach, differential response, al-lowed caseworkers to place families in a non-inves-tigatory track when risks to child safety were low. Inthis track, caseworkers assessed and addressed fam-ily needs in an effort to prevent removing a childfrom the home.

Differential response is one of the most recentchanges in a department that historically has facedrepeated challenges. In the early 1990s, the American

Civil Liberties Union accused DCFS in a lawsuit offailing to keep children safe. As a result, a federalconsent decree provided standards for child wel-fare practice, including child placement, upfront as-sessment, caseload sizes, and protective services.1

Over the past 20 years, DCFS has radically changedits practices resulting in reduced caseloads and

Differential Response to Child Abuse and Neglect in Illinois What is differential response? How might Illinois use it to support families?By Kristin Abner and Rachel A. Gordon

N E E D T O K N O W

1 Kosanovich, A., & R.M. Joseph. 2005. “Child Welfare Con-sent Decrees: Analysis of Thirty-Five Court Actions from1995 to 2005.” Child Welfare League of America. Retrievedfrom http://www.cwla.org/advocacy/consentdecrees.pdf.

Abner Gordon

The Illinois Report 201350

fewer out-of-home placements—and Illinois hasimplemented new service delivery approaches, oneof which is differential response.2

Even with these changes, the challenges at DCFScontinue. In early 2012, the media called attentionto very high caseloads that violated a federal man-date for DCFS investigators’ monthly workload.Even so, DCFS faces the potential for additionalfunding cuts due to the current state budgetdeficit.3 In this climate, it is unclear whether andhow the differential response approach might con-tinue in the future. The five-year pilot program wasfunded by outside sources and ended in May 2012,but the evaluation was not yet completed in early2013. In this chapter, we describe the Illinois pro-gram and offer insights from other states’ programs.

What is Differential Response?

In the early to mid-1990s, researchersand advocates began to question theuniform treatment approach in childwelfare, which failed to recognize fam-ilies’ differing risk levels.4 Familiesexperiencing lower risk (such aspoverty-related neglect versus severephysical abuse) were often investigatedby child welfare caseworkers as the re-sult of a report, but if the case was notformally opened, services were notoffered.5 This situation was problematic

because these families were treated in an adversar-ial way even when they were not at risk for thetypes of extreme neglect and abuse that most peo-ple associate with child protective services. That is,these parents were often doing their best to raisetheir children under extremely impoverished cir-cumstances. The situation was also problematic be-cause these lower-risk families needed different

types of services than those offered bychild welfare agencies, such as cash as-sistance, emergency housing, or foodsubsidies. Because the system was notdesigned to refer those families to ap-propriate services, they often endedup receiving no help at all.6 Differen-tial response offered a non-adversarialway to respond to these lower-riskfamilies and to connect them withneeded services. In doing so, differen-tial response also allowed child protec-tive services to focus their inves -tigative resources on the most severecases.7

C H A P T E R 5

“Differentialresponse offered

a non-adversarial wayto respond tothese lower-riskfamilies and toconnect themwith neededservices.”

Differential Response across the country isalso referred to as:• Multiple Response• Alternative Response• Dual Track• Family Assessment Response

2 McDonald, Jess, Connie Flower, and Mike Sumski. 2005. “The Child Welfare Workforce Crisis: Implications for Leadership.” CommonGround, New England Association of Child Welfare Commissioners and Directors; Illinois Department of Children and Family Servicesn.d. “Signs of Progress in Child Welfare Reform.” Retrieved from http://www.jessmcdonald.com/f/Signs_of_Progress.pdf.

3 http://articles.chicagotribune.com/2012-06-10/news/ct-met-dcfs-triplett-20120610_1_dcfs-reforms-dcfs-workers-budget-cuts.

4 Conley, Amy. 2007. “Differential Response: A Critical Examination of a Secondary Prevention Model.” Children and Youth Services Review29: 1454–1468; Schene, Patricia A. 1998. “Past, Present, and Future Roles of Child Protective Services.” Future of Children 8 (1): 23-38.

5 Ibid.

6 U.S. Department of Health and Human Services, Administration for Children and Families, Administra tion on Children, Youth andFamilies, Children’s Bureau. 2010. “Child Maltreatment 2009.” Available from http://www.acf.hhs.gov/programs/cb/stats_research/index.htm#can.

7 Conley, Amy. 2007. “Differential Response: A Critical Examination of a Secondary Prevention Model.” Children and Youth Services Review29: 1454–1468; Waldfogel, Jane. 1998. The Future of Child Protection: How to Break the Cycle of Abuse and Neglect. Cambridge, MA: Har-vard University Press.

Institute of Government & Public Affairs • igpa.uillinois.edu 51

Differential response interventions began in 1994in Missouri and Florida. By early 2012, there were17 states implementing the approach in full and an-other 13 had a program that included at least somecomponents.8

Differential response implementation differs acrossthe country, but there is consensus on the core ele-ments of differential response. These include: • Different intervention responses are allowed for

different kinds of maltreatment reports. • The type of response is determined by the pres-

ence of imminent danger, risk level, previousreports, report source, and case characteristics.

• The type of response can change based on addi-tional information gathered (i.e. cases in the dif-ferential response track can move to theinvestigation track if child safety becomes a risk).

• Services in the differential response track arevoluntary.

• Caseworkers connect families with existing serv-ices in their communities and neighborhoods.

• A formal determination of maltreatment is notmade/substantiated, and the perpetrators andvictims are not identified.9

There is still variation in how states implementthese core features. For example, states vary inwhich families may participate in differential re-sponse, although most exclude the most seriouscases, such as those involving physical or mentalinjury, sexual abuse, abandonment, or medical

neglect.10 The suite of services offered to lower-riskfamilies also varies across states, but typically in-cludes economic support (housing, transportation,employment services), substance abuse programs,family counseling, and parenting classes.11

Two of the most salient hallmarks of differential re-sponse are its voluntary nature and its reliance oncommunity partnerships. Unlike an investigation, inwhich families must cooperate,families can choose whether ornot they will participate in differ-ential response services. How-ever, some states require serviceseven if the family declines. Insome states, if child risk is higher,the family might be re-assignedto the investigatory track, whereservices are mandatory. Otherstates mandate services throughcourt intervention but do not re-quire an investigation.12 Case-workers in differential responseprograms typically approach parents as partners, asharp contrast to the standard adversarial approachto child protective services. Differential response

The number of differential responseeligible families in some areas in thecity of Chicago.

“Two of the mostsalient hallmarksof differentialresponse are itsvoluntary natureand its reliance oncommunitypartnerships.”

8 Ibid; Waldfogel, Jane. 2009. “Differential Response.” Pp. 139-155 in Preventing Child Maltreatment: Community Approaches, edited byK. A. Dodge and D.L. Coleman. New York, NY: The Guilford Press; Crane, Kelly. 2012. “State Legislative Experience with DifferentialResponse.” Council on Contemporary Families Annual Conference, Family Impact Seminar, Chicago, IL.

9 Merkel-Holguin, Lisa, Caren Kaplan, and Alina Kwak. 2006. “National Study on Differential Response in Child Welfare.” Denver, CO:American Humane Association and the Child Welfare League of America, p. 10.

10 Kaplan, Caren and Lisa Merkel-Holguin. 2008. “Another Look at the National Study on Differential Response in Child Welfare.” ProtectingChildren 23 (1&2): 5-22.

11 National Quality Improvement Center on Differential Response in Child Protection, 2009b. “Online Survey of State Differential ResponsePolicies and Practices Findings Report. Washington, DC: U.S. Department of Health and Human Services, Administration for Childrenand Families, Children’s Bureau.

12 Kaplan, Caren and Lisa Merkel-Holguin. 2008. “Another Look at the National Study on Differential Response in Child Welfare.” ProtectingChildren 23 (1&2): 5-22.

50/ S Q M I L E

The Illinois Report 201352

caseworkers also individualize their approaches tofamilies, aiming to connect them with various com-munity-based organizations and to reinforce infor-mal supports in neighborhoods and families.13

Indeed, differential response calls for “informal andnatural helpers, drawn from families and commu-nities, to play a much more active role in childprotection.”14

These core features of differential response can ap-peal to a bipartisan agenda. On the conservativeside, differential response shifts responsibility fromthe government to neighborhoods and communi-

ties. On the liberal side, it offersa safety net to families strugglingwith structural barriers, likepoverty, rather than criminaliz-ing poor families. However, dif-ferential response programs alsoface political and practical chal-lenges. For example, neighbor-hoods and communities vary intheir capacity to support fami-lies. The very communities withfamilies most in need of differen-tial response may also be theplaces with the fewest services to

offer. Indeed, early data from Illinois’ differentialresponse implementation revealed that the city ofChicago had some areas with more than 50 differ-ential response-eligible families per square mile.15

In other areas, services may be available, but

inaccessible, due to insufficient or nonexistent pub-lic transportation.16

The History of Differential Response in Illinois

Differential response began in Illinois as a way toreduce recurrence of child mistreatment in families,offer child protective services with a family cen-tered approach, promote community well-being,and address racial disproportionality in the childwelfare system. In 2008, the Illinois Department ofChildren and Family Services hired the Childrenand Family Research Center at the University ofIllinois to compile existing literature, which gener-ated interest in the beginning of a differential re-sponse model in Illinois. Based on the findings, thedepartment put together a group of stakeholders in2009 to formulate a strategy for developing differ-ential response criteria, determining appropriateoutcome measures for the evaluation, identifyingcomputer system changes, and formulating publicawareness campaign.17,18

In August 2009, Governor Pat Quinn signed theDifferential Response Program Action (SB807), andin December 2009, the National Quality Improvement

C H A P T E R 5

Year Governor Quinn signed theDifferential Response Program Action

13 Waldfogel, Jane. 1998. The Future of Child Protection: How to Break the Cycle of Abuse and Neglect. Cambridge, MA: Harvard UniversityPress, p. 138.

14 Ibid.

15 McEwen, Erwin. 2010. “Differential Response.” Presentation at the Illinois Child Welfare Data Summit: 1st Annual Leadership Summit.Chicago, IL.

16 In short, as Crain and Tonmyr argue, “it is critical to acknowledge that a weakness of some differential response systems is the as-sumption that community support services are available.” Crain, Jennifer and Lil Tonmyr. 2007. “Differential Response Models of ChildProtection and Implications for the Canadian Incidence Study of Reported Child Abuse and Neglect – 2008.” Pp. 21-26 in Canada’sChildren: Child and Youth Maltreatment, edited by the Child Welfare League of Canada. Ottawa, Canada: Author. (p. 22).

17 Jones, Womazetta, William Wolfe, Tamara Fuller, and Kathleen Kearney. 2010. “Putting It All Together: Lessons Learned from Imple-menting Differential Response in Illinois.” 2010 Conference on Differential Response in Child Welfare.

18 Source for flowchart: http://www.state.il.us/DCFS/docs/CFS_1050-49_Differential_Response_Brochure.pdf

“The verycommunities withfamilies most inneed of differentialresponse may alsobe the places withthe fewest services

to offer.”

2009

Institute of Government & Public Affairs • igpa.uillinois.edu 53

Center on Differential Response in Child Protec-tive Services (QIC-DR) awarded Illinois a grant topilot differential response for five years and toevaluate the program’s effectiveness, including arandomized control trial. The program, calledPathways to Strengthening and Supporting Families(PSSF), was implemented statewide in November2010 (Figure 1).

Illinois’ differential response criteria focused onlower-risk neglect allegations. Families were screenedfor the differential response track, as opposed to the

child welfare investigations track, when a report ofchild abuse or neglect came into the Department ofChildren and Family Services’ reporting hotline withany of the following allegations: lock out; inade-quate food, shelter, or clothing; environmental neg-lect; mental injury; medical neglect; inadequatesupervision; or risk of harm due to neglect. Childphysical and sexual abuse allegations were not eli-gible for differential response. Additional criteria forthe Illinois demonstration and evaluation included:• The family must not have had a previous or

pending indicated report of abuse or neglect; • The family’s address and other identifying infor-

mation was available;• The alleged perpetrator was the parent or legal

guardian of the child;• The children were not wards of the court; and• It was not necessary to take protective custody

of the child.

Illinois implemented a paired team approach toservice where one representative from DCFS ac-companied one local community-based serviceworker (SSF worker) for the initial visit with the

Figure 1Illinois DCFS Pathways to Strengthening and Supporting Families

Report of Alleged Abuse/Neglect to 24-Hour Hotline

Screened Families: Meet State and

Local Criteria Eligible for Family Assessment

Mandatory Investigation

Initial Screening for CPS

Control Group (Investigation)

Experimental Group(Non-Investigational)

Eligibility for Investigational or

Family Assessment Path

TRADITIONALINVESTIGATORS

Random Assignment

TRADITIONALINVESTIGATORS

SPECIAL DCFS UNIT

AND PRIVATE AGENCY

HOTLINEWORKERS

HOTLINEWORKERS

ELECTRONICDECISION

Source: Information from http://www.state.il.us/DCFS/docs/CFS_1050-49_Differential_Response_Brochure.pdf

“ Beginning January 1, 2010, the Department of Children and

Family Services may implement a 5-year demonstration of a

‘differential response program’ in accordance with criteria,

standards, and procedures prescribed by rule. The program

may provide that, upon receiving a report, the Department

shall determine whether to conduct a family assessment or

an investigation as appropriate to prevent or provide a

remedy for child abuse or neglect.”

— SB807

The Illinois Report 201354

family. These visits occurred within three days ofthe report. The SSF worker then acted as a coachand advocate for the family. The SSF worker had amaximum of 12 cases at one time, and served as anagent of change for the family.19

The home visit included an interview with the childto assess his/her developmental level and variousother health and risk assessments for both child andfamily. The SSF worker stayed in contact with thefamily on a weekly basis in-person to provide short-term support, for example teaching a mother how todiaper her new baby. Or the worker might connectthe mother to federal and statewide food assistanceprograms or help her secure reliable childcare. TheSSF worker could also help the family identify its ex-isting social support network, which was assessedin terms of its helpfulness, intensity, durability, ac-cessibility, proximity, reciprocity, and size.20 The caseremained open for 90 days, but the family could re-quest extensions for another 90 days.21

The Children and Family Research Center wasexpected to complete its evaluation of the state dif-ferential response program by mid-2013. The eval-uation includes a process component, an outcomecomponent, and a cost analysis.22 The randomizedcontrol trial (RCT) offering services to familiesended on May 25, 2012. The process evaluationcomponent includes implementation informationfrom steering committee meetings to case tracking.

Outcome data will be provided through adminis-trative case records, caseworker case closing re-ports, surveys from families after case closure, focusgroups with DCFS staff and community providers,interviews with families, and observations of fam-ily-caseworker interactions.23 Once evaluation ma-terials become available, Illinois will assess whetherto re-institute the differential response track.

Differential Response in Other States

Differential response varies considerably across thecountry, which makes it difficult to generalize butalso provides many alternative models for Illinois toconsider. Overall, 21 states have completed or arecurrently conducting some sort of evaluation of dif-ferential response.24 Minnesota, Ohio, and New Yorkhave completed a randomized control trial (RCT)experimental evaluation of differential response(Figure 2). Illinois and Colorado are in the process of

C H A P T E R 5

19 Jones, Womazetta. 2012a. “Differential Response in Illinois – What Is It?” Council on Contemporary Families Annual Conference, FamilyImpact Seminar, Chicago, IL; Fuller, Tamara. 2012. “Differential Response: Sounds Great, But Does it Work?” Council on ContemporaryFamilies Annual Conference, Family Impact Seminar, Chicago, IL.

20 Illinois Department of Children and Family Services, Division of Service Support, Office of Training and Professional Development.2010. Illinois’ Pathways to Strengthening and Supporting Families Training Modules. Springfield, IL.

21 Fuller, T.L., Kearney, K.A., & Lyons, S.L. (2012). Differential Response in Illinois: 2011 Site Visit Report. Urbana, IL: Children and Family ResearchCenter, University of Illinois at Urbana-Champaign. http://cfrc.illinois.edu/pubs/rp_20120327_DifferentialResponseInIllinois2011SiteVisitReport.pdf.

22 Ibid.

23 Children and Family Research Center. (2012). An Introduction to Differential Response. http://cfrc.illinois.edu/pubs/bf_20120101_AnIntroductionToDifferentialResponse.pdf

24 National Quality Improvement Center on Differential Response in Child Protection, 2011. “Differential Response in Child ProtectiveServices: A Literature Review, Version 2.” Washington, DC: U.S. Department of Health and Human Services, Administration for Childrenand Families, Administration for Children, Youth, and Families, Children’s Bureau.

For More Information

A report on Illinois' differential response research anddemonstration project is expected by the end of 2013.Please check the University of Illinois' Children and Family Re-search Center web site for its release at http://www.cfrc.illi-nois.edu/differentialresponse.php.

For more information on differential response in Illinois,please contact Womazetta Jones, Project Director for Differ-ential Response at womazetta.jones @illinois.gov.

Institute of Government & Public Affairs • igpa.uillinois.edu 55

finishing RCTs. Ohio is doing an extended evalua-tion that is expected to be finished in 2013. Otherstates have completed quasi-experimental and non-experimental evaluations.25 Quasi-experimental de-signs have matched and compared families or sitesbased on similar characteristics. Non-experimentaldesigns have included case reviews or pre-and post-data comparison.26 We focus primarily on resultsfrom the RCTs and quasi-experiments because theyprovide the most rigorous evidence of the causal im-pact of the programs.

The following is an overview of previous state eval-uations. For a list of the evaluations, please see theend of this chapter.

Child Safety and Recidivism

Child safety is an important outcome for differen-tial response evaluations because children remainin the home after they come to the attention of childprotective services.

Overall, evaluations of differential response showthat children are not at increased risk when placedinto the differential response track as opposed tothe investigation track. These evaluations generallydefined child safety in terms of caseworkers’ as-sessments and by using re-reports into the childwelfare system. Other outcomes like child injuriesor deaths have been less often examined.

The equivalent safety of children in both tracksfrom the caseworker perspective has been shownin several states, including Minnesota, Ohio, andMissouri. In Missouri, researchers coded the safety

concerns and severity descriptions in case files,comparing pilot to control counties. Researchersfound improved safety for children in the pilot (dif-ferential response) counties with reported neglectof basic needs, lack of supervision and proper care,and less serious physical and verbal abuse, such asinjuries resulting from discipline.27

Equivalent recurrence rates were descriptively doc-umented with data from five states in the NationalChild Abuse and Neglect Data System (NCANDS),revealing that repeated reports to the system wereabout the same for families in each track.28 Evalua-tions in five states also showed that recurrence was

25 Fuller, Tamara. 2012. “Differential Response: Sounds Great, But Does it Work?” Council on Contemporary Families Annual Conference,Family Impact Seminar, Chicago, IL.

26 National Quality Improvement Center on Differential Response in Child Protection, 2011. “Differential Response in Child ProtectiveServices: A Literature Review, Version 2.” Washington, DC: U.S. Department of Health and Human Services, Administration for Childrenand Families, Administration for Children, Youth, and Families, Children’s Bureau.

27 Siegel, G. L., & Loman, L. A. (1997). Missouri family assessment and response demonstration: Final evaluation report. St. Louis, MO: Instituteof Applied Research.

28 Ortiz, Mary Jo, Gila R. Shusterman, and John D. Fluke. 2008. “Outcomes for Children with Allegations of Neglect Who Receive AlternativeResponse and Traditional Investigations: Findings for NCANDS.” Protecting Children 23 (1&2): 57-70.

Figure 2Differential Response Evaluations by State

RCT EvaluationsRCT Evaluations (Ongoing)Quasi-Experimental EvaluationsQuasi-Experimental Evaluations(Ongoing)

* Expected Completion in 2013

**

Source: See footnote 24 on page 54.

The Illinois Report 201356

either lower in the differential response than the in-vestigatory track (in Kentucky, Missouri, and Min-nesota) or no different between the two tracks (inNew York and North Carolina). After re-reporting,three states also found that the results of investiga-tions were more favorable or no different for fami-lies originally in the differential response track.Specifically, fewer differential responsefamilies had children removed in sub-sequent investigations in both Min-nesota and Ohio. In Nevada, there wasno significant difference in child re-moval rates.

Service Delivery and Take-Up

Differential response arose from thepremise that lower-risk families need dif-ferent kinds of services than higher-riskfamilies. Voluntary service delivery isalso a core component of differential re-sponse, thus it is important to examinewhether families offered services actu-ally accept them.

If families refuse to participate in differential re-sponse, then the case is either closed or, in somestates, moved into the investigatory track. In Illinois,the case is closed if the family refuses to receive serv-ices. In contrast, when in the investigatory track,families can be required by the court to participatein services.

State evaluations indicate that families in the dif-ferential response track receive more services, morequickly than families in the investigations track. InMissouri, differential response families receivedservices within 17 days of the incident report asopposed to 34 days for families receiving investi-gations. In Minnesota, 54 percent of differential re-sponse families reported receiving specific servicesother than case management as opposed to 36 per-cent of the control group of investigations families.This service advantage is especially true in termsof basic items needed for children, like diapersand formula and economic help like assistancewith utility bills (Colorado; Minnesota; Missouri;New York; Ohio). The counties in North Carolina

implementing differential response also increasedthe number of minutes per case on “frontloading”services pre- and post-implementation from (344minutes per child to 441 minutes per child, respec-tively). Frontloading is defined as bringing servicesand supports more quickly to families. Differentialresponse workers were also more likely to link

families to community resources, whichis an important component of differen-tial response. For example, in New Yorkdifferential response caseworkers re-ferred families to neighborhood organ-izations (11 percent more than investi-gations), community action groups (12percent more), and self-help groups (13percent more).

Family Satisfaction and Engagement

Differential response caseworkers in-volve families as partners, rather thanas clients, thus family satisfaction andengagement are important outcomes.

Consistently, state evaluations have found thatfamilies receiving differential response were moresatisfied with the intervention that they have re-ceived, and caseworkers also report that familiesare more satisfied (Minnesota; Missouri; New York;North Carolina; Ohio).

The Minnesota evaluation, for example, docu-mented this greater satisfaction and involvementamong families in the differential response track.29

Workers in Minnesota likewise reported that fami-lies in differential response were more cooperative.In Ohio, families reported being offered more serv-ices, deemed their caseworker more helpful, andwere more likely to participate in services thanfamilies in the investigative track.30 Families inNew York’s differential response track reported

C H A P T E R 5

“Consistently,state evaluationshave found thatfamilies receivingdifferentialresponse weremore satisfiedwith the

intervention thatthey havereceived…”

29 Loman, L. Anthony. 2009. “Differential Response and Fam-ily Poverty: Evidence from Evaluations.” Conference on Dif-ferential Response in Child Welfare.

30 Ibid.

Institute of Government & Public Affairs • igpa.uillinois.edu 57

more positive feelings after the initial meeting withcaseworkers, and reported their caseworker lis-tened and respected them more than investigativetrack families did. And, if the family had prior ex-perience with child protective services, they noteda better experience when in the differential re-sponse track.

Family satisfaction is one of the few areas in whichfindings are available for Illinois’ differential re-sponse evaluation. So far, there is evidence thatfamilies in the differential response track are moresatisfied with the quality and substance of interac-tions with workers than other families. Differentialresponse families also report that they have morepathways to receive more services and are satisfiedwith the help they receive. Caseworkers likewisereport that the differential response approach trackallows them to make more progress with clients.31

Considerations for the Future

As already noted, differential response programscan appeal to lawmakers from both sides of the aisle,which speaks to its potential long-term viability.Consistent with conservative ideology, differentialresponse shifts the role of child protection from thefederal and state governments to local governmentand community-based organizations. By offeringvoluntary services to families, it also promotes fam-ily responsibility and accountability. On the liberalside, differential response programs recognize thatsituational forces, such as poverty, contribute to fam-ilies coming into contact with the child welfare sys-tem. Such families are connected with social safetynets rather than criminally investigated.

Yet differential response also faces political and prac-tical challenges. We highlight four main considera-tions and implications for Illinois lawmakers toconsider: (a) maintaining child safety, (b) understand-ing the implications of voluntary service delivery, (c)building informal supports and community capacity,and (d) collaborating across human service silos.

Maintaining Child Safety. Child protective services al-ways face a tension between a desire to keep familiesintact and a desire to protect children from harm.Public and political support can be quickly under-mined when a child is seriously injured or killedafter being left at home, despite a mistreatment re-port to the state. Due to this possibility, states imple-menting differential response have excluded thehighest-risk cases from participating. States also pro-vide continuous safety and risk monitoring for chil-dren who are in the differential response track. InIllinois, the community-based worker has weekly in-person contact with families in the differential re-sponse track. Continued vigilance to child safety willbe essential for maintaining support for future dif-ferential response legislation.

Understanding the Implications of Voluntary ServiceDelivery. As discussed, families’ voluntary partici-pation is a core feature of differential response. InIllinois, about 90 percent of families participatedwhen offered differential response services. We donot yet fully understand why some of these fami-lies accepted services and why others did not, andthis knowledge will be important to increasing par-ticipation if the program is reinstated.32 More infor-mation is also available from other states aboutshort-term versus long-term follow-up with fami-lies. For example, it is not clear whether differentialresponse families remain more likely to receiveservices after contact with their caseworker ends.Such longer-term follow-ups will also be importantto future programming.

31 Fuller, Tamara. 2012. “Differential Response: Sounds Great,But Does it Work?” Council on Contemporary Families An-nual Conference, Family Impact Seminar, Chicago, IL.

32 Schene, Patricia A. 2005. “The Emergence of DifferentialResponse.” Protecting Children 20 (2&3): 4-7.Source: (Crane 2012)

Considerations for Policymakers:• Service array• Safety issues• Family participation in services• Reoccurrence of families in DCFS system• Definitions of abuse and neglect and determinations of

levels of risk• Cost savings• Community resources

The Illinois Report 201358

Building Informal Supports and Community Capacity. Asdiscussed above, differential response relies on exist-ing community services. Such resources will varyacross communities, and may be least available in theareas with the highest concentration of families inneed of services. In the future, sub-state evaluationswould allow Illinois and other states to better under-stand whether differential response is more success-ful in some communities than others. Illinois’evaluation will include a sub-state analysis. More in-formation is also needed about how caseworkers es-tablish linkages with local community agencies, and

whether this relationship buildingvaries across locales and types ofagencies. For example, Missouri’sevaluation indicates that workersreported a better relationship withcommunity agencies and schools,but no change in the relationshipwith the juvenile court and police,as a result of differential response.

Collaborating Across Human ServiceSilos. More cost-benefit analysesof differential response are alsoneeded. Differential response hasthe potential to increase efficien-

cies within state child protection agencies becausecaseworkers can focus investigations on the highest-risk families. However, to the extent that differentialresponse caseworkers duplicate efforts of otherhuman service agencies, state spending may be lessefficient across agencies. Currently, state cost-benefitanalyses of differential response have shown mixedresults. Ohio’s evaluation showed that differentialresponse cost about $100 more per family when cal-culated over a 15-month period. However, differen-tial response in Minnesota cost around $200 less perfamily when considered long-term (the mean lengthfor counties for the second follow-up was 453 days).It is also unclear how community-based differentialresponse workers are connecting families to otherhuman service systems, such as Temporary Assis-tance for Needy Families (TANF; cash assistance) orthe Supplemental Nutrition Assistance Program(SNAP; food stamps). In order to truly enable orga-nizational change within DCFS and across agencies,multiple stakeholders must be involved. Differential

response will likely be best sold across state agenciesas one component of a wider effort to more effec-tively deliver a system of care to clients. TippecanoeCounty in Indiana recently reformed its child welfaresystem after a highly publicized case involving achild’s death. Community organizations plannedpublic forums to raise awareness about the preva-lence of child abuse and neglect in the county, thetypes of services available for families, and to fosterideas from the community about moving forward tokeep children safe. The county also began to fostercollaboration between agencies as a way to preventchild mistreatment and improve service delivery tofamilies in the caseload.33 Externally, the programneeds to be well described to the media, legislators,judges, and other stakeholders, so that they under-stand this paradigm shift in child welfare and itslong-term policy goals.

Conclusions

In Illinois and the nation, child protective servicesface a continual tension between keeping childrensafe and keeping families intact. Differential re-sponse represents the most recent shift in childwelfare practice to help address this tension, rec-ognizing that families differ in their level of riskand needed response. However, the impact of thisprogram not only for families, but also state gov-ernments, is only partially understood.

Unfortunately, this innovative approach is beingconsidered at the same time that state budgets areshrinking.34 The short-term investment in startingthe program might lead to increased efficiencieswithin and across state agencies, and to longer-term benefits to the extent that fewer families comeback into contact with child protective services and

C H A P T E R 5

33 Biggs-Reed, P., Smith Grossman, A., Rush, L., & Wilson, S.(2008). “White Paper: Understanding Child Abuse andNeglect in Tippecanoe County.” Retrieved from http://www.ourkidstippecanoe.org/child_abuse_neglect_white_paper1.pdf.

34 Waldfogel, Jane. 2009. “Differential Response.” Pp. 139-155in Preventing Child Maltreatment: Community Approaches,edited by K. A. Dodge and D.L. Coleman. New York, NY: TheGuilford Press.

“The recessionreduced states’abilities to fundthese short-terminvestments. Yetthe recession alsoaccentuated theneed for theprogram.”

Institute of Government & Public Affairs • igpa.uillinois.edu 59

to the extent that more families are connected withservices to get and keep them on their feet.35

The recession reduced states’ abilities to fund theseshort-term investments. Yet the recession also accen-tuated the need for the program. In a study of the ef-fect of the current economic recession on childwell-being, children in poverty were at higher riskfor both abuse and neglect, and reports of child neg-lect increase during times of economic recession.36

The impending report on the full evaluation of dif-ferential response in Illinois will importantlysharpen understanding of whether differential re-sponse met its goals, and whether it has the poten-tial to bolster existing policy and foster the creationof new programs targeted to vulnerable families—not only in Illinois, but also across the country.

State Evaluations

Center for Child and Family Policy. 2006. MultipleResponse System (MRS) Evaluation Report tothe North Carolina Division of Social Services(NCDSS). Duke University.

Huebner, R. A. (2005). Program evaluation of themultiple response system: Kentucky.

Loman, L. Anthony and Gary L. Siegel. 2004a.“Minnesota Alternative Response Evaluation:Final Report.” A Report of the Institute ofApplied Research, St. Louis, Missouri.

Loman, L. A., & Siegel, G. (2004b). Differentialresponse in Missouri after five years: Final report. St.Louis: Institute of Applied Research.

Loman, L. A., Filonow, C.S., & Siegel, G. (2010).Ohio alternative response evaluation: Final report. St.Louis, MO: Institute of Applied Research.

Ruppel, Joanne, Yufan Huang, Gail Haulenbeek.2011. “Differential Response in Child ProtectiveServices in New York State Implementation,Initial Outcomes and Impacts of Pilot Project.”New York State Office of Children and FamilyServices.

Siegel, G. L., & Loman, L. A. (1997). Missouri familyassessment and response demonstration: Finalevaluation report. St. Louis, MO: Institute ofApplied Research.

Siegel, G. & Loman, L. (2006). Extended follow-upstudy of Minnesota’s Family

Assessment Response: Final report. St. Louis, MO:Institute of Applied Research.

Winokur, M., et al. 2012. Colorado year 1 site visitfinal report. Fort Collins, CO: Colorado StateUniversity Social Work Research Center. �

Family Impact Seminar

This chapter is a component of the 2012 Illinois Family Impact Sem-inar, on the topic of Differential Response to Child Abuse and Neglect:Where is Illinois and What Can We Learn from Other States? The 2012Family Impact Seminar was part of the Council on ContemporaryFamilies Annual Conference, co-sponsored by the University-BasedChild and Family Policy Consortium. The event, “Crossing Bound-aries: Public and Private Roles in Assuring Child Well-Being,” was heldin Chicago on April 27 and 28. At the 2012 Family Impact Seminar,hosted by IGPA and several co-sponsors, researchers, practitioners,and experts on family policy gathered to discuss the effectivenessof differential response, and ways to ensure the approach’s successin Illinois. The panel included widely respected experts on the dif-ferential response approach, including:

Womazetta Jones, Project Director, Differential Response, Illinois De-partment of Children and Family Services

Tamara Fuller, Director, Children and Family Research Center, Schoolof Social Work, University of Illinois at Urbana-Champaign

Kelly Crane, Child Welfare Policy Specialist at the National Confer-ence of State Legislatures

Joel Rosch, Senior Research Scholar, Policy Liaison, Center for Childand Family Policy, Duke University

Acknowledgments

The Illinois Family Impact Seminar is an annual series directed byDr. Rachel Gordon at the University of Illinois’ Institute of Govern-ment and Public Affairs. The series connects decision makers withresearch evidence to inform pressing family policy issues. The IllinoisFamily Impact Seminars benefit from the good advice of membersof the Policy Network for Family Impact Seminars, directed by KarenBogenschneider at the University of Wisconsin-Madison and themembers of the Illinois Family Impact Seminars advisory committee.Additional materials and video of the 2012 seminar can be found atigpa.uillinois.edu/pe/fis.

35 Loman, L. Anthony. 2005. “Criminal Arrests for SeverePhysical and Sexual Abuse.” A Report of the Institute of Ap-plied Research, St. Louis, Missouri.

36 Sell, Katherine, Sarah Zlotnik, Kathleen Noonan, and DavidRubin. 2010. “The Recession and Child Maltreatment.” TheEffect of the Recession on Child Wellbeing. First Focus,Washington, DC.