May 10, 2018
May 10, 2018
The federal government’s Centers for Medicare and Medicaid Services (CMS) is starting to let states take away Medicaid health coverage for otherwise-eligible people who cannot get work, or enough work. The Kasich Administration, at the bidding of the legislature, asked CMS for permission to try it out here in a demonstration program affecting the 700,000 Ohioans who gained health care through the federally-funded Medicaid expansion in 2014.
This issue brief provides an overview of the research CMS cites in a guidance letter to state Medicaid directors to explain this policy change. None of the research they cite provides evidence that removing health insurance will improve health in the long or short run. A number of the studies included in reviews find that insecure jobs – the kind of jobs participants will get – actually harm health. While some of the studies find a not-surprising correlation between unemployment and bad health, the causal relationship is not clear. Some of the studies cited are more than 40 years old; some are based on evidence from other countries or are based on populations that have universal health coverage. In short, these studies do nothing to support the policy they are being used to justify.
The vast majority (73 percent) of Ohio working-age Medicaid enrollees belong to a family with at least one working person in the household. Working-age Medicaid beneficiaries who can work, do. Most of the rest have a disability or are caregivers. The Kasich Administration wants to implement a new condition of eligibility that will require Ohioans enrolled under the Medicaid expansion to prove they have worked 80 hours in a month, or that they are exempt from the requirements. If they don’t do either, the state will take away their Medicaid coverage. The governor’s proposal projects thousands will lose coverage.
Good health is essential in order to work, not a reward for it. When polled by the Ohio Department of Medicaid, over half of Medicaid expansion enrollees said Medicaid made it easier to secure and maintain employment and almost 75 percent of job seekers said Medicaid made it easier to look for work. Under the proposed Medicaid regulations, even Ohioans who work substantial hours could lose health care coverage. Low-wage jobs, in fields like retail, food service, and construction, are often unstable, with fluctuating hours. The Center on Budget and Policy Priorities estimates that 46 percent of low-wage workers could lose Medicaid coverage under a work requirement policy like Kentucky, which has the same 80-hours-per-month requirement Ohio proposes. Even Ohioans meeting the requirements would be forced to jump hurdles to access care.
Ohio’s working-age Medicaid population includes an estimated 60,000 people with disabilities that could limit ability to work. The Ohio proposal does not provide the estimated $370 million needed to implement work requirements. It does not fund a thorough assessment of who is exempt based on disability. Disabled people with limited mobility and resources may struggle to get paperwork proving they qualify for an exemption.
Ohio’s proposed eligibility requirement would apply to all 700,000 in the Medicaid expansion population. While many would be eligible for an exemption, all would need to prove such eligibility. They, too would have to jump through hoops to access health care. Many could lose coverage because they lose paperwork, do not understand requirements, or their documentation gets lost in the system. 
People need to be healthy in order to work. Obviously extremely ill people often cannot work. People with cancer, spinal cord injuries, schizophrenia and serious and chronic illnesses often cannot work. For others, regular health care can prevent or mitigate the effects of illness. The clear causation is that good health care helps people work.
It’s tough to make a case for reversing the cause and effect. An enormous amount of research has been conducted on work, health, employment and unemployment, particularly during the 1980s, when dislocation and plant closure affected huge numbers of workers. More recent studies have examined broader aspects, including geographical and socio-economic factors. This body of research has complex and sometimes contradictory findings, but none that support denying health coverage to otherwise eligible people due to insufficient work hours.
Key statements in the CMS guidance are presented below with a review of contradictory elements embedded in the study or studies.
Review of CMS’s statements and contradictions
1. “While high-quality health care is important for an individual’s health and well-being, there are many other determinants of health. It is widely recognized that education, for example, can lead to improved health by increasing health knowledge and healthy behaviors.”
This statement footnotes a 2002 study entitled “Accumulated labor market disadvantage and limiting long term illness,” published in the International Journal of Epidemiology. It is based on 20 years of data on men in the United Kingdom (1971-1991). Its applicability to Medicaid policy in the United States in 2018 is severely limited because the data is old, the study is based solely on men, and the research was done in the United Kingdom, which has universal health care. Although CMS cited it to support its statement about education, this study does not focus on the role of education, health knowledge nor healthy behavior as outcomes.
The study examines interaction between socio-economic status, unemployment, and disabling illness that prevents a person from working. The authors note that their research could not demonstrate causation, only correlation. They found that both low socio-economic status and periods of unemployment correlate with disabling illness. The authors note that less secure jobs of the type held by working people in low socio-economic status have periods of unemployment, which raises the risk they will develop long term illnesses.
Many with income so low they qualify for Medicaid have jobs that are not secure and provide irregular schedules. For people with low-wage jobs, such as food service, construction, or retail, work hours often fluctuate, leaving them short of the required minimum in some months even as they exceed it in others. Low-wage jobs are also unstable, with frequent job losses that leave people unable to find work in some months. The authors close by acknowledging the challenge of reducing health inequality in the context of long-term economic decline.
2. “One comprehensive review of existing studies found strong evidence that unemployment is generally harmful to health, including higher mortality; poorer general health; poorer mental health; and higher medical consultation and hospital admission rates.
The British Department for Work and Pensions commissioned “Is Work Good For Your Health And Well-Being?” in 2006 as part of their review of Britain’s social insurance and safety net programs. CMS cited it to support the above statement. Some studies in the review were conducted during deindustrialization, when thousands were losing well paid jobs. The economic circumstances were different: the problem then was the immediate loss of well paid jobs. Today’s problem is the pervasiveness of low-wage jobs, with far too few secure jobs with adequate pay and benefits. The review included research from many countries with universal health care, a much different context than in the United States. Authors note that “interventions which simply force claimants off benefits are more likely to harm their health and well-being,” which essentially is what Ohio policymakers are proposing to do if Medicaid expansion participants cannot get enough work. This insight is not discussed in the CMS guidance letter.
A critical summary statement drawn from this review is: “Re-employment of unemployed adults improves various measures of general health and well-being, such as self-esteem, self-rated health, self-satisfaction, physical health, financial concerns.” The findings of these studies are complex; several point out the harm that some jobs can have on mental health. Vinokur et.al reviewed the outcomes of an intervention where people received assistance in getting work. They found the greatest benefit accrued to people who volunteered to participate – not the case under the proposed Medicaid requirements. Kessler et.al found that when participants were reemployed in insecure jobs, their re-employment did not reverse the health-damaging effect of job loss. Ferrie et. al. noted studies that found reemployment in unsatisfactory work or insecure work caused mental health problems. Another equally strong summary statement in the review concluded “After leaving benefits, many claimants go into poorly paid or low-quality jobs and insecure, unstable or un-sustained employment. Many go on to further periods of unemployment or sickness, and further spell(s) on the same or other social security benefits.”
No study cited in the review looks at or supports a policy that takes away health care for people who don’t meet hours of work every month
3. “CMS recognizes that a broad range of social, economic, and behavioral factors can have a major impact on an individual’s health and wellness, and a growing body of evidence suggests that targeting certain health determinants, including productive work and community engagement, may improve health outcomes. For example, higher earnings are positively correlated with longer lifespan.”
To support that statement, CMS cites the study “The association between income and life expectancy in the United States, 2001-2014,” published in the Journal of the American Medical Association in 2016. The study does find that higher income is associated with better health, but raised questions about whether that would be true among poor workers. The authors caution that their findings do not necessarily imply that income has a causal effect on life expectancy: Giving someone more money may not increase their lifespan. Causation could occur in either direction, as people in poor health tend to make less money. Nothing in this study validates cutting of health insurance for these low-income Ohioans.
4. “Another academic analysis found strong evidence for a protective effect of employment on depression and general mental health.”
This statement cites “Health effects of employment: a systemic review of prospective studies,” published in 2014 in the BMJ Journal: Occupational and Environmental Medicine. This study reviews research on employment and health. The review notes that low-quality jobs can lead to reduced health while high-quality jobs can lead to improved health. The authors specifically state that the results may be bi-directional (i.e. healthier people may be more likely to work, rather than the other way around). The study does not claim that the low-quality jobs that work requirement participants may get will improve health.
5. “A 2013 Gallup poll found that unemployed Americans are more than twice as likely as those with full-time jobs to say they currently have or are being treated for depression.”
This statement references findings from a 2014 Gallop survey, entitled “In U.S., Depression Rates Higher for Long-Term Unemployed.” The author notes that psychologists have long associated unemployment with psychological ailments, but states: “The causal direction of the relationship, though, is not clear from Gallup's data. It is possible that unemployment causes poor health conditions such as depression, or it could be that having such conditions makes it harder to land a job. One explanation is that depression makes it harder to find and maintain a job.” The poll did not adjust for differences in income, which could also explain depression. Terminating Medicaid benefits for failing to meet a mandatory work requirement is likely to increase depression without access to non-emergency care or treatment—a concern which CMS did not address in its letter to the states.
6. “Other community engagement activities such as volunteering are also associated with improved health outcomes, and it can lead to paid employment.”
The part of the statement that refers to improved health outcomes is cited to an insurance company survey. The survey does not adjust for income, health status or ability to volunteer so is of little value to the policy. It also cites a 2013 study entitled “Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of Volunteers,” The article starts with the United Nations’ definition of volunteering, described as an act of free will. The Medicaid guidance letter suggests enrollees who cannot get enough hours of work could volunteer hours of time in order to keep their health care coverage. But by definition, this is coerced, not voluntary. The findings of this study are therefore not relevant to the CMS assertion. No study is cited to back up the CMS claim that volunteering will lead to paid employment.
The Kasich Administration has asked the federal government to waive Medicaid rules so they can take away coverage for working age enrollees who cannot get 80 hours per month of work. The proposal – like the Medicaid guidance letter inviting such proposals - points to research to justify this action, stating: “It is widely recognized that poverty, food insecurity, housing, and employment status can impact an individual’s overall health. Indeed, there is a strong connection between improved health and being employed.”
The first study the Ohio proposal cites is “The Social Determinants of Health: It’s Time to Consider the Causes of the Causes” (2014). The authors look at many factors affecting health, from housing to education to public health. When they address work, they focus on how work conditions - manual labor, work schedules, sick leave – can harm health. They do not show that eliminating Medicaid for people unable to get required work hours will improve health. The proposal also cites the “HealthyPeople.gov” website, which has an outline of social determinants of health. There is a search tool, but “work” or “employment” are not included in the list of topic areas. It highlights the Kaiser Family Foundation article: “Beyond Health Care: The Role of Social Determinants in Promoting Health and Health Equity,” which provides an overview of federal and state policy used to improve health outcomes. The article does not include denying Medicaid to those who lack sufficient work hours. It mentions work as it outlines things people need to get work, including public transportation and skills training. The Kasich proposal contains no commitment of funds to help with transit costs or to provide additional workforce development. Instead, the proposal asks CMS for approval to use Medicaid money for such purposes, in clear contradiction of the guidance CMS provided to states.
Ohio’s proposal cites the Ohio Department of Medicaid’s evaluation of the Medicaid expansion, which found access to health coverage made it easier for people to work and easier for job seekers to look for work. Medicaid expansion has made Ohio a healthier, more productive state. Thanks to Governor Kasich’s leadership, 700,000 Ohioans are now covered by Medicaid and have access to mental health and addiction treatment. People who received coverage under Medicaid expansion say it’s now easier to work and look for work. It provides a crucial support to low-income Ohioans who are doing their best to contribute to society and be self-sufficient.
The Trump Administration’s Medicaid officials have cited a body of research to justify denying Medicaid health coverage to otherwise eligible people if they don’t work 80 hours a month. Some studies they reference are old and examine different times and different economic contexts. Some are European and study people who have universal health coverage and don’t live under the threat of losing care. Some studies identify the detrimental health effects of insecure work, which is a condition of many low-wage jobs held by Medicaid beneficiaries. The most closely related finding, from the major review of studies from the United Kingdom’s Work and Pensions Program, warns of poor health outcomes of people cut off from social benefit programs.
Most working age Medicaid enrollees live in families where at least one member works. The vast majority of Medicaid enrollees are working, disabled or caretakers. The question is: should people be denied coverage if they are not able to get 80 hours a week of work, and is cutting off health coverage in months when they cannot get the hours good for their health? Is cutting off health care for disabled people or caretakers who fail to get proper documentation going to be good for their health? No research explains, justifies or even addresses the notion that denying access to health care based on hours worked is a way to improve health. Yet the Trump and Kasich Administrations are moving forward with a proposal to do just that.
The Kasich Administration’s assessment of Ohio’s Medicaid expansion found access to health care made it easier for the majority of working enrollees to keep their job and job seekers to persevere in their search. The current program has helped more than 700,000 Ohioans and has supported working people. It should continue without the harmful new eligibility requirement.
For funding or other assistance that made this work possible, we thank:
The Gund Foundation
The Center on Budget and Policy Priorities
Steve Wagner, Universal Health Care Action Network of Ohio
 Community Engagement State Medicaid Director Letter, Centers for Medicare and Medicaid Services, January 11, 2018 at https://www.medicaid.gov/federal-policy-guidance/downloads/smd18002.pdf
 Kaiser Family Foundation, Intersection of Medicaid and work, KFF https://kaiserf.am/2FKvMv4
 Wendy Patton, Medicaid works, work requirements don’t, Policy Matters Ohio, Feb.6, 2018 at https://bit.ly/2qYtCkX
 Ohio Department of Medicaid, Group VIII Work Requirement and Community Engagement 1115 Demonstration Waiver at https://bit.ly/2p4MC19
 Ohio Department of Medicaid, Ohio Group VIII Assessment: A Report to the general Assembly, 2017 at https://bit.ly/2mfXeXn
 Aviva Aron-Dine, Raheem Chadhrey and Matt Broaddus, Many Working People Could Lose Health Coverage Due to Medicaid Work Requirements, April 11, 2018 at https://bit.ly/2v5yWs6
 An independent analysis by the National Disability Institute in February 2018 using American Community Survey data estimated that as many as 60,000 Ohioans who are Non-SSI Disabled and currently rely on Medicaid would be at risk of losing their coverage. Nanette Goodman, Sr. Researcher, National Disability Institute, Analysis of the 2016 American Community Survey (ACS) One-Year Public Use Microdata Sample (PUMS). (February 20, 2018)
 Loren Anthes, Medicaid Work Requirements Budget Neutrality Analysis, March 13, 2018 at https://www.communitysolutions.com/research/medicaid-work-requirement-waiver-analysis-budget-neutrality/
 Kaiser Family Foundation, Implications of Emerging Waivers on Streamlined Medicaid Enrollment and Renewal Processes, February 28, 2018 at https://kaiserf.am/2HXNvAJ
 Mel Bartley and Ian Plewis, Accumulated labour market disadvantage and limiting long-term illness: data from the 1971–1991 Office for National Statistics' Longitudinal Study, International Journal of Epidemiology, Volume 31, Issue 2, 1 April 2002, Pages 336–341, at https://academic.oup.com/ije/article/31/2/336/617702
 Two longitudinal surveys from the Bureau of Labor Statistics that follow workers show that over time, workers with lower levels of educational attainment spend more weeks unemployed and have more short job spells. See: https://www.bls.gov/news.release/nlsoy.t03.htm, https://www.bls.gov/news.release/nlsyth.t05.htm, and https://www.bls.gov/news.release/nlsyth.t03.htm.See Bureau of Labor Statistics, “Job Openings and Labor Turnover, January 2017,” https://www.bls.gov/news.release/archives/jolts_03162017.htm; Arindrajit Dube, William Lester, and Michael Reich, “Minimum Wage Shocks, Employment Flows and Labor Market Frictions,” University of California Berkeley Institute for Research on Labor and Employment, October 2014, http://irle.berkeley.edu/files/2013/Minimum-Wage-Shocks-Employment-Flows-and-Labor-Market-Frictions.pdf; Susan Lambert, Anna Haley-Lock, and Julia Henly, “Schedule flexibility in hourly jobs: unanticipated consequences and promising directions,” Community, Work, and Family, Vol. 15, No. 3, August 2012, https://ssascholars.uchicago.edu/sites/default/files/work-scheduling-study/files/136688032e20122e662803_0.pdf; David Weil. “Enforcing Labour Standards in Fissured Workplaces: the U.S. Experience,” Economic and Labour Relations Review, Vol. 22, No. 2, https://bit.ly/1EUGkAD; Lawrence Katz and Alan S Kruger, the rise and nature of alternative work arrangements in the united states, 1995-2015, September 2016, http://www.nber.org/papers/w22667.pdf. On working conditions faced by enrollees subject to work requirements in other programs, see Jeffrey Grogger and Lynn A. Karoly, Welfare Reform: Effects of a Decade of Change, Harvard University Press, 2005. Cited in Stacy Dean, Ed Bolen and Byrnne Keith-Jennings, “Making SNAP work requirements harsher will not improve outcomes for low-income people,” Center on Budget and Policy Priorities, March 1, 2018 at https://www.cbpp.org/research/food-assistance/making-snap-work-requirements-harsher-will-not-improve-outcomes-for-low
Aviva Aron-Dine et.al., Op.Cit.
G. Waddell, and A.K. Burton, Is Work Good For Your Health And Well-Being? (2006) EurErg Centre for Health and Social Care Research, University of Huddersfield, UK
 Pavetti, Op.Cit.
 R.D. Caplan, A.D. Vinokur, R.H. Price, M. van Ryn, “Job seeking, re-employment and mental health: a randomized field experiment in coping with job loss.” Journal of applied psychology 1989, 74:759-769.; R.M. Cohn, The effect of employment status change on self-attitudes. Social Psychology, 1978, 41:81-93; J.E. Ferrie, P. Martikainen, M.J. Shipley, S.A. Stanfeld, M.G. Marmot and G.D. Smith, Employment status and health after privatization in white color civil servants: prospective cohort study.2001, BMJ 322:647-651; R.C. Kessler, J. Blake Turner, J.S. House. Unemployment, reemployment and emotional functioning in a community sample, 1989, American Sociological Review, 54:648-657; R. Payne and J.G. Jones, Social class and reemployment: changes in health and perceived financial circumstances, 1987, Journal of Occupational Behavior 8:175-184 and A. Vinocour and R.D. Caplan, Attitudes and social support: determinants of job seeking behavior and well-being among the unemployed, 1987 , Journal of Applied Social Psychology, 17:1007-1024.
 J. Bowling, N. Coleman, J. Wapshott, H. Carpenter. Destination of benefit leavers (in-house report 132) 2004. Department for work and pensions, London; J. Bacon, Moving between sickness and unemployment, 2002. Labor Market Trends 110:195-205, K. Ashworth, Y. Hartfree and A. Stephenson, Well enough to work? DSS Research Report #145, Corporate Document Services, Leeds; W.W. Daniel, How the unemployed Fare after they find new jobs, 1983 Policy Studies 3:246-260., A. Hedges and W. Sykes, Moving between sickness and health, DWP Research Report No. 151, 2001 Corporate document services, Leeds; and P. Juvonen-Posti, T. Kallanrants, S-L Eksyma, K. Piiraubeb and S. Kuikaanniemi-Keinanen, Into work through tailored paths: A two-year follow-up of the return-to-work rehabilitation and re-employment project, International Journal of Rehabilitation Research, 2002, 25: 313-330.
 Raj Chetty; Michael Stepner; David Cutler, The Association Between Income and Life Expectancy in the United States, 2001-2014, Journal of the American Medical Association (JAMA), April 26, 2016 at https://bit.ly/2JS5kAN
 Dr. LaDonna Pavetti, in comments to CMS from the National Health Law Project, “HHS’s Invitation to Impose Punitive Work Requirements Lacks Support, January 11, 2018 at https://bit.ly/2w9b0Es
 M. Van der Noordt, H. Jzelenberg,, M. Droomers, and K. Proper, Health effects of employment: a systemic review of prospective studies. BMJournals. Occupational and Environmental Medicine. 2014: 71 (10). https://www.ncbi.nlm.nih.gov/pubmed/24556535
 S. Crabtree.. In U.S., Depression Rates Higher for Long-Term Unemployed. (2014). Gallup. https://bit.ly/2JVSgur
 Pavetti, Op.Cit.
 United Health Group. Doing good is good for you. 2013 Health and Volunteering Study. https://bit.ly/18AXZzw
 Jenkins, C. Dickens, A. Jones, K. Thompson-Coon, J. Taylor, R. and Rogers,M. .Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers BMC Public Health 2013. 13 (773)
 Ohio Department of Medicaid, Public Notice: Group VIII Work Requirement and Community Engagement 1115 Demonstration Waiver
 “The Social Determinants of Health: It’s Time to Consider the Causes of the Causes.” Paula Braveman, MD, MPH and Laura Gottlieb, MD, MPH, “Public Health Rep. 2014 Jan-Feb 129 (suppl2): 19-31
 “Social Determinants of Health” available at healthypeople.gov;
 Group VIII Assessment, Op. Cit.
1 of 22