Here’s the Evidence: A Policymaker’s Guide to Evidence-Based Disability Policy

Dec 12, 2017

David Stapleton speaks on Capitol Hill during a 2015 briefing.
David Stapleton speaks on Capitol Hill during a 2015 briefing.

During a recent meeting with staff from the Senate Health, Education, Labor and Pensions (HELP) Committee, I was asked a question central to Mathematica’s recent and ongoing disability research: “Our committee is dedicated to helping more individuals with disabilities find and sustain jobs. Where do you recommend we focus our energy and resources?”

The answer to this question is surprisingly clear: investments in “early interventions”—services and other supports designed to help people become financially self-sufficient before they become dependent on disability benefits—targeted to two populations: (1) workers whose continued self-sufficiency is threatened by injury or illness, and (2) youth who have long-term medical conditions or functional limitations as they transition to working age. For decades, federal efforts to improve the economic status of people with disabilities have focused on helping adult beneficiaries of Social Security Disability Insurance (SSDI) and Supplemental Security Income (SSI) work and earn enough money to give up their benefits. It appears that none of those efforts have paid off in a substantial way. That doesn’t mean they never will, but available evidence indicates that investing in early interventions will lead to more rapid gains in the economic status of adults with disabilities.

The first target population is workers who experience “needless work disability” after an injury or the onset of a chronic condition. Mathematica’s work under the Department of Labor’s Stay-at-Work/Return-to-Work Policy Collaborative, found that, instead of receiving the medical, rehabilitative, and other services and supports that would help them stay at work, many workers with injuries or chronic conditions fall through the cracks in a fragmented support system. Further, the incentives in the support system nudge such workers toward disability benefits. In 2015, almost 2.4 million workers applied for SSDI benefits. Based on historical data, about 40 percent of them will ultimately receive awards. We do not know how many of them experience needless work disability, but the evidence suggests the number is quite large.

For example, needless work disability seems particularly common among workers with lower back pain and other musculoskeletal conditions. In 2016, almost 37 percent of SSDI awards were made to workers with “diseases of the musculoskeletal system and connective tissue,” about half of which involve lower back pain. The Centers of Occupational Health & Education (COHE) program in Washington State, which provides care coordination and financial incentives to facilitate return to work when the condition is work-related (that is, covered by the state’s workers’ compensation system), offers a promising example of how needless work disability can be prevented. A rigorous 2011 study of the pilot program found a 30 percent reduction in lost workdays over the first 12 months for those with lower back problems. Results were not as strong for all workers in the pilot, but were nonetheless impressive: a 21 percent reduction in lost work days and a 7 percent reduction in medical expenditures, net of the expenditures for the new services. The new services way more than paid for themselves in the first 12 months alone. A follow-up analysis found preliminary evidence that the number of workers entering SSDI over the next eight years decreased by about 25 percent.

As I said to the Senate HELP Committee staff during that meeting, there is strong evidence on the services and supports that can help individual workers avoid needless work disability. The focus of federal and state efforts should therefore be on developing policies and programs that can ensure such services and supports are efficiently delivered to the right workers at the right time. Over the past few years, Mathematica researchers and others have identified several promising approaches, as well as opportunities to test them. For example, the three states with state-run short-term disability insurance programs—California, New Jersey, and Rhode Island—could pilot-test case coordination and behavioral interventions like those delivered by COHE—in their case outside the workers’ compensation system. Other states could test similar approaches for their own employees, and accountable care organizations could test the use of job retention as a quality metric. These examples just scratch the surface of opportunities.

The second target population for early intervention is transition-age youth with disabilities. Recognizing the potential for improving their outcomes in adulthood, the Social Security Administration launched the Youth Transition Demonstration (YTD) more than a decade ago. YTD adopted an approach that emphasized provision of early work experience based on evidence that such experience would make a substantial difference. Focusing on youth ages 14 to 25, most of whom received SSI as children, YTD projects also provided benefits counseling, career counseling, job development, job placement, and services to support continued employment. Our findings showed that the YTD approach can increase the proportion of youth employed by 7 percentage points as of the third year after enrollment, provided that services are appropriately targeted and delivered as intended.

More recently, Mathematica used data from a randomized evaluation of the Job Corps program, the National Job Corps Study, to understand impacts for the 470 youth who, in a baseline survey, responded “yes” when asked if they had medical conditions that limited their activities. Although Job Corps was originally designed to support economically disadvantaged youth, not those with disabilities, our analysis revealed new information about the program’s impacts for youth with limiting medical conditions. We found positive, large, and significant impacts on employment and earnings over four years. The program also significantly reduced dependence on long-term disability benefits. The employment and earnings impacts in the last two years were at least twice the size of the corresponding impacts for enrollees without limiting medical conditions. Taken together, the YTD and Job Corps findings suggest concrete ways to meet state and national policy goals for improving adult outcomes for youth with disabilities and reducing their reliance on disability benefits in the long term.

Most working-age adults strive to be productive and self-sufficient; it’s built into our DNA. Success is instrumental not just to our economic well-being, but also to how we feel about ourselves and each other—our mental and social well-being. In fact, evidence from surveys that Mathematica has conducted for the Social Security Administration shows that a large share of adult disability program beneficiaries still have such goals. Unfortunately, very few realize them. Efforts that help beneficiaries to be productive and self-sufficient early on—thereby lowering the number of workers and young people who become needlessly dependent on disability benefits in the first place—would go a long way to reduce the number of people facing this frustrating predicament.


The opinions expressed are those of the author(s) and do not represent those of Mathematica Policy Research.

Recent Comments

Join the conversation: You can register for an account to comment on Evidence in Action. Log in to comment through this account or through Facebook, Twitter, LinkedIn, or Google+.

Log in | Register

View the comments policy

Evidence in Action: Comments Policy

We encourage comments on the Evidence in Action blog—all viewpoints are welcome. Commenters can register through our simple form to create an account for Evidence in Action. Commenters can log in through this account or through their social media accounts. Comments are moderated, and we reserve the right to edit, reject, or remove any that include off-topic statements or links; abusive, vulgar, or offensive content; personal attacks; or spam. Those who violate this policy will be blocked from commenting in the future.

Users who log in through a social media account will be identified by information associated with that account (i.e., a Twitter handle or the user name registered with a Facebook, LinkedIn, or Google+ account). Your comment will not include links to your social media account. Mathematica will not post to your social media account.

Feel free to email us with any questions.