A Disability Policy Research Forum

1. Mathematica Policy Research Presents A Disability Policy Research Forum On The Air March 18, 2010 12:00 to 1:30 pm

2. Staying Employed: Early Intervention for Adults with Potentially Disabling Conditions Speakers: Henry Ireys, Mathematica Policy Research Gilbert Gimm, Mathematica Policy Research Patricia Owens, Government Accountability Office Disability Policy Research Forum March 18, 2010

3. Demonstration to MaintainIndependence and Employment (DMIE): Preliminary Findings from the National Evaluation Presentation at the Disability Policy Research Forum March 18, 2010 Henry Ireys ● Gilbert Gimm ● Boyd Gilman●Noelle Denny-Brown ● Sarah Croake

4. National DMIE Evaluation Goals • Address the primary question: • Can an early intervention program providing medical and employment assistance prevent or delay the loss of work and independence due to a physical or mental health condition before a person becomes disabled? • Build on state evaluations in Kansas, Texas, Minnesota, and Hawaii • Synthesize lessons learned from cross-state comparisons

5. Research Questions • Who enrolled in the DMIE across the four states? • What were the early impacts of the program on disability applications and employment outcomes? • What lessons were learned from implementing the program?

6. Preview of Early Impacts • No difference in the percentage employed or average hours worked 12 months after DMIE enrollment • Lower percentage of participants applying for disability benefits 12 months after DMIE enrollment

7. DMIE Eligibility Criteria • Working at the time of enrollment • Age 18 to 64 • Not currently applying for or receiving disability benefits at the time of enrollment

8. DMIE Program Components • Enhanced medical services • Wraparound coverage (dental, vision) beyond existing Medicaid coverage; expedited mental health visits • Employment supports • Peer support; vocational rehabilitation services

9. DMIE Program Components (cont’d.) • Intensive, person-centered case management • Wellness navigator; life coaching • Subsidies • Coverage of deductibles and co-payments; premium subsidies

10. DMIE Target Populations by State • Minnesota • Working adults with severe mental illness in public insurance programs • Texas • Working adults with severe mental illness or behavioral health/physical conditions in safety-net program

11. DMIE Target Populations by State (cont’d.) • Kansas • Working adults with physical and mental health conditions in state high-risk insurance pool • Hawaii • Privately insured working adults with diabetes

12. DMIE Enrollment Total at Baseline (n=4,099)

13. Baseline Health Characteristics and Age at Enrollment, by State Note: SF-12 health scores are norm-based, with 50.0 representing the national average. Lower scores indicate worse health.

14. Baseline Employment Characteristics and Education, by State

15. Evaluation Design and Analysis • Randomized design in all four states • Control group (“business as usual”) • Treatment group (offered additional services) • Intent to treat (ITT) analysis • Regression-adjusted estimates • Accounts for participant age, withdrawals, enrollment year, and prior applications or baseline hours worked

16. Data Sources • State-level survey and administrative data • Rounds 1 and 2 • SSA 831 file on disability applications • Data through fall 2009 • Ticket Research File, Master Earnings File • Data through 2008; data to analyze one-year impacts on payments and earnings will be available in fall 2010 • Site visit interviews; descriptive reports

17. Preliminary Results • Impact on employment • Percentage employed • Monthly hours worked • Impact on disability benefits • Percentage applying for SSA disability benefits • Lessons learned about implementation

18. Percentage Employed in Texas and Minnesota, 12 Months After Enrollment Note: p = .55

19. Average Hours Worked in Last Month in Texas and Minnesota, 12 Months After Enrollment Note: p = .76

20. Percentage Applying for Disability Benefits in Texas and Minnesota, 12 Months After Enrollment Note: p = .05

21. Percentage Applying for Disability Benefits in Three States, 12 Months After Enrollment Texas Minnesota Kansas Note: p = .12, .20, and .47, respectively

22. Lessons Learned • DMIE can be implemented in diverse settings to serve various target populations • Flexibility for states to design/customize benefits • Program diversity strengthens evaluation • Building DMIE around existing programs makes it easier to identify candidates from a “captive pool” • Obtain information to focus recruitment effort • Program services build on existing benefits, can be deployed more quickly with existing network

23. Lessons Learned (cont’d.) • Participants value person-centered case management • Key component of program design in every state (system navigation, life coaching) • Helps participants address barriers to employment and obtain services • Working adults must be recruited at the right point on the disability trajectory • Too early: services may not be needed • Too late: services may not help prevent disability

24. Summary • Early findings on impacts • Evidence that early intervention programs can reduce disability applications • No short-term impact on employment • Robust model for early intervention • Can be implemented with diverse populations; flexible enough for states to customize benefits • Implementation findings • Provide initial foundation for “best practices” in early intervention

25. Final Thoughts • The power of a good idea • Evidence matters • Leadership matters, too

26. Our Partners Claudia Brown, Melissa Hulbert, Joe Razes, Steve Hrybyk (Disabled & Elderly Health Programs Group, CMS) Becky Ozaki, Jean Isip Schneider, Tammy Tom, Denise Uehara (Hawaii) Jean Hall, Jan Moore, Jenifer Telshaw, Mary Ellen Wright (Kansas)

27. Our Partners (cont’d.) Tom Bohman, Dena Stoner, Lynn Wallisch, Tim Weatherby (Texas) Jennifer Brya, Karen Linkins, Maryalice Mowry, Katy Olson (Minnesota)

28. Contact Information Henry T. Ireys, Ph.D., and Gilbert Gimm, Ph.D. Mathematica Policy Research 600 Maryland Avenue, SW, Suite 550 Washington, DC 20024 hireys@mathematica-mpr.com ggimm@mathematica-mpr.com Access reports/issue briefs on the DMIE at www.disabilitypolicyresearch.org www.mathematica-mpr.com/disability/dmie.asp

29. DISABILITY POLICY FORUMMARCH 18, 2010 COMMENTS ON INITIAL FINDINGS OF NATIONAL EVALUATION OF DEMONSTRATION TO MAINTAIN INDEPENDENCE AND EMPL0YMENT (DMIE)

30. DISABILITY POLICY FORUMMARCH 18, 2010 About GAO • The government accountability office (GAO) is the audit, evaluation, and investigative arm of congress whose work is to help improve the performance and accountability of the federal government. • GAO evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help congress make informed oversight, policy, and funding decisions.

31. DISABILITY POLICY FORUMMARCH 18, 2010 • GAO has previously reported that “demonstration projects examining the impact of social programs aim to provide evidence of the feasibility or effectiveness of a new approach or practice and are inherently complex and difficult to conduct.” (GAO 05-19). • Measuring outcomes, ensuring consistency and quality of data collected at various site locations establishing a causal connection between outcomes and program activities, and separating out the influence of extraneous factors can raise formidable technical and logistical problems. (GAO 08-1053).

32. DISABILITY POLICY FORUMMARCH 18, 2010 PURPOSE OF DMIE: TO ADDRESS THE QUESTION : Can an early intervention program of medical and employment assistance prevent or delay the loss of work and independence due to a physical or mental health condition before a person becomes disabled? PERSONS IN DEMONSTRATION WERE WORKERS WITH POTENTIALLY DISABLING CONDITIONS

33. DISABILITY POLICY FORUMMARCH 18, 2010 • Demonstration itself is an example of inherent difficulties as discussed in the prior GAO reports. • While the generic principle was the same, there were in effect four different populations and interventions. • Evidence in all the populations that persons with chronic health conditions do work and want to continue to work. • Even controlling for differences, it is difficult to say precisely what was learned that can be the basis of policy decisions. • This is not the complete evaluation and mainly tells us what happened under four different scenarios.

34. DISABILITY POLICY FORUMMARCH 18, 2010 • But it does provide addition clues to early intervention that can increase participation of persons with disabilities.

35. DISABILITY POLICY FORUMMARCH 18, 2010 • What do we mean by “potentially” disabling conditions. • Need to have a frame of reference. • IOM in 1991 “Disability in America” • Disability is the expression of a physical or mental limitation in a social context – the gap between a person’s capabilities and the demands of the environment….It is the interaction of their physical or mental limitations with social and environmental factors that determines whether they have a disability. • One major conceptual framework is the functional framework first described by Saad Nagi in the 1950s

36. DISABILITY POLICY FORUMMARCH 18, 2010

38. DISABILITY POLICY FORUMMARCH 18, 2010 Work Disability Programs are generally designed to pay persons who cannot work as defined by a particular program definition. Few designed to consider work participation, increasing work capabilities and job placement as part of a disability program process – Australia, European Countries

39. DISABILITY POLICY FORUMMARCH 18, 2010 MEDICAL AND TECHNOLOGICAL ADVANCES, THE SHIFT TOWARD SERVICE- AND KNOWLEDGE-BASED JOBS, AND CHANGING SOCIETAL VIEWS—SHOULD BE INCREASING THE OPPORTUNITY FOR MORE PEOPLE WITH DISABILITIES TO WORK AND SUPPORT THEMSELVES (CURRENT ECONOMIC CONDITIONS ASIDE). CURRENT DATA INDICATES THIS IS NOT HAPPENING. A CHALLENGE IS TO EXAMINE DISABILITY POLICY AND PROGRAMS IN LIGHT OF NEW AND PROMISING CONCEPTS AND PRACTICES THAT HAVE POTENTIAL TO INCREASE WORK PARTICIPATION.

41. DISABILITY POLICY FORUMMARCH 18, 2010 • Any objective findings that inform the policy issue of increasing work participation of persons at risk of “work” disability are important. • Sometimes findings come in the form of better framed questions concerning what can contribute to improvements and who needs to be involved. • Early interventions may reduce SSDI applications and the jury is out on the impact on employment outcomes per se. • States are shown as important in early intervention programs.

42. DISABILITY POLICY FORUMMARCH 18, 2010 • GAO RECENT FORUM ON IMPROVING WORK OUTCOMES OF FOR ADULTS WITH DISABILITIES • Expert Group of 60. • Two surveys which identified interventions to help people stay at work or return to work. • Forum of Experts to discuss policy areas identified through surveys. • Report in the late summer.

43. DISABILITY POLICY FORUMMARCH 18, 2010 • IN CLOSING I CAN ONLY AGREE WITH: • THE POWER OF A GOOD IDEA • EVIDENCE MATTERS • LEADERSHIP MATTERS TOO

44. DISABILITY POLICY FORUMMARCH 18, 2010 • RELATED GAO STUDIES • Federal Disability Assistance: Wide Array of Programs Needs to be Examined in Light of 21st Century Challenges. GAO-05-626. • Results-Oriented Government: Practices That Can Help Enhance and Sustain Collaboration among Federal Agencies. GAO-06-15. • Highlights of a GAO Forum: Modernizing Federal Disability Policy. GAO-07-934SP. • A Call for Stewardship: Enhancing the Federal Government’s Ability to Address Key Fiscal and Other 21st Century Challenges. GAO-08-93SP. • Federal Disability Programs: More Strategic Coordination Could Help Overcome Challenges to Needed Transformation. GAO-08-635. • Management Controls Needed to Strengthen Demonstration Projects GAO-08-1053

45. Q & A