Examination of Changes in Health Status Among Michigan Medicaid Expansion Enrollees From 2016 to 2017

This survey study examines the self-reported health of enrollees in Michigan’s Medicaid expansion, the Healthy Michigan Plan, among respondents to surveys in 2016 and 2017.


Introduction
Under the Patient Protection and Affordable Care Act (ACA), Medicaid eligibility expanded to provide coverage to a greater portion of the low-income, nonelderly adult population. To date, 36 states and the District of Columbia have expanded their Medicaid programs. 1 Several studies have shown that expanding Medicaid has improved access to primary care and medical homes, 2 increased uptake of preventive services, 3,4 and reduced disparities in insurance coverage across multiple demographic groups, including by race, marital status, and age. 5,6 Racial/ethnic minority groups have experienced the largest gains in coverage as a result of the ACA, especially the Latino/Hispanic population. 7 Research examining the association between having insurance coverage and improved health has been mixed but more recently suggests an association with health improvement. [8][9][10] In studies examining expansions in health insurance under the ACA, reductions in emergency department use and improved self-reported health were observed among enrollees 1 year after expansion, 11,12 and improved survival has also been noted among individuals with end-stage kidney disease. 13 Medicaid expansion in particular may have large health effects, because the Medicaid expansion specifically targets low-income populations that have been historically underinsured or uninsured, have been medically underserved, and have more chronic health conditions. 14 Most state and national studies examining the effect of Medicaid expansion on health have compared health outcomes for low-income persons in expansion and nonexpansion states. These studies have found mixed results, with some findings suggesting reductions in poor health days in expansion states [15][16][17] and other findings suggesting no changes in self-reported health status. 3,4 To our knowledge, no longitudinal studies of the health status of actual Medicaid expansion enrollees to learn whether those who enrolled experience improved health over time have been published to date. Health in the United States is patterned strongly along both socioeconomic and racial/ethnic lines, 18 and Medicaid plays a large role in providing health care coverage to low-income racial/ethnic minority populations that would otherwise be uninsured. 7 Medicaid expansion reaches a broader population of low-income individuals, such as childless adults and those with incomes of 100% to 138% of the federal poverty level (FPL), who have been historically underserved. Given national goals to eliminate health disparities within racial/ethnic minority groups 19  than 675 000 low-income adults (aged 19-64 years) were enrolled. The HMP covers essential health benefits required by the ACA and benefits such as dental and vision care, home health services, and family planning services. Michigan has documented disparities in health by race/ethnicity, income, and geography. 21 The purpose of this study was to examine longitudinal changes in enrollees' selfreported health status and days of poor health over time in racial, ethnic, urban/rural, and very-lowincome subgroups.

Study Design, Data Sources, and Study Population
This study used 2 waves of survey data gathered after HMP implementation in April 2014. The surveys were administered as part of an evaluation of HMP for the Michigan Department of Health and Human Services and the Centers for Medicare & Medicaid Services. As an evaluation of a public program, the University of Michigan and Michigan Department of Health and Human Services institutional review boards deemed the study exempt; therefore, no informed consent was required.
We told all of those who were selected to participate that their participation was voluntary and asked for their verbal permission to continue with the survey. This study followed the American Association for Public Opinion Research (AAPOR) reporting guideline.
An initial telephone survey was administered to HMP enrollees between January 1 and October 31, 2016, 2 years after HMP implementation. The sample included enrollees aged 19 to 64 years with HMP enrollment at least 12 months before sampling and at least 9 months in an HMP managed care plan; preferred language of English, Spanish, or Arabic; and a complete Michigan address and telephone number in the state's enrollment files. We used random sampling stratified by income and geographic region and conducted telephone interviews with enrollees. Additional survey methods have been described elsewhere. 22

Measures
The surveys measured demographic characteristics such as self-reported race/ethnicity and educational attainment, health status, access to and use of health care services, and health risks and behaviors using standard measures from established national surveys. [25][26][27] Other demographic characteristics, including age, geographic region, and enrollment status, were obtained from the state's Medicaid files. Enrollees with at least 1 chronic physical health condition were identified using Medicaid claims in the 24-month period before survey sampling, the 12-month period after survey sampling, and/or self-report in the 2016 or 2017 survey. Health status was assessed by the following survey items: "In general, would you say your health is…(excellent, very good, good, fair, poor [responses fair and poor were grouped together for this analysis])?" "For how many days during the past 30 days was your physical health not good?" "For how many days during the past 30 days was your mental health not good?" and "During the past 30 days, for how many days did poor physical or mental health keep you from doing your usual activities, such as self-care, work, or recreation?"

Discussion
This study is among the first, to our knowledge, to use a longitudinal panel survey of Medicaid expansion enrollees to assess the longer-term association with change in health after Medicaid expansion among historically marginalized groups. In this study, we found improved health for Medicaid expansion enrollees in Michigan from 2016 to 2017, including across all age groups and in populations with a long history of health disparities, such as non-Hispanic black individuals, those with incomes of less than 35% of the FPL, and those residing in the Detroit metropolitan area. Our findings align with those of other work using national data and different study designs (eg, secondary analysis of national data between expansion and nonexpansion states) that have shown improvements in health among racial/ethnic groups and chronic disease and behavioral health subgroups who gained Medicaid coverage. 10,16,29 Our study is unique given the longitudinal panel nature of the design that allowed for assessment of the health status of actual Medicaid expansion enrollees to learn whether those who enrolled experienced improved health over time.
The Detroit metropolitan area has the largest Hispanic and non-Hispanic black population in the state of Michigan and saw its uninsured rate reduced by more than half as a result of the ACA insurance expansions. 30 Improved health status among Detroit-area enrollees may be attributed to the ability to now access health care, coupled with more investments to the health care infrastructure in Detroit as a result of increased reimbursement from a more widely insured population.
Results document improvements in self-reported health status and days of poor physical health across all respondents and in critical subgroups such as those with chronic conditions, those with lower income, and non-Hispanic black respondents. Our findings support previous studies that have found Medicaid expansion to be associated with reductions in poor health days among chronically ill adults. 10,15 Our findings also align with the randomized Oregon Health Insurance study that showed improvements in self-reported health in enrollees in the first 2 years after gaining Medicaid  In contrast to our physical health findings, we did not see significant changes in days of poor mental health or days of usual activities missed. Our findings differ from those of national serial crosssectional studies, which found improvements in poor mental health days and activity limitations in Medicaid expansion states compared with nonexpansion states. 2 The Oregon Health Insurance Study also found reduced depression in the first 2 years after gaining Medicaid coverage. 30 Other research has shown that despite increases in coverage of services for mental health and substance abuse disorders, 33 low rates of treatment remain a concern. 34 Low-income individuals with behavioral health conditions may require high degrees of outreach and engagement to observe consistent improvements in mental health measures. These measures of mental health status may require longer longitudinal assessment or a focus on particular subgroups to observe improvements in self-reported mental health.

Limitations
There are limitations to this study. Measurements of health status relied on individuals' perceptions of their own health. Although reports of improved health from surveys may reflect social desirability or recall bias, self-report is considered a valid and reliable measure for assessing health, and our reports of change over time relied on assessing health status separately for each survey, rather than asking respondents to reflect on changes in health. 35 The potential for bias due to sampling only those who consented to follow-up and also due to loss to follow-up, despite our strong consent and follow-up rates, is another limitation. Our analysis of nonresponse bias indicated little difference in demographic characteristics between those who did and did not complete the follow-up survey and weights adjusted for sampling and nonresponse bias. The subgroup that was no longer enrolled and uninsured was a small category, which may explain the wide CIs in those estimates. We measured changes between 2 short-term points. Longer-term follow-up of Medicaid expansion enrollees will be useful to confirm and extend our findings from this study. Our findings showed improvements in self-reported health among enrollees over time. Causal inference could not be determined from this observational study. Finally, state-specific contextual factors may limit generalizability of our findings to other states.