Abstract
This survey study examines the demographic and health characteristics associated with the employment status of current Medicaid expansion enrollees in Michigan.
The Affordable Care Act (ACA) expanded Medicaid coverage to approximately 11 million working-age adults. Critics have raised concerns about providing Medicaid to adults capable of working. Several states have proposed work requirements in Section 1115 Medicaid waivers. Although none were approved during the Obama administration, the Trump administration is willing to consider such provisions. Prior analyses estimated that half of adults eligible for ACA Medicaid expansion were employed, and 62% of nondisabled adults were working or in school. Although these national estimates of Medicaid-eligible individuals are valuable, less is known about the employment experience of actual enrollees in Medicaid expansion states and which health characteristics may keep them from working. Complementary state-level analyses are needed as individual states consider whether to propose work requirements. This study examined the demographic and health characteristics associated with the employment status of current Medicaid expansion enrollees in Michigan, which expanded Medicaid under a Section 1115 waiver to nonelderly adults with incomes at or below 133% of the federal poverty level who do not otherwise qualify for Medicaid or Medicare based on disability or other criteria.
Methods
The study was deemed to be exempt from approval by the institutional review boards of the University of Michigan and Michigan Department of Health and Human Services. All survey participants provided verbal consent. From January 1 through October 31, 2016, we conducted a computer-assisted telephone survey in English, Arabic, and Spanish of Medicaid expansion (Healthy Michigan Plan) enrollees with at least 12 months of coverage (response rate, 54%). Sampling was stratified by income and Michigan region. Measures included demographic and health characteristics. Multivariable logistic regression analysis incorporating sampling and nonresponse weights were used to examine the association between demographic and health characteristics and the outcomes of being out of work or unable to work vs employed, adjusting for age, sex, race, health status, presence of chronic health conditions, and functional limitations. Analyses were conducted using Stata software (version 14.2; StataCorp). Two-sided P < .05 was considered to be significant.
Results
Among 4090 surveyed Healthy Michigan Plan enrollees (weighted sample, 379 627; 48.4% male and 51.6% female; mean [SD] age, 40.36 [12.96] years), 48.8% reported that they were employed or self-employed; 27.6% were out of work, 11.3% were unable to work, 2.5% were retired, 5.2% were students, and 4.5% were homemakers. Table 1 presents demographic and health characteristics of enrollees out of work or unable to work compared with employed enrollees. In multivariable analyses (Table 2), enrollees were more likely to report being out of work if they were older, male, African American, or in fair or poor health or had mental health conditions or functional limitations. Enrollees were more likely to report being unable to work if they were older, male, or in fair or poor health or had chronic health conditions or functional limitations.
Table 1. Demographic and Health Characteristics for Michigan Medicaid Expansion Enrollees by Employment Status.
| Enrollee Respondent Characteristic | Employment Status, Weighted % (95% CI)a | P Valueb | Total Weighted Sample, % (95% CI) (n = 379 627) |
||
|---|---|---|---|---|---|
| Out of Work (n = 104 534) |
Unable to Work (n = 42 720) |
Employed (n = 185 435) |
|||
| Age, y | |||||
| 19-34 | 34.8 (30.9-38.9) | 14.8 (10.6-20.2) | 45.8 (43.0-48.6) | <.001 | 39.9 (37.9-41.9) |
| 35-50 | 37.7 (33.8-41.8) | 43.1 (37.6-48.8) | 34.2 (31.6-36.8) | 34.0 (32.2-36.0) | |
| 51-64 | 27.5 (24.4-30.8) | 42.1 (36.8-47.5) | 20.0 (18.3-21.9) | 26.1 (24.6-27.6) | |
| Male | 57.2 (53.3-61.1) | 53.9 (48.3-59.4) | 45.5 (42.7-48.3) | <.001 | 48.4 (46.5-50.4) |
| Race | |||||
| White | 55.2 (51.1-59.2) | 70.3 (64.7-75.4) | 62.2 (59.5-64.9) | <.001 | 61.3 (59.4-63.2) |
| Black or African American | 34.4 (30.6-38.5) | 21.9 (17.3-27.3) | 24.2 (21.8-26.8) | 25.9 (24.2-27.7) | |
| Other | 5.9 (4.4-7.9) | 4.3 (2.5-7.3) | 9.4 (7.9-11.2) | 8.8 (7.7-10.0) | |
| >1 Race | 4.4 (3.0-6.5) | 3.6 (2.1-6.1) | 4.1 (3.1-5.5) | 4.0 (3.3-4.9) | |
| Ethnicity | |||||
| Hispanic or Latino | 4.6 (3.1-6.6) | 3.3 (1.8-6.0) | 6.1 (4.9-7.6) | .43 | 5.2 (4.4-6.2) |
| Arab, Chaldean, or Middle Eastern | 2.7 (1.7-4.1) | 1.2 (0.3-4.8) | 7.3 (5.9-9.0) | <.001 | 6.2 (5.3-7.2) |
| Income, % federal poverty level | |||||
| 0%-35% | 79.1 (76.5-81.5) | 73.8 (69.4-77.8) | 33.7 (31.3-36.3) | <.001 | 51.7 (50.7-52.7) |
| 36%-99% | 15.0 (12.9-17.3) | 13.9 (10.9-17.6) | 38.1 (36.1-40.1) | 28.5 (27.6-29.3) | |
| ≥100% | 5.9 (4.7-7.4) | 12.2 (9.6-15.4) | 28.1 (26.5-29.8) | 19.8 (19.2-20.5) | |
| Veteran | 3.9 (2.6-5.8) | 5.9 (3.7-9.2) | 2.3 (1.6-3.3) | .001 | 3.4 (2.7-4.2) |
| Health status | |||||
| Excellent, very good, or good | 66.1 (62.3-69.6) | 26.2 (21.5-31.5) | 80.3 (78.1-82.4) | <.001 | 70.1 (68.4-71.9) |
| Fair or poor | 33.7 (30.1-37.4) | 73.4 (68.1-78.1) | 19.6 (17.5-21.9) | 29.7 (28.0-31.5) | |
| Chronic health condition | 74.0 (69.9-77.6) | 94.0 (90.6-96.2) | 62.3 (59.5-65.0) | <.001 | 69.2 (67.3-71.0) |
| Physical health condition | 65.1 (60.9-69.0) | 87.5 (82.6-91.2) | 53.8 (51.0-56.6) | <.001 | 60.8 (58.8-62.8) |
| Diabetes | 11.4 (9.3-13.9) | 22.3 (17.9-27.4) | 8.8 (7.5-10.4) | <.001 | 10.8 (9.7-12.1) |
| Hypertension | 37.6 (33.8-41.5) | 54.2 (48.5-59.8) | 24.9 (22.7-27.3) | <.001 | 31.3 (29.6-33.1) |
| Cardiovascular disease | 10.4 (8.2-13.2) | 22.9 (18.3-28.2) | 7.1 (5.9-8.6) | <.001 | 9.8 (8.7-11.0) |
| Asthma | 16.1 (13.5-19.1) | 26.6 (21.9-31.9) | 14.7 (12.9-16.6) | <.001 | 17.1 (15.7-18.6) |
| COPD | 11.2 (9.2-13.6) | 23.7 (19.3-28.8) | 7.6 (6.2-9.1) | <.001 | 10.5 (9.5-11.7) |
| Cancer | 2.7 (1.8-4.1) | 10.2 (7.4-14.0) | 2.8 (2.1-3.6) | <.001 | 3.7 (3.2-4.4) |
| Mental health condition | 35.3 (31.7-39.1) | 61.7 (56.1-66.9) | 25.2 (22.9-27.7) | <.001 | 32.2 (30.4-34.0) |
| Mood disorder | 33.7 (30.1-37.4) | 59.6 (54.1-65.0) | 23.5 (21.2-25.9) | <.001 | 30.5 (28.7-32.3) |
| Other | 0.2 (0.0-1.1) | 1.2 (0.5-2.8) | 0.8 (0.4-1.8) | .008 | 0.8 (0.4-1.3) |
| Functional impairment (≥14 of past 30 d) | |||||
| Physical | 24.4 (21.2-27.9) | 68.8 (63.2-73.8) | 13.3 (11.6-15.3) | <.001 | 22.9 (21.3-24.5) |
| Mental | 25.0 (21.7-28.7) | 48.4 (42.7-54.1) | 11.6 (10.1-13.4) | <.001 | 19.9 (18.3-21.5) |
Numbers in column heads indicate weighted numbers for the population.
Generated from χ2 analyses that included all categories of employment, including employed or self-employed, out of work, unable to work, homemaker, student, and retired.
Table 2. Multivariable Logistic Regression Analysis of Association Between Michigan Medicaid Expansion Enrollee Demographic and Health Characteristics and Being Out of Work or Unable to Work.
| Characteristic | Employment Outcomesa | |||
|---|---|---|---|---|
| Out of Work | Unable to Work | |||
| AOR (95% CI) | P Valueb | AOR (95% CI) | P Valueb | |
| Age, y | ||||
| 19-34 | 1 [Reference] | NA | 1 [Reference] | NA |
| 35-50 | 1.29 (0.99-1.67) | .06 | 2.34 (1.45-3.75) | <.001 |
| 51-64 | 1.67 (1.29-2.17) | <.001 | 4.20 (2.64-6.65) | <.001 |
| Male (reference, female) | 1.80 (1.45-2.23) | <.001 | 1.88 (1.35-2.63) | <.001 |
| Race | ||||
| White | 1 [Reference] | NA | 1 [Reference] | NA |
| Black or African American | 1.93 (1.50-2.49) | <.001 | 1.16 (0.76-1.78) | .48 |
| Other | 0.75 (0.50-1.11) | .15 | 0.51 (0.25-1.06) | .07 |
| >1 Race | 1.25 (0.72-2.18) | .42 | 1.02 (0.49-2.15) | .95 |
| Fair or poor health | 1.47 (1.15-1.89) | .003 | 3.52 (2.42-5.11) | <.001 |
| Chronic health condition (reference, none) | ||||
| Physical | 1.11 (0.88-1.42) | .38 | 1.73 (1.08-2.79) | .02 |
| Mental | 1.47 (1.16-1.87) | .001 | 2.61 (1.82-3.73) | <.001 |
| Functional limitation (reference, none) | ||||
| Physical | 1.43 (1.07-1.92) | .02 | 5.10 (3.54-7.33) | <.001 |
| Mental | 1.95 (1.46-2.60) | <.001 | 2.29 (1.56-3.37) | <.001 |
Abbreviation: AOR, adjusted odd ratio.
Each column represents a different multivariable logistic regression model relative to the reference category of employed enrollees.
Generated from multivariable logistic regression incorporating sampling and nonresponse weights and adjusted for age, sex, race, health status, presence of chronic health conditions, and functional limitations.
Discussion
Our findings have key implications for proposed work requirement policies for Medicaid expansion enrollees. First, more than half of Michigan enrollees were already working or students and thus would not be affected by work requirements. Second, most enrollees who were unable to work reported significant barriers to employment, such as poor health, chronic conditions, older age, or functional limitations. Work requirements could disrupt coverage for such vulnerable individuals who may not meet formal criteria for disability. Third, although those who were out of work reported better health and fewer functional limitations, the proportion of Medicaid expansion enrollees overall who were not working and possibly able to work if employment were available remained small. Study limitations include self-reported outcomes, single state data, and lack of information about whether enrollees were looking for work or had other barriers, such as caregiving responsibilities (thus, the proportion of those not working or looking for work may be lower than our estimates). States should consider the administrative costs of implementing a work requirement program to identify and enforce employment for relatively few individuals and the risk of coverage interruptions for vulnerable individuals with chronic health conditions.
References
- 1.Musumeci M. Medicaid and work requirements. Kaiser Family Foundation. http://kff.org/medicaid/issue-brief/medicaid-and-work-requirements. March 23, 2017. Accessed September 29, 2017.
- 2.Garfield R, Rudowitz R, Damico A Understanding the intersection of Medicaid and work. Kaiser Family Foundation. http://kff.org/medicaid/issue-brief/understanding-the-intersection-of-medicaid-and-work. February 15, 2017. Accessed September 29, 2017.
- 3.Price TE, Verma S. Letter to the Nation’s Governors. Washington, DC: US Department of Health and Human Services; 2017. https://www.hhs.gov/sites/default/files/sec-price-admin-verma-ltr.pdf. Accessed September 29, 2017. [Google Scholar]
- 4.Ku L, Brantley E Myths about the Medicaid expansion and the “able-bodied.” Health Affairs Blog. http://healthaffairs.org/blog/2017/03/06/myths-about-the-medicaid-expansion-and-the-able-bodied. March 6, 2017. Accessed September 29, 2017.
- 5.Ayanian JZ. Michigan’s approach to Medicaid expansion and reform. N Engl J Med. 2013;369(19):1773-1775. [DOI] [PubMed] [Google Scholar]