Abstract
Objectives. To explore whether and how the Affordable Care Act (ACA) affects the relationship between employment and health insurance coverage, health care utilization, and health outcomes among recently incarcerated men aged 18 to 64 years in the United States.
Methods. With data from the National Survey on Drug Use and Health (NSDUH), we used a difference-in-differences approach to compare changes in outcomes by employment status among recently incarcerated men.
Results. Uninsurance declined significantly among recently incarcerated men after ACA implementation. As the uninsured rate of unemployed men fell below that of their employed counterparts, the ACA helped to fully eliminate the effect of employment on insurance coverage among recently incarcerated men. The employment gap in diabetes widened after ACA implementation as unemployed men saw significant increases in diagnosed diabetes. Employment disparities in hospital visits, diagnosed hypertension, and reported mental illness also declined in the period following ACA implementation, but these changes were not statistically significant.
Conclusions. These findings highlight how the ACA, by providing a new route to health care, reduces the confounding forces associated with employment that are linked to both incarceration and health.
Mass incarceration is a key driver of stratification and inequality in economic outcomes including employment, income, and wealth, and an emerging body of research indicates that it has profound effects on health for formerly incarcerated individuals, their families, and the communities in which they live.1–4 The mechanisms through which incarceration affects health, however, are less well understood. Incarceration is both a correlate and cause of poor health outcomes. High rates of criminal justice contact are concentrated in historically disadvantaged communities and among people with high rates of chronic health conditions, such as mental health and substance use disorders, who have limited access to medical care.5–8 Constitutionally mandated access to care for persons while incarcerated contributes to the diagnoses and treatment of a wide range of communicable and chronic illnesses, theoretically improving health outcomes.9–14 At the same time, high rates of unemployment, low income, and limited access to health insurance are also predictive of comparatively poor health for persons who have experienced incarceration, their families, and the communities in which they live.15–20
Prisons and jails in the United States have become key sites for interventions addressing mental health needs, substance use issues, and routine care for chronic illness.5–7, 20 Correctional facilities are constitutionally mandated to provide minimal health care, and, for some people, prisons and jails have historically been their only sources of medical attention.11–14 Arguments that emphasize the importance of behavior and those that emphasize the structural determinants of health both suggest that low education, unemployment, and low income should be correlated with poor health outcomes among those with criminal justice contact. However, the high correlation between criminal justice contact and other markers of disadvantage, such as low levels of formal schooling, unemployment, and low income, make it hard to disentangle the mechanisms through which having been incarcerated affects health.
The recent enactment of the Affordable Care Act (ACA) provides a new opportunity to examine the pathways through which incarceration affects health. By establishing new pathways to health insurance outside of the labor market, the ACA disentangles access to health care from not only employment but also other social–structural determinants of health, such as education and income, that are likewise historically important determinants of health insurance coverage.21 Existing studies have found that formerly incarcerated individuals saw significant increases in health insurance coverage and a continued reliance on acute care services in the first 2 years following implementation of the ACA.22–26 Whether these trends have continued in more recent years and the extent to which changes in health insurance and acute care usage vary by employment status remains unknown. Researchers also have yet to provide national estimates on how overall and employment patterns in health outcomes among justice-involved populations have changed since the ACA.
To address these gaps in knowledge, we analyzed multiple years of a national, cross-sectional survey to explore whether and how the ACA affects the relationship between employment and (1) health insurance coverage, (2) health care utilization, and (3) chronic health conditions among recently incarcerated men. We evaluated the ACA’s effect on employment by assessing whether preexisting disparities in the outcomes of interest between employed and unemployed men diminished after implementation of the ACA. The purpose of this analysis was to draw attention to how the ACA influences the relationship between incarceration and health by reducing the importance of employment for outcomes including health insurance coverage, health care utilization, and the diagnosis and reporting of chronic health conditions.
METHODS
We used data from the National Survey on Drug Use and Health (NSDUH) to create our analytic data set. Conducted annually by the Substance Abuse and Mental Health Services Administration, the NSDUH is a nationally representative, cross-sectional survey of the noninstitutionalized US population aged 12 years and older. The sampling frame contains residents of households as well as individuals living in noninstitutionalized group quarters and temporary housing units, including halfway houses and homeless shelters.27 We restricted the sample to recently incarcerated men aged 18 to 64 years. Individuals were identified as recently incarcerated if they reported being on probation or parole in the previous 12 months.
The sample consisted of individuals drawn from periods before and after implementation of the ACA’s Medicaid expansion and Health Insurance Marketplace. These policy changes are considered the most comprehensive reforms of the ACA and were expected to make roughly half of the justice-involved population newly eligible for health insurance coverage.24,25,28 Open enrollment for coverage through the Marketplace and expanded Medicaid began in 2013, but the benefits of these programs did not become active until January 1, 2014.29 In view of this unique period and in line with previous research,23 we excluded all 2013 data from our analyses and used the 2009–2012 and 2014–2017 survey years to construct the samples in the pre- and post-ACA implementation periods, respectively.
Measures
Outcome variables.
The outcomes of interest included a range of measures intended to capture health insurance coverage, health care utilization, and chronic health conditions. We evaluated health insurance coverage by using a single binary variable to determine whether an individual was uninsured. We classified an individual as uninsured if that individual reported being without health insurance at the time of the interview. We evaluated health care utilization based on the use of acute care services in hospitals and emergency department (ED) settings. Two binary variables indicated whether respondents reported they (1) stayed overnight or longer as an inpatient in a hospital or (2) were treated in an ED during the past 12 months. We examined chronic health conditions by using another series of binary variables that indicated whether respondents reported having ever been diagnosed with (1) diabetes or (2) hypertension and whether they reported having any type of (3) mental illness or (4) substance use disorder during the past 12 months. NSDUH assessed whether respondents had past-12-month substance use disorders (alcohol use disorders only, illicit drug use disorders only, both alcohol use and illicit drug use disorders, or none) according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria.30 NSDUH assessed whether respondents had any type of mental illness in the past 12 months based on the DSM-IV criteria for having a mental, behavioral, or emotional disorder other than a developmental or substance use disorder.
Key independent variables.
The key independent variables were (1) an indicator of the timing of the survey with respect to the ACA’s launching of the Medicaid expansion and the Health Insurance Marketplace (1 = survey year of 2014, 2015, 2016, or 2017, after the ACA implementation; 0 = survey year of 2009, 2010, 2011, or 2012, before the implementation), (2) respondents’ employment status (treatment group: without full-time employment; and the control group: with full-time employment), and (3) the interaction term between the indicator of the ACA implementation and the employment variable. We determined employment based on how this status confers access to employer-sponsored health insurance. Individuals were identified as having full-time employment if they reported having a job or being enrolled in school or the military on a full-time basis. We refer to the group with full-time employment as employed and to all others as unemployed for the sake of simplicity, but note that the unemployed group includes individuals who reported being employed or enrolled in school or the military on a part-time basis, in addition to those who reported being fully unemployed or not enrolled.
Control variables.
To control for the confounding effects of factors related to the outcomes of interest, our analyses adjusted for a battery of individual health and sociodemographic characteristics. These control variables were self-rated health, age, race/ethnicity, education, household income, and marital status. To help rule out the possibility that unobserved confounders over time could not account for the observed estimates, all analyses also adjusted for year fixed effects.
Statistical Analyses
Similar to recent research investigating differential effects of the ACA,21–23 our statistical analyses followed a difference-in-differences (DID) approach. This approach allowed us to determine if preexisting disparities in the outcomes of interest between employed and unemployed men diminished after implementation of the ACA. The statistical validity of the DID approach relies on the assumption that trends in the outcomes of interest were similar (parallel) for both the treatment (unemployed) and comparison (employed) groups before the policy intervention (ACA implementation). We tested this assumption by assessing interactions in the pre-ACA period between the employment indicator and a year-based time variable in regression models estimated for each outcome of interest. These tests (shown in Table A, available as a supplement to the online version of this article at http://www.ajph.org) confirmed that the trends in the outcomes of interest in the pre-ACA period were not significantly different for the employed and unemployed groups.
Implementing a DID approach to compare pre- to post-ACA disparities, we estimated multivariate logistic regression models for each outcome of interest, adjusting for year fixed effects as well as all health status and sociodemographic measures, and including indicators for ACA implementation, employment, and the interaction between ACA implementation and employment. The interaction term was the DID estimator measuring the ACA’s effect on disparities in the outcomes of interest between employed and unemployed men. We report regression-adjusted means in the pre- and post-ACA periods for both employed and unemployed men, the disparities in these values in both periods, and the pre- to post-ACA changes in these disparities (the DID estimation) for all outcomes based on the predictive margins of the interaction terms, with all other covariates held at their means.
We applied survey weights created by NSDUH analysts to make estimates nationally representative for the noninstitutionalized population, and we adjusted standard errors for the survey’s complex sampling design. We used Stata version 15 SE (StataCorp LP, College Station, TX) to conduct the analyses.
RESULTS
Table 1 presents descriptive statistics of sociodemographic and health characteristics for the sample of recently incarcerated men and shows how these traits vary by employment status. We conducted significance testing for comparisons by using the Wald F test. Compared with employed men, unemployed men were significantly more likely to be older, to be Black, and to have lower levels of formal schooling. Household income and marital status also varied significantly by employment. More than 40% of unemployed men had a household income below 100% of the federal poverty line (as defined by the US Census Bureau in each survey year) while the same was true for roughly 20% of employed men. Nearly 27% of employed men were married compared with 18% of unemployed men. Consistent with their observed sociodemographic disadvantages, unemployed men were also significantly more likely than employed men to report their overall health status as fair or poor. Table B (available as a supplement to the online version of this article at http://www.ajph.org) compares sociodemographic and health characteristics within each group over time and shows that these traits remained overwhelmingly consistent across the study periods.
TABLE 1—
Sociodemographic and Health Characteristics of Recently Incarcerated Men (Aged 18–64 Years): National Survey on Drug Use and Health, United States, 2009–2012 and 2014–2017
| Unweighted No.a (Weighted %b) |
||||
| Characteristic | Overall (n = 7157) | Employed (n = 4140) | Unemployed (n = 3017) | P |
| Age, y | ||||
| 18–25 | 4301 (30.7) | 2514 (31.7) | 1787 (29.2) | .1 |
| 26–34 | 1336 (26.6) | 829 (30.1) | 507 (22.0) | < .001 |
| 35–49 | 1199 (29.0) | 689 (29.5) | 510 (28.4) | .54 |
| 50–64 | 321 (13.7) | 108 (8.7) | 213 (20.4) | < .001 |
| Race/ethnicity | ||||
| White | 3817 (55.4) | 2303 (56.8) | 1514 (53.6) | .09 |
| Black | 1314 (19.6) | 626 (16.4) | 688 (23.8) | < .001 |
| Latino | 1349 (19.6) | 848 (21.4) | 501 (17.2) | .011 |
| Other | 759 (5.3) | 363 (5.4) | 314 (5.4) | .94 |
| Education | ||||
| < High school | 2228 (28.9) | 1110 (24.5) | 1118 (34.8) | < .001 |
| High school/GED | 2812 (37.6) | 1556 (36.4) | 1256 (39.2) | .13 |
| Some college | 1765 (26.2) | 1209 (29.7) | 556 (21.4) | < .001 |
| College graduate | 352 (7.3) | 265 (9.4) | 87 (4.6) | < .001 |
| Household income, % of povertyc | ||||
| < 100 | 2153 (28.6) | 932 (19.5) | 1221 (40.8) | < .001 |
| 100–200 | 2051 (28.4) | 1144 (26.9) | 907 (30.4) | .026 |
| > 200 | 2893 (43.1) | 2012 (53.6) | 881 (28.8) | < .001 |
| Marital status: married | 1138 (23.1) | 745 (26.8) | 393 (18.1) | < .001 |
| Self-rated health status: fair or poor | 935 (16.4) | 358 (9.6) | 577 (25.6) | < .001 |
Note. GED = General Educational Development test. Significance testing conducted with adjusted Wald F test to account for the survey’s weights and complex sampling design.
Reported numbers represent the unweighted sample sizes.
Percentages are weighted to be nationally representative.
As defined by the US Census Bureau in each survey year.
Health Insurance
Table 2 shows that the ACA helped to dramatically reduce uninsurance, and it also, notably, helped to completely eliminate the effect of employment on insurance coverage among recently incarcerated men. Among the overall sample of recently incarcerated men, the adjusted uninsured rate declined by 15.4 percentage points, from 42.8% in the pre-ACA period to 27.4% in the post-ACA period. In the pre-ACA period, uninsurance was significantly higher among unemployed men. Uninsurance decreased by 8.0 percentage points among employed men and by 26.0 percentage points among unemployed men, indicating that unemployed men saw substantially larger benefits in coverage from the ACA. The disparity in health insurance coverage among recently incarcerated men fully disappeared in the post-ACA period as the prevalence of uninsurance among unemployed men fell below that of their employed counterparts (25.1% compared with 29.1%), and the difference between these groups was no longer statistically significant.
TABLE 2—
Adjusted Differences in Health Insurance Coverage Among Recently Incarcerated Men (Aged 18–64 Years): National Survey on Drug Use and Health, United States, 2009–2012 and 2014–2017
| Uninsured | Pre-ACA,a % (95% CI) | Post-ACA,b % (95% CI) | Pre- vs Post-ACA,c Percentage Point Δ (95% CI) |
| Overall | 42.8 | 27.4 | −15.4 (−18.8, −12.2) |
| Employment status | |||
| Employed | 37.1 | 29.1 | −8.0 (−12.6, −3.3) |
| Unemployed | 51.1 | 25.1 | −26.0 (−31.2, −20.6) |
| Employment disparityd | 14.0 (8.0, 19.9) | −4.0 (−9.2, 1.2) | −18.0 (−25.5, −10.4) |
Note. ACA = Affordable Care Act; CI = confidence interval. Results represent adjusted predictive margins. All estimates adjusted for age, race/ethnicity, education, household income, marital status, and self-rated health.
2009–2012.
2014–2017.
Reported number represents the value in the pre-ACA column subtracted from the value in the post-ACA column (post-ACA–pre-ACA).
Reported number represents the value in the employed row subtracted from the value in the unemployed row (unemployed–employed).
Health Care Utilization
Table 3 indicates that patterns of hospital stays and ED visits among recently incarcerated men have partially changed following implementation of the ACA but that employment disparities in the use of these acute care services have remained persistent over time. Hospital stays in the overall sample of recently incarcerated men declined significantly from 10.4% in the pre-ACA period to 7.4% in the post-ACA period. ED visits, by contrast, did not change significantly across periods. Consistent with previous employment inequalities in health insurance coverage, hospital stays and ED visits were significantly higher among unemployed men in the pre-ACA period. Although employment differences in health insurance coverage disappeared in the post-ACA period (Table 2), unemployed men remained significantly more likely to have hospital stays and ED visits, and the employment gaps in use of these acute care services remained similar over time.
TABLE 3—
Adjusted Differences in Health Care Utilization Among Recently Incarcerated Men (Aged 18–64 Years): National Survey on Drug Use and Health, United States, 2009–2012 and 2014–2017
| Pre-ACA,a % (95% CI) | Post-ACA,b % (95% CI) | Pre- vs Post-ACA,c Percentage Point Δ (95% CI) | |
| Hospital stay past 12 mo | |||
| Overall | 10.4 | 7.4 | −3.0 (−5.2, −0.1) |
| Employment status | |||
| Employed | 8.5 | 5.9 | −2.6 (−5.3, 1.0) |
| Unemployed | 13.4 | 10.2 | −3.2 (−6.8, 0.4) |
| Employment disparityd | 4.9 (1.7, 8.1) | 4.3 (1.3, 7.3) | −0.6 (−4.8, 3.8) |
| ED visit past 12 mo | |||
| Overall | 39.3 | 36.2 | −3.1 (−7.0, 0.8) |
| Employment status | |||
| Employed | 36.6 | 32.3 | −4.3 (−9.9, 1.1) |
| Unemployed | 43.1 | 42.1 | −1.0 (−7.2, 5.2) |
| Employment disparityd | 6.5 (1.3, 12.7) | 9.8 (4.1,15.9) | 3.3 (−5.1, 12.1) |
Note. ACA = Affordable Care Act; CI = confidence interval; ED = emergency department. Results represent adjusted predictive margins. All estimates adjusted for age, race/ethnicity, education, household income, marital status, and self-rated health.
2009–2012.
2014–2017.
Reported number represents the value in the pre-ACA column subtracted from the value in the post-ACA column (post-ACA–pre-ACA).
Reported number represents the value in the employed row subtracted from the value in the unemployed row (unemployed–employed).
Chronic Health Conditions
Table 4 presents the ACA’s effects on several chronic health conditions. In the overall sample of recently incarcerated men, there was a significant decline in diagnosed hypertension (from 9.6% in the pre-ACA period to 6.1% in the post-ACA period), but there were no significant changes in diagnosed diabetes, reported mental illness, or reported substance use disorder. Levels of diagnosed hypertension, reported mental illness, and reported substance use disorder were significantly higher among unemployed men in both the pre- and post-ACA periods, and employment disparities in these chronic health conditions remained similar over time. Employment differences in diagnosed diabetes, by contrast, changed significantly following ACA implementation. In the pre-ACA period, there was not a significant difference in diagnosed diabetes between employed and unemployed men (2.5% compared with 2.9%, respectively). The level of diagnosed diabetes in the post-ACA period remained similar among employed men but increased significantly among unemployed men. As unemployed men became significantly more likely to have diagnosed diabetes, implementation of the ACA was associated with a significant widening of the employment gap in diabetes among recently incarcerated men.
TABLE 4—
Adjusted Differences in Chronic Health Conditions Among Recently Incarcerated Men (Aged 18–64 Years): National Survey on Drug Use and Health, United States, 2009–2012 and 2014–2017
| Pre-ACA,a % (95% CI) | Post-ACA,b % (95% CI) | Pre- vs Post-ACA,c Percentage Point Δ (95% CI) | |
| Diabetes | |||
| Overall | 2.6 | 3.3 | 0.7 (−0.6, 1.9) |
| Employment status | |||
| Employed | 2.5 | 2.4 | −0.1 (−1.6, 1.4) |
| Unemployed | 2.9 | 5.2 | 2.3 (0.4, 4.2) |
| Employment disparityd | 0.4 (−2.1, 1.2) | 2.8 (1.1, 4.4) | 2.4 (1.6, 4.6) |
| Hypertension | |||
| Overall | 9.6 | 6.1 | −3.5 (−5.7, −1.1) |
| Employment status | |||
| Employed | 7.8 | 5.1 | −2.7 (−5.2, −0.2) |
| Unemployed | 12.5 | 7.9 | −4.6 (−8.3, −1.1) |
| Employment disparityd | 4.7 (1.4, 8.0) | 2.8 (0.4, 5.1) | −1.9 (−5.9, 2.0) |
| Mental illness | |||
| Overall | 26.2 | 27.8 | 1.6 (−2.0, 5.2) |
| Employment status | |||
| Employed | 21.4 | 23.7 | 2.3 (−2.2, 6.9) |
| Unemployed | 33.6 | 34.0 | 0.4 (−5.1, 5.8) |
| Employment disparityd | 12.2 (6.7, 17.8) | 10.3 (5.1, 15.5) | −1.9 (−9.3, 5.4) |
| Substance use disorder | |||
| Overall | 36.6 | 34.9 | −1.7 (−5.3, 5.4) |
| Employment status | |||
| Employed | 34.8 | 30.7 | −4.1 (−8.5, 2.9) |
| Unemployed | 39.1 | 40.9 | 1.8 (−3.7, 7.1) |
| Employment disparityd | 4.3 (0.9, 9.3) | 10.2 (5.3, 15.0) | 5.9 (−0.9, 12.5) |
Note. ACA = Affordable Care Act; CI = confidence interval. Results represent adjusted predictive margins. All estimates adjusted for age, race/ethnicity, education, household income, marital status, and self-rated health.
2009–2012.
2014–2017.
Reported number represents the value in the pre-ACA column subtracted from the value in the post-ACA column (post-ACA–pre-ACA).
Reported number represents the value in the employed row subtracted from the value in the unemployed row (unemployed–employed).
DISCUSSION
We evaluated the ACA’s effects on the relationship between employment and health insurance coverage, health care utilization, and chronic health conditions among recently incarcerated US men aged 18 to 64 years. Our results showed that the effect of employment on health insurance among recently incarcerated men fully disappeared following ACA implementation. Unemployed men saw their risk of being uninsured cut in half after implementation of the ACA, and they are now less likely to be uninsured than employed men. Despite their improved levels of health insurance coverage, we found that unemployed men remained significantly more likely than employed men to use acute care services in hospital and ED settings. With respect to changes in chronic health conditions, we found that the employment disparity in diagnosed diabetes widened significantly in the period following ACA implementation as unemployed men saw a unique increase in their level of diagnosed diabetes over time. We found no significant changes in employment disparities for diagnosed hypertension, reported mental illness, or reported substance use disorder. Taken together, these findings suggest that employment status no longer determines health insurance coverage among recently incarcerated men in the context of the ACA, but that unemployed men remain disadvantaged in terms of their use of acute care services and chronic health conditions.
Our findings build on recent studies that observed substantial increases in health insurance coverage and persistent use of acute care services among justice-involved populations in the first 2 years following enactment of the ACA.22,23,26 We show that the level of health insurance coverage among recently incarcerated men has continued to increase in the 4 years since ACA implementation (2014–2017) and that improvements in coverage have been largely concentrated among unemployed men. Extending the analysis of previous research,23 we also show that recently incarcerated men have maintained a high level of reliance on acute care in ED settings over time. We contribute new findings to the literature on health care utilization among justice-involved individuals by showing how implementation of the ACA was associated with a significant decline in hospital stays among recently incarcerated men.
We further contribute to work in this area by demonstrating how pre- and post-ACA patterns of hospital stays and ED visits among recently incarcerated men have varied by employment status. Despite the ACA helping to completely eliminate the effect of employment on health insurance coverage, our results show that the ACA has done little to affect employment disparities in the use of acute care services. Coupled together, these findings suggest that improvements in health insurance coverage have not yet fully translated into better use of health care for this population. Whether and how the ACA affects recently incarcerated men’s access to health care more broadly, however, remains unclear. Utilization of health care, including acute care services like those observed in this study, only partially approximate access to care.31 Future work investigating the ACA’s influence on the use of health care among recently incarcerated men should consider how other outcomes that tap into health care access, such as the use of primary care, are also affected.
We add to existing literature that explores how the ACA affects outcomes among justice-involved populations by investigating several chronic health conditions not yet examined in previous work, to our knowledge. Our results suggest that the ACA led to a significant decline in diagnosed hypertension among the overall population of recently incarcerated men and a significant increase in diagnosed diabetes among those who were unemployed. These results do not necessarily reflect actual changes in the prevalence of these chronic health conditions, however, given that our measures estimate the receipt of a diagnosis from a doctor or other medical provider, which depends on access to and use of care. Accordingly, our findings may reflect changes in health care utilization. Future research should investigate whether patterns in the use of health care explain the significant changes in diagnosed hypertension and diabetes that we observed in this study.
The nonsignificant changes we found in the reporting of mental illness and substance use disorder among recently incarcerated men also warrant further investigation. Existing work shows that the receipt of treatment of substance use disorders among justice-involved individuals did not change significantly in the first year following ACA implementation.26 In light of these findings, future studies should explore whether inadequate access to treatment of mental illness and substance use disorder among recently incarcerated men explains their persistently high levels of such chronic health conditions.
Limitations
Although we controlled for available covariates and applied well-established analytic strategies,21–23 it remains possible that there are unobserved factors that could account for some of the changes we observed in our outcomes. For example, our results may be subject to biases attributable to significant variation in the implementation of the ACA over time and across states or may reflect other aspects of the ACA including the dependent care provision. Other historical events, such as spillover effects from the economic recession from 2007 to 2009, might have also influenced the patterns we observed.
Also important to note is that our results reflect an imperfect representation of recently incarcerated men in the United States. The NSDUH data identify people who reported being on probation or parole in the past year. Parole is a period of correctional supervision in the community following a prison term; probation is also a period of supervision in the community and is generally an alternative to prison but often involves time in jail. Our sample of recently incarcerated men therefore excludes individuals released from prison who were not required to serve time on parole and may include people who serve time on probation but did not spend time in jail. In addition, the residential-based sampling frame of the NSDUH excludes individuals who are homeless or unstably housed, so recently incarcerated individuals who experience residential instability may be underrepresented in our study.
Conclusions
By helping to eliminate the effect of employment on health insurance among recently incarcerated men, the ACA generates important implications for our understanding of the relationship between incarceration and health. Previous research has established that the experience of incarceration is negatively associated with a number of health outcomes, although the causal mechanisms linking incarceration to health are a source of ongoing research and debate.11,14,17,18,32 Research has also established that recently incarcerated individuals face a heightened risk of unemployment.3,4,33 Our findings highlight how the ACA reduces the confounding forces associated with employment that have been historically linked to both incarceration and health. Evidence that the ACA has reduced the effect of employment on health insurance status for recently incarcerated men suggests that the ACA enables future research to more clearly identify psychosocial and biological pathways or behavioral mechanisms whereby incarceration may have an impact on health outcomes.32
Existing literature on the health consequences of incarceration commonly emphasize the stress-inducing effects of incarceration and the behavioral impacts of incarceration on health.9,10,14,17 Our research draws attention to how connections to institutions—such as the prison system or the labor market—shape health insurance, health care use, and health conditions in ways that complicate the identification of psychosocial and biological or behavioral mechanisms.31,32 Historical patterns of labor market exclusion coupled with contemporary concentrations of criminal justice contact uniquely affect access to health care and health outcomes of Black Americans.1–4,33–35 Future research, therefore, should acknowledge not only how incarceration affects health but also how access to health care may help reduce racial inequalities in health that may be reflected in, and reinforced by, criminal justice contact and system involvement.
ACKNOWLEDGMENTS
C. M. Gutierrez was supported by the Population Science Training Program grant (T32 HD007168) and the Population Research Infrastructure Program (P2C HD050924) awarded to the Carolina Population Center at the University of North Carolina at Chapel Hill by the Eunice Kennedy Shriver National Institute of Child Health and Human Development. B. Pettit was supported by the Population Research Infrastructure Program awarded to the Population Research Center (P2C HD042849) at the University of Texas at Austin by the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
Note. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
CONFLICTS OF INTEREST
The authors declare that they have no conflict of interest.
HUMAN PARTICIPANT PROTECTION
Our analysis was exempt from human participant review because it fell under the University of North Carolina’s policy for research that uses publicly available data sets with de-identified respondents.
REFERENCES
- 1.Mueller-Smith I. The criminal and labor market impacts of incarceration. 2015. Available at: http://sites.lsa.umich.edu/mgms/wp-content/uploads/sites/283/2015/09/incar.pdf. Accessed April 20, 2019.
- 2.Western B. Punishment and Inequality in America. New York, NY: Russell Sage Foundation; 2006. [Google Scholar]
- 3.Western B, Pettit B. Black–White wage inequality, employment rates, and incarceration. Am J Sociol. 2005;111(2):553–578. [Google Scholar]
- 4.Sykes BL, Maroto MA. A wealth of inequalities: mass incarceration, employment, and racial disparities in US household wealth, 1996 to 2011. RSF. 2016;2(6):129–152. [Google Scholar]
- 5.Bronson J, Berzofsky M. Indicators of mental health problems reported by prisoners and jail inmates, 2011–12. Bureau of Justice Statistics. 2017. Available at: http://www.bjs.gov/content/pub/pdf/imhprpji1112.pdf. Accessed March 12, 2019.
- 6.Bronson J, Stroop J, Zimmer S, Berzofsky M. Drug use, dependence, and abuse among state prisoners and jail inmates, 2007–2009. Bureau of Justice Statistics. 2017. Available at: http://www.bjs.gov/content/pub/pdf/dudaspji0709.pdf. Accessed March 12, 2019.
- 7.Maruschak LM, Berzofsky M, Unangst J. Medical problems of state and federal prisoners and jail inmates, 2011–12. Bureau of Justice Statistics. 2015. Available at: http://www.bjs.gov/content/pub/pdf/mpsfpji1112.pdf. Accessed March 12, 2019.
- 8.Lara-Millán A. Public emergency room overcrowding in the era of mass imprisonment. Am Sociol Rev. 2014;79(5):866–887. [Google Scholar]
- 9.Binswanger IA, Stern MF, Deyo RA et al. Release from prison—a high risk of death for former inmates. N Engl J Med. 2007;356(2):157–165. doi: 10.1056/NEJMsa064115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Binswanger IA, Krueger PM, Steiner JF. Prevalence of chronic medical conditions among jail and prison inmates in the USA compared with the general population. J Epidemiol Community Health. 2009;63(11):912–919. doi: 10.1136/jech.2009.090662. [DOI] [PubMed] [Google Scholar]
- 11.Chang Z, Lichtenstein P, Larsson H, Fazel S. Substance use disorders, psychiatric disorders, and mortality after release from prison: a nationwide longitudinal cohort study. Lancet Psychiatry. 2015;2(5):422–430. doi: 10.1016/S2215-0366(15)00088-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Patterson EJ. Incarcerating death: mortality in US state correctional facilities, 1985–1998. Demography. 2010;47(3):587–607. doi: 10.1353/dem.0.0123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rosen DL, Wohl DA, Schoenbach VJ. All-cause and cause-specific mortality among Black and White North Carolina state prisoners, 1995–2005. Ann Epidemiol. 2011;21(10):719–726. doi: 10.1016/j.annepidem.2011.04.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wang EA, Wang Y, Krumholz HM. A high risk of hospitalization following release from correctional facilities in Medicare beneficiaries: a retrospective matched cohort study, 2002 to 2010. JAMA Intern Med. 2013;173(17):1621–1628. doi: 10.1001/jamainternmed.2013.9008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hatzenbuehler ML, Keyes K, Hamilton A, Uddin M, Galea S. The collateral damage of mass incarceration: risk of psychiatric morbidity among nonincarcerated residents of high-incarceration neighborhoods. Am J Public Health. 2015;105(1):138–143. doi: 10.2105/AJPH.2014.302184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lee H, Wildeman C, Wang EA, Matusko N, Jackson JS. A heavy burden: the cardiovascular health consequences of having a family member incarcerated. Am J Public Health. 2014;104(3):421–427. doi: 10.2105/AJPH.2013.301504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Schnittker J, John A. Enduring stigma: the long-term effects of incarceration on health. J Health Soc Behav. 2007;48(2):115–130. doi: 10.1177/002214650704800202. [DOI] [PubMed] [Google Scholar]
- 18.Schnittker J, Massoglia M, Uggen C. Out and down: incarceration and psychiatric disorders. J Health Soc Behav. 2012;53(4):448–464. doi: 10.1177/0022146512453928. [DOI] [PubMed] [Google Scholar]
- 19.Wildeman C, Mueller C. Mass imprisonment and inequality in health and family life. Annu Rev Law Soc Sci. 2012;8(1):11–30. [Google Scholar]
- 20.Wilper AP, Woolhandler S, Boyd JW et al. The health and health care of US prisoners: results of a nationwide survey. Am J Public Health. 2009;99(4):666–672. doi: 10.2105/AJPH.2008.144279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gutierrez CM. The institutional determinants of health insurance: moving away from labor market, marriage, and family attachments under the ACA. Am Sociol Rev. 2018;83(6):1144–1170. doi: 10.1177/0003122418811112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Winkelman TN, Kieffer EC, Goold SD, Morenoff JD, Cross K, Ayanian JZ. Health insurance trends and access to behavioral healthcare among justice-involved individuals—United States, 2008–2014. J Gen Intern Med. 2016;31(12):1523–1529. doi: 10.1007/s11606-016-3845-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Winkelman TN, Choi H, Davis MM. The Affordable Care Act, insurance coverage, and health care utilization of previously incarcerated young men: 2008–2015. Am J Public Health. 2017;107(5):807–811. doi: 10.2105/AJPH.2017.303703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cuellar AE, Cheema J. As roughly 700,000 prisoners are released annually, about half will gain health coverage and care under federal laws. Health Aff (Millwood) 2012;31(5):931–938. doi: 10.1377/hlthaff.2011.0501. [DOI] [PubMed] [Google Scholar]
- 25.Regenstein M, Rosenbaum S. What the Affordable Care Act means for people with jail stays. Health Aff (Millwood). 2014;33(3):448–454. doi: 10.1377/hlthaff.2013.1119. [DOI] [PubMed] [Google Scholar]
- 26.Saloner B, Bandara SN, McGinty EE, Barry CL. Justice-involved adults with substance use disorders: coverage increased but rates of treatment did not in 2014. Health Aff (Millwood) 2016;35(6):1058–1066. doi: 10.1377/hlthaff.2016.0005. [DOI] [PubMed] [Google Scholar]
- 27.Gfroerer JC, Larson SL, Colliver JD. Drug use patterns and trends in rural communities. J Rural Health. 2007;23(1):10–15. doi: 10.1111/j.1748-0361.2007.00118.x. [DOI] [PubMed] [Google Scholar]
- 28.Garfield R, Licata R, Young K. The uninsured at the starting line: findings from the 2013 Kaiser survey of low-income Americans and the ACA. The Henry J. Kaiser Family Foundation. 2014. Available at: http://www.kff.org/wp-content/uploads/2014/02/8552-the-uninsured-at-the-starting-line7.pdf. Accessed April 20, 2019.
- 29.Wachino V, Artiga S, Rudowitz R. How is the ACA impacting Medicaid enrollment? The Henry J. Kaiser Family Foundation. 2014. Available at: http://www.kff.org/wp-content/uploads/2014/05/8584-how-is-the-aca-impacting-medicaid-enrollment2.pdf. Accessed at April 20, 2019.
- 30.Diagnostic and Statistical Manual of Mental Disorders. Fourth Edition. Washington, DC: American Psychiatric Association; 1994. [Google Scholar]
- 31.Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36(1):1–10. [PubMed] [Google Scholar]
- 32.Esposito MH, Lee H, Hicken MT, Porter LC, Herting JR. The consequences of contact with the criminal justice system for health in the transition to adulthood. Longit Life Course Stud. 2017;8(1):57–74. doi: 10.14301/llcs.v8i1.405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Western B, Sirois C. Racialized re-entry: labor market inequality after incarceration. Soc Forces. 2019;97(4):1517–1542. [Google Scholar]
- 34.Neal DA, Johnson WR. The role of premarket factors in Black–White wage differences. J Polit Econ. 1996;104(5):869–895. [Google Scholar]
- 35.Travis J, Western B, Stephens R. The Growth of Incarceration in the United States: Exploring Causes and Consequences. Washington, DC: National Academy Press; 2014. [Google Scholar]