As of February 2020, 28 states have sought or received US Centers for Medicare & Medicaid Services approval to impose work requirements as a condition of Medicaid eligibility.1 Centers for Medicare & Medicaid Services requires states to exempt medically frail Medicaid beneficiaries from work requirements2,3 but gives states flexibility in defining such exemptions. After accounting for state definitions of medical frailty exemptions, it is uncertain whether nonexempt beneficiaries not meeting work requirements are disproportionately ill or medically unable to work.4,5 We characterized the proportion of nonexempt beneficiaries not meeting work requirements and determined whether they (1) differ in health from beneficiaries fulfilling them and (2) do not meet the requirements for health-related reasons.
We conducted a cross-sectional analysis using the Agency for Healthcare Research and Quality 2014 and 2015 Medical Expenditure Panel Survey (MEPS) Household Component, an annual nationally representative household survey. We used the demographic, labor, and health-related data in MEPS to model Medicaid work requirement criteria, including medical frailty. As we had insufficient data for individual states, we modeled work requirements nationally across all Medicaid beneficiary respondents. We conducted 4 parallel analyses, varying only the medical frailty exemption criteria used in each analysis. Given the nature of the data set, institutional review board approval was waived. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
For each analysis, we first excluded beneficiaries who are commonly exempted in all proposed state work requirement plans: those 17 years or younger, 65 years or older, pregnant, categorized as disabled on Supplemental Security Income, full-time students, primary caretakers of dependents younger than 6 years, or dually enrolled in Medicare. Next, we excluded medically frail beneficiaries—also exempt—using medical frailty designation methods from one of 4 states with clear, publicly described methods that could be modeled within MEPS: Arkansas, Indiana, Michigan, and New Hampshire (eAppendix in the Supplement). Finally, using weekly work hours reported in MEPS, we classified remaining nonexempt beneficiaries as either fulfilling (working 20 or more hours per week) or not meeting (working less than 20 hours per week) the 20-hour workweek requirement proposed by 90% of states.1
Using logistic regression adjusting for income, we compared the adjusted prevalence of self-reported fair or poor general or mental health among beneficiaries fulfilling and not meeting work requirements (eAppendix in the Supplement). Among beneficiaries not meeting work requirements, we estimated the proportion who self-reported the inability to work owing to illness or disability despite previous employment. We used Stata/SE version 14.0 (StataCorp), weighting all proportions accounting for MEPS survey design.
Among 20 508 respondents representing 59 908 525 Medicaid beneficiaries nationwide in 2014 and 2015, 54 323 166 (75.7%) met the common work requirement exemptions. Depending on the state medical frailty exemption definition used, an additional 1.2% to 2.9% of beneficiaries who did not meet the common work requirement exemptions met state medical frailty exemptions. Of the remaining 21.4% to 23.1% of nonexempt beneficiaries (ie, those subject to work requirements), 10.2% to 10.6% fulfilled work requirements and 11.1% to 12.6% did not meet them (Figure 1).
Among nonexempt beneficiaries, those not meeting work requirements had significantly poorer general and mental health than those fulfilling them (Figure 2A and B). Controlling for income, 20.3% to 26.3% of beneficiaries not meeting work requirements reported fair or poor general health compared with 13.5% to 14.3% of beneficiaries fulfilling work requirements, and 15.5% to 19.6% of those not meeting work requirements reported fair or poor mental health compared with 6.0% to 7.3% of those fulfilling work requirements. Among beneficiaries not meeting work requirements, 15.4% to 19.6% reported an inability to work for health-related reasons (Figure 2C).
Among Medicaid beneficiaries subject to work requirements, those not meeting them were disproportionately sicker than those fulfilling them and often reported health-related barriers to work. Our findings suggest that state medical frailty definitions for Medicaid beneficiaries may incompletely identify medical inability to work.
Our analysis has limitations. The outcome measures, although widely used and valid health and labor status indicators, were self-reported. We analyzed Medicaid beneficiaries nationwide; results may differ state by state.6 Our estimates may be conservative; MEPS lacks some work requirement criteria variables (eg, job training), so we may have overclassified healthy underemployed beneficiaries as not meeting requirements and thereby underestimated fair or poor health in this group of beneficiaries. Similarly, we likely underestimated health-related inability to work, since MEPS limits this question to previously employed respondents. Additionally, our study is cross-sectional and cannot establish causality. These limitations notwithstanding, our findings suggest that Medicaid work requirements may lead some beneficiaries not meeting common work requirement and medical frailty exemptions to lose health insurance coverage because of underlying illness and an inability to work.
Accepted for Publication: March 5, 2020.
Corresponding Author: David M. Silvestri, MD, MBA, MHS, Office of Quality & Safety and Office of Ambulatory Care, New York City Health + Hospitals, 125 Worth St, Ste 427, New York, NY 10013 (david.m.silvestri@gmail.com).
Published Online: May 4, 2020. doi:10.1001/jamainternmed.2020.1039
Author Contributions: Dr Silvestri had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Silvestri, Gluck.
Acquisition, analysis, or interpretation of data: Silvestri, Ross.
Drafting of the manuscript: Silvestri, Gluck.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Silvestri.
Administrative, technical, or material support: Gluck.
Study supervision: Gluck, Ross.
Conflict of Interest Disclosures: Dr Silvestri has received grants from the National Center for Advancing Translational Science. Dr Ross has received grants from the US Food and Drug Administration; Johnson & Johnson; Medical Devices Innovation Consortium; Agency for Healthcare Research and Quality; National Heart, Lung, and Blood Institute; Laura and John Arnold Foundation; US Centers for Medicare & Medicaid Services; Medtronic; and Blue Cross Blue Shield Association. No other disclosures were reported.
Funding/Support: Dr Silvestri completed this work at Yale University while supported in part through Clinical and Translational Science Award grant TL1 TR001864 from the National Center for Advancing Translational Science, a component of the National Institutes of Health.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The contents of this work are solely the responsibility of the authors and do not necessarily reflect the official views of NYC Health + Hospitals, Yale University, or the National Institutes of Health.
3.Exempt Individuals. 42 CFR §440.315(f) (2013).