Abstract
Background
Studies have shown that adults with a history of incarceration have elevated cardiovascular (CVD) risk. Research on racial/ethnic group differences in the association between incarceration and CVD risk factors of hypertension and hyperglycemia is limited.
Objective
To assess racial/ethnic group differences in the association between incarceration and hypertension and hyperglycemia.
Design
We performed a secondary data analysis using the National Longitudinal Survey of Adolescent to Adult Health (Add Health). Using modified Poisson regression, we estimated the associations between lifetime history of incarceration reported during early adulthood with hypertension and hyperglycemia outcomes measured in mid-adulthood, including incident diagnosis. We evaluated whether associations varied by self-reported race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, and Asian).
Participants
The analytic sample included 4,015 Add Health respondents who self-identified as non-Hispanic White, Non-Hispanic Black, Hispanic, and Asian, and provided incarceration history and outcome data.
Main Measures
Outcome measures included (1) hypertension (2) systolic blood pressure ≥ 130 mmHg, and (3) hyperglycemia.
Key Results
In non-Hispanic Black and non-Hispanic White participants, there was not evidence of an association between incarceration and measured health outcomes. Among Hispanic participants, incarceration was associated with hyperglycemia (Adjusted Risk Ratio (ARR): 2.1, 95% Confidence Interval (CI): 1.1–3.7), but not with hypertension risk. Incarceration was associated with elevated systolic blood pressure (ARR: 3.1, CI: 1.2–8.5) and hypertension (ARR: 1.7, CI: 1.0–2.8, p = 0.03) among Asian participants, but not with hyperglycemia risk. Incarceration was associated with incident hypertension (ARR 2.5, CI 1.2–5.3) among Asian subgroups.
Conclusions
Our findings add to a growing body of evidence suggesting that incarceration may be linked to chronic disease outcomes. Race/ethnic-specific results, while limited by small sample size, highlight the need for long-term studies on incarceration’s influence among distinct US groups.
KEY WORDS: incarceration, racial/ethnic disparities, hypertension, hyperglycemia
INTRODUCTION
Racial/ethnic disparities in hypertension and type 2 diabetes-related morbidity and mortality are substantial and persistent in the US.1 As compared to White patients, Black individuals experience higher rates of hypertension-related events, such as fatal and nonfatal strokes, cardiovascular disease (CVD), and renal disease.2 Considerable variations regarding hypertension awareness and treatment access perpetuate along racial lines.3 Likewise, the prevalence of diabetes is highest among communities of color, particularly among Asian, Hispanic, and non-Hispanic Black populations, and has only been shown to increase over time.4–6 The root causes of such disparities are complex and interrelated with social determinants of health and known structural barriers.7 Identifying upstream determinants of hypertension and diabetes is crucial, given these are strong risk factors for CVD, the leading cause of death in the US.8
Given that the criminal legal system disproportionately burdens communities of color, we suggest that interfacing with the criminal legal system may be one possible determinant of hypertension and diabetes risk. A variety of mechanisms may contribute towards criminal legal involvement (CLI) facilitating an increased risk of hypertension and hyperglycemia. High levels of obesity and metabolic syndrome exist among those impacted by CLI, both at the individual level as well as among neighborhoods with high rates of incarceration, and likely contribute towards hypertension and diabetes.9,10 Increased stress as a consequence of CLI may lead to hormonal dysregulation and increases in catecholamine levels. Upregulated stress hormones can cause increased blood pressure as well as poor glycemic control.11–13 Several existing challenges experienced by those with a history of CLI, such as food insecurity14 and disruptions in care15, may impede a patient’s ability to adhere to recommended dietary restrictions and medication regimens.
There is evidence of an association between CLI and CVD risk16, including increased risk of uncontrolled hypertension and left ventricular hypertrophy.13,17 While the prevalence of diabetes has been shown to be higher among incarcerated populations than non-institutionalized peers16,18,19, other studies have found no association between a history of CLI and diabetes.13 Despite the disproportionately increased risk of incarceration among communities of color20, there remains a gap in the literature pertaining to disparate health risks linked to CLI along racial/ethnic differences, specifically among Hispanic and Asian populations.
In this study, we seek to expand our understanding of the associations between incarceration and outcomes of hypertension and hyperglycemia with a focus on racial/ethnic group differences. A better understanding of hypertension and hyperglycemia risk among formerly incarcerated groups may have significant implications for both individual health and population-level racial and ethnic disparities.
METHODS
Study Design & Sample Population
We performed a secondary data analysis of the National Longitudinal Study of Adolescent to Adult Health (Add Health). Add Health is an ongoing school-based longitudinal cohort using a nationally representative US sample of 20,000 individuals who were in grades 7–12 in 1994–95, and was designed to study adolescent health-relevant social behaviors and their outcomes into adulthood. To date, the study is composed of 5 total waves: Wave I (1994–1995, grades 7–12, N = 20,745), Wave II (1996, grades 8–12, N = 14,738), Wave III (2001–2002, ages 18–26, N = 15,197), Wave IV (2008–2009, ages 23–32, N = 15,701), and Wave V (2016–2018, ages 33–43, N = 12,300). At each wave of data collection, participants completed a survey assessing sociodemographic characteristics, life experiences, and health behaviors; biological data (e.g., anthropometric, cardiovascular measures) were also collected. Further details of the Add Health study are described elsewhere.21,22 The analytic sample for this study included 4,015 respondents who self-identified as non-Hispanic White, non-Hispanic Black, Hispanic, and Asian and provided incarceration history and Wave V biological data. The majority of participants from Wave V were not able to be included in this analysis due to missing biomarker data.
Measures
Exposure: Lifetime History of Incarceration
Participants were considered to have a history of incarceration if they self-reported having ever spent time in a jail, prison, juvenile detention center or other correctional facility at the Wave IV survey.
Outcomes: Hypertension, Elevated Systolic Blood Pressure (SBP), and Hyperglycemia
Measures of hypertension and hyperglycemia were obtained during Wave V. Participants were classified as having hypertension if they met at least one of the following three criteria: (1) they had an average SBP measurement ≥ 130 mmHg (2) they had an average diastolic blood pressure measurement ≥ 80 mmHg, or (3) they indicated that they had used a prescription antihypertensive medication within the past four weeks. Three blood pressure measurements for each participant were obtained using a sphygmomanometer. An average of the second and third SBP measurements contributed to the final recorded systolic reading, with the same approach applied for diastolic measurements. In addition, we evaluated high SBP, defined as ≥ 130 mmHg, irrespective of diastolic pressure and medication use. SBP was evaluated as an independent outcome given increasing recognition of independent associations between SBP variability and coronary atherosclerosis, cardiac remodeling, stroke, and all-cause mortality.23–25 Hyperglycemia was defined using hemoglobin a1c values derived from respondent blood samples obtained via venipuncture, with values of 5.7% or greater meeting diagnostic criteria. Hyperglycemia was used as an indicator of diabetes risk, combining prediabetic and diabetic range hemoglobin a1c values as a means to optimize power given limitations in sample size.
Covariates
Sociodemographic Characteristics
We measured the following self-reported sociodemographic characteristics: race/ethnicity (reported at Wave I), categorized as non-Hispanic White, non-Hispanic Black, Hispanic, and Asian; age (reported at Wave IV), categorized as 28 and younger, 29–30, and 31 and older; sex (reported at Wave IV), defined as male and female; education level (reported at Wave IV), defined as less than high school, completed high school, and completed college or higher; and socioeconomic status (reported at Wave III), defined using self-reported concern about being unable to pay utility/housing bills in the past 12 months as a proxy for functional poverty. The covariates were chosen based on review of the literature on variables potentially associated with the exposure (incarceration) and/or the outcome (hypertension, elevated SBP, or hyperglycemia). We elected to use pre-incarceration behavioral risk factors measured at Wave III in an attempt to ensure that these factors preceded the incarceration exposure measured at Wave IV. We measured educational attainment at Wave IV because it was closest to the time of exposure measurement and likely better reflective of current education.
Depression
Comorbid depressive symptoms were evaluated at Wave III using a modified version of the Center for Epidemiologic Studies Depression Scale (CES-D) to measure symptoms of depression.26,27 Possible scores ranged from 0 to 27, with a score greater than or equal to 16 indicating a risk for clinical depression.
Substance Use
Self-reported substance use covariates were measured using the Wave III survey. These included affirmative answers to any smoking in the past 30 days, any binge drinking (e.g., ≥ 5 drinks in one sitting) in the past 12 months, any marijuana use in the past 12 months, any cocaine use in the past 12 months, any methamphetamine use in the past 12 months, any prescription opioid misuse (e.g., taken opioids not prescribed to them, taken in larger amounts than prescribed) in the past 12 months, and any other illicit drug use (e.g., LSD, heroin) in the past 12 months.
Analysis
We conducted analyses with Stata version 15 among participants who self-identified as non-Hispanic White, non-Hispanic Black, Hispanic, and Asian and had biological data measured at Wave V. In bivariate analyses, we estimated the prevalence of hypertension and hyperglycemia by participant baseline characteristics. We estimated adjusted risk ratios (ARRs) and 95% confidence intervals (CIs), using modified Poisson regression with robust variance estimation.28,29 Adjusted models controlled for covariates (sociodemographic characteristics, depressive symptoms, and substance use). Models were stratified by racial/ethnic group to identify disparities. All analyses accounted for Add Health’s complex survey design.
We then evaluated longitudinal associations with incident outcomes by excluding individuals with prevalent outcomes of interest by Wave IV. Incident hypertension was measured by excluding respondents who either had a self-reported diagnosis of hypertension or who had systolic ≥ 130 mmHg or diastolic blood pressure measurements ≥ 80 mmHg at Wave IV. For incident hyperglycemia, respondents were excluded from the analysis if the individual either had a self-reported diagnosis of diabetes or had a hemoglobin a1c measurement of ≥ 5.7% at Wave IV.
RESULTS
Respondent Characteristics of the Analytic Sample
Of the analytic sample, 19.9% of respondents were 31 years or older by Wave IV, 34.3% were between ages 29 and 30, and the remaining 45.8% were 28 years of age or younger (Table 1). In regards to sex assigned at birth, 57.8% of respondents identified as female. Among race/ethnicity categories, 75.1% identified as non-Hispanic White, 13.3% as non-Hispanic Black, 9.2% as Hispanic, and 2.4% as Asian. In this sample, 48.0% (N = 1,928) met diagnostic criteria for hypertension, 27.5% (N = 1,106) had a SBP greater than or equal to 130 mmHg, and 11.9% (N = 476) were hyperglycemic. Of non-Hispanic White participants, 11.2% reported a history of incarceration, with rates of 15.2%, 15.0%, and 3.9% across non-Hispanic Black, Hispanic, and Asian respondents, respectively.
Table 1.
Respondent Characteristics in Wave I, Wave III and IV and Corresponding Rates of Hypertension, Systolic Blood Pressure ≥ 130 mmHg, and Hyperglycemia in Wave V
| Variable | N (%) in the Total Sample (N = 4,015) |
N (%) with Hypertension (N = 2,129) |
N (%) with Systolic Blood Pressure ≥ 130 mmHg (N = 1,154) |
N (%) with Hyperglycemia (N = 441) |
|---|---|---|---|---|
| Age: Wave IV | ||||
| < 28 years | 1667 (45.8%) | 792 (48.3%) | 414 (25.9%) | 170 (10.1%) |
| 29–30 years | 1561 (34.3%) | 815 (54.2%) | 448 (31.9%) | 191 (11.8%) |
| > 31 years | 787 (19.9%) | 437 (57.7%) | 243 (32.7%) | 122 (15.0%) |
| Sex: Wave IV | ||||
| Male | 1608 (42.2%) | 1021 (64.1%) | 661 (43.1%) | 195 (11.5%) |
| Female | 2409 (57.8%) | 907 (39.0%) | 445 (19.2%) | 281 (11.3%) |
| Race/Ethnicity: Wave I | ||||
| Non-Hispanic White | 2544 (75.1%) | 1208 (48.6%) | 683 (28.7%) | 205 (8.3%) |
| Non-Hispanic Black | 739 (13.3%) | 423 (59.3%) | 263 (38.0%) | 171 (26.8%) |
| Hispanic | 522 (9.2%) | 217 (48.0%) | 118 (25.6%) | 76 (16.0%) |
| Asian | 212 (2.4%) | 80 (36.3%) | 42 (14.8%) | 24 (8.1%) |
| Education level: Wave IV | ||||
| Less than high school | 200 (5.5%) | 123 (59.0%) | 77 (34.8%) | 33 (19.0%) |
| Completed high school | 475 (12.8%} | 256 (54.1%) | 159 (34.9%) | 84 (18.2%) |
| Completed college | 3342 (81.6%) | 1549 (48.3%) | 870 (28.1%) | 359 (9.8%) |
|
Insufficient Income for Rent, Utilities (Past 6 Months): Wave III |
496 (14.2%) | 271 (55.5%) | 159 (31.4%) | 83 (18.5%) |
| CES-D score indicative of depressive symptoms: Wave III | 394 (10.3%) | 220 (58.6%) | 126 (36.7%) | 56 (15.9%) |
| Substance Use: Wave III | ||||
| Smoking past 30 days | 1063 (32.3%) | 542 (52.0%) | 322 (31.3%) | 109 (8.7%) |
| Binge drinking past 12 months | 1775 (52.6%) | 852 (49.4%) | 499 (30.0%) | 172 (9.8%) |
| Marijuana use past 12 months | 1105 (32.6%) | 530 (47.5%) | 313 (28.2%) | 105 (8.4%) |
| Cocaine use past 12 months | 235 (7.0%) | 128 (51.5%) | 76 (32.2%) | 19 (7.5%) |
| Meth use past 12 months | 83 (2.2%) | 47 (56.2%) | 27 (37.7%) | *cell count < 10 |
| Prescription opioid use past 12 months | 682 (20.2%) | 333 (47.7%) | 195 (29.6%) | 82 (10.3%) |
| Other illicit drug use past 12 months | 349 (10.1%) | 161 (45.7%) | 90 (26.5%) | 28 (6.6%) |
* Cell counts with less than 10 respondents are withheld due to risk of deductive disclosure
Respondent Characteristics by Outcome (Hypertension, SBP > 130 mmHg, and Hyperglycemia
Of those meeting a diagnosis of hypertension, there was a higher prevalence among males as compared to females (64.1% to 39.0%), and higher among non-Hispanic Black groups (59.3%) than all other racial/ethnic respondents.
Males exhibited a higher prevalence of elevated SBP (43.1%) as compared to females (19.2%). An increased prevalence of elevated SBP was evident among those with an education level of completing high school (34.9%) compared to those who completed less than high school (34.8%) and those who completed college (28.1%).
Prevalence of hyperglycemia increased as age group increased, with the highest prevalence (15.0%) seen among adults age 31 years and older. Hyperglycemia was more common among non-Hispanic Black individuals (26.8%) as compared to non-Hispanic White, Hispanic, and Asian groups, with rates of 8.3%, 16.0%, and 8.1% respectively. Hyperglycemia was more prevalent among those with less than high school completed (19.0%) compared to 18.2% among those who completed high school and 9.8% among those who completed college. Respondents with insufficient income saw higher rates of hyperglycemia (18.5%) compared to 10.3% with sufficient income, as did those at risk for clinical depression (15.9% with depressive symptoms compared to 10.9% without).
Incarceration and CVD Risk Factors
Hypertension
Among non-Hispanic White, non-Hispanic Black, and Hispanic participants, there was little evidence of an association between incarceration and hypertension, with ARRs near the null (Table 2). Though cell counts were less than 10, Asian participants showed a significantly increased risk of hypertension among those with a history of incarceration, even after adjusting for key covariates (ARR: 1.7, CI: 1.0–2.8, p = 0.03).
Table 2.
Adjusted Risk Ratios (RRs) and 95% Confidence Intervals (CIs) for the Associations between Incarceration in Wave IV and Hypertension, Systolic Blood Pressure ≥ 130 mmHg, and Hyperglycemia in Wave V among Different Racial/Ethnic Groups
| Non-Hispanic White | Non-Hispanic Black | Hispanic | Asian | |||||
|---|---|---|---|---|---|---|---|---|
| n (%) | Adjusted RR (95% CI) |
n (%) | Adjusted RR (95% CI) |
n (%) | Adjusted RR (95% CI) |
n (%) | Adjusted RR (95% CI) |
|
| HYPERTENSION DIAGNOSTIC CRITERIA* | ||||||||
| No Incarceration | 1121 (49.9%) | Referent | 376 (60.9%) | Referent | 190 (49.0%) | Referent | 72 (39.6%) | Referent |
| Incarceration | 153 (59.2%) | 1.0 (0.9–1.2) | 77 (76.0%) | 1.1 (0.9–1.4) | 41 (56.2%) | 1.0 (0.7–1.5) | 13 (72.2%) | 1.7 (1.0–2.8) |
| SYSTOLIC BLOOD PRESSURE ≥ 130 mmHg | ||||||||
| No Incarceration | 596 (27.6%) | Referent | 213 (35.8%) | Referent | 99 (25.9%) | Referent | 34 (13.5%) | Referent |
| Incarceration | 87 (37.6%) | 1.1 (0.9–1.4) | 50 (51.1%) | 1.2 (0.8–1.7) | 19 (23.7%) | 0.7 (0.3–1.3) | cell count < 10† | 3.1 (1.2–8.5) |
| HYPERGLYCEMIA | ||||||||
| No Incarceration | 187 (8.8%) | Referent | 152 (28.6%) | Referent | 57 (13.6%) | Referent | 20 (8.3%) | Referent |
| Incarceration | 19 (4.6%) | 0.5 (0.3–1.1) | 25 (26.7%) | 0.7 (0.3–1.5) | 19 (28.4%) | 2.1 (1.1–3.7) | cell count < 10† | 0.4 (0.1–3.8) |
* Hypertension Diagnostic Criteria: systolic blood pressure ≥ 130 mmHg, diastolic blood pressure ≥ 80 mmHg, and/or self-reported use of anti-hypertensive medication
† Cell counts with less than 10 respondents are withheld due to risk of deductive disclosure
SBP ≥ 130 mmHg
Non-Hispanic Black respondents with a history of incarceration had a higher prevalence of elevated SBP (51.1%) as compared to those never incarcerated (35.8%). In multivariable models, the ARR for the relationship between incarceration and SBP among non-Hispanic Black participants was 1.2 (CI: 0.8–1.7). Non-Hispanic White respondents saw a similar trend for the relationship between incarceration and elevated SBP risk, with an ARR of 1.1 (CI: 0.9–1.4). Though cell counts were small, among Asian participants, there was evidence that incarceration was associated with elevated SBP (ARR: 3.1, CI: 1.2–8.5). Associations were not observed among Hispanic participants.
Hyperglycemia (Hemoglobin a1c ≥ 5.7%)
Among Hispanic participants, 28.4% with a history of incarceration met diagnostic criteria for hyperglycemia versus 13.6% with no incarceration history. After adjustment, Hispanic participants showed an increased risk for hyperglycemia (ARR: 2.1, CI: 1.1–3.7). Among non-Hispanic White, non-Hispanic Black, and Asian subgroups, differences in incarceration history did not yield major differences in prevalence of hyperglycemia.
Incident Outcomes of Hypertension, SBP and Hyperglycemia
After excluding respondents with either SBP ≥ 130 mmHg, diastolic pressure ≥ 80 mmHg, or self-reported diagnosis of hypertension at time of Wave IV, the Asian subgroup yielded the strongest association between lifetime history of incarceration and incident hypertension at Wave V (ARR 2.5, CI: 1.1–5.3), though the association with incident elevated SBP approached the null (Table 3). Other groups did not show associations among outcomes of incident hypertension or incident elevated SBP. Hispanic subgroups had an increased risk for incident hyperglycemia (ARR, 1.5 CI: 0.1–18.3), though not significant. No relationships were seen among other racial/ethnic subgroups for incident hyperglycemia.
Table 3.
Adjusted Risk Ratios (RRs) and 95% Confidence Intervals (CIs) for the Associations between Incarceration in Wave IV and Incident Hypertension, Incident Systolic Blood Pressure ≥ 130 mmHg, and Incident Hyperglycemia in Wave V among Different Racial/Ethnic Groups
| Non-Hispanic White | Non-Hispanic Black | Hispanic | Asian | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| n (%) | Adjusted RR (95% CI) |
n (%) | Adjusted RR (95% CI) |
n (%) | Adjusted RR (95% CI) |
n (%) | Adjusted RR (95% CI) |
||||
| INCIDENT HYPERTENSION DIAGNOSTIC CRITERIA* | |||||||||||
| No Incarceration | 355 (30.0%) | Referent | 124 (37.7%) | Referent | 59 (23.6%) | Referent | 19 (18.0%) | Referent | |||
| Incarceration | 39 (31.8%) | 1.0 (0.7–1.5) | 27 (62.0%) | 1.3 (0.9–2.1) | 12 (31.1%) | 1.3 (0.4–3.0) | cell count < 10† | 2.5 (1.2–5.3) | |||
| INCIDENT SYSTOLIC BLOOD PRESSURE ≥ 130 mmHg | |||||||||||
| No Incarceration | 135 (11.8%) | Referent | 64 (19.4%) | Referent | 25 (10.4%) | Referent | cell count < 10† | Referent | |||
| Incarceration | 19 (16.6%) | 1.1 (0.6–2.0) | 12 (35.6%) | 1.2 (0.5–3.0) | cell count < 10† | 0.4 (0.1–1.7) | cell count < 10† | 1.1 (0.1–15.1) | |||
| INCIDENT HYPERGLYCEMIA | |||||||||||
| No Incarceration | 61 (3.9%) | Referent | 18 (8.7%) | Referent | 13 (4.9%) | Referent | cell count < 10† | Referent | |||
| Incarceration | cell count < 10† | 0.3 (0.1–1.0) | cell count < 10† | 0.1 (0.0–0.8) | cell count < 10† | 1.5 (0.1–18.3) | cell count < 10† | 0.0 (0.0–0.0) | |||
*Hypertension Diagnostic Criteria: systolic blood pressure ≥ 130 mmHg, diastolic blood pressure ≥ 80 mmHg, and/or self-reported use of anti-hypertensive medication
† Cell counts with less than 10 respondents are withheld due to risk of deductive disclosure
DISCUSSION
To our knowledge, our study is the first to examine how incarceration may influence chronic disease outcomes across multiple major racial/ethnic groups, including Hispanic and Asian groups. In analyses focused on estimating the independent effect of incarceration on outcomes, we observed that incarceration may potentially play a role in elevating cardiometabolic risk in some but not all groups. Among Asian participants, a history of incarceration was independently associated with a threefold increased risk of having a SBP ≥ 130 mmHg than compared with never-incarcerated peers and also was a strong independent correlate of a diagnosis of hypertension. Among Hispanic groups, incarceration was not linked to elevations in hypertension or elevated SBP risk, but there was evidence to suggest incarceration may be associated with hyperglycemia risk.
Thus far, the prevalence of CVD risk factors among currently incarcerated individuals has been well described in the literature, noting a higher degree of diabetes, hypertension, obesity, and tobacco use among those in correctional settings compared with the general population.30–32 Less is understood regarding post-release impacts, with a 2020 systematic review noting an increased incidence of hypertension after release, yet inconclusive findings for other CVD risk factors.9 Further, few studies have evaluated the impacts on different racial/ethnic groups, especially among Hispanic and Asian populations.
Our study is one of the first to examine these relationships in the Asian population. Despite small cell counts, we were able to document a longitudinal association between a history of incarceration and hypertension risk. We acknowledge that there is a significant degree of heterogeneity within who identifies as Asian, and there may be more groups impacted by incarceration in a deleterious manner than we were able to capture. Currently, there is limited data on Asian populations within health services research and there is an urgent need to oversample this population so we can properly reach all groups affected by social determinants of health.33–36
Our findings appear to contradict the existing literature surrounding incarceration and CVD. A prior study by Wang and colleagues found that, within the Coronary Artery Risk Development In Young Adults (CARDIA) study, all individuals with a history of incarceration were more likely to have elevated SBP, hypertension, and incident hypertension compared to never incarcerated peers.13 Yet, upon evaluating racial (White/Black) and sex-based (women/men) differences, results only showed a statistically significant relationship between incarceration and hypertension among Black men, without an increased risk for incident hypertension or elevated SBP. Possible reasons for such differences may be related to the data source itself, as Add Health is a nationally representative sample and CARDIA is a community-based cohort sample. Additional studies should be conducted to further confirm the effects of incarceration on hypertension.
The association between incarceration and hyperglycemia has been poorly characterized. Prior studies on the relationship between incarceration and diabetes have revealed mixed findings. Wang et al. noted no difference in diabetes between formerly incarcerated and never incarcerated individuals.13 Our results note an association between a history of incarceration and the development of hyperglycemia among Hispanic subgroups. This is of public health concern as this population is already disproportionately burdened by diabetes in regards to both prevalence and incidence.37 Future inquiry is needed to better elucidate the relationship between a lifetime history of incarceration and hyperglycemia specific to racial/ethnic group differences.
Our study has several limitations. The Asian population used in our analytic sample had a very small sample size (n = 212), which limits the statistical power of the analysis and our ability to detect true differences, especially in the evaluation of incident diagnoses where we further restricted to those without prior diagnosis. The combination of all hemoglobin a1c values ≥ 5.7% into a single outcome of hyperglycemia, which was done to optimize power, may consolidate a spectrum of disease states into a single variable and thereby limit the interpretation of results. It is possible that at the time of Wave V outcome data collection, participants may have either died or were re-incarcerated, which would lead to differential loss in follow up. If a large percentage of those individuals had a higher incidence of these chronic conditions as compared to participants who remained in the community, this would bias our observed associations. However, retention in Add Health is approximately 70–80% across waves and in non-response analyses performed by Add Health, loss to follow up introduced only minimal bias across results.38 While our analyses included all levels of socioeconomic status in order to optimize generalizability, future analyses should explore the intersections of how incarceration works in tandem with adverse social determinants, such as poverty and other measures of low socioeconomic status. Though we had robust outcome measurement based on biomarker data, models may have been limited by self-reported covariate information via the nature of survey items. Other limitations include underreporting of survey variables.
Despite such limitations, this study yields noteworthy evidence of worsened health disparities among Hispanic and Asian groups affected by incarceration. The criminal legal system is a powerful driving force that perpetuates cycles of poverty and influences health outcomes. Through disrupted social support networks39,40, discriminatory hiring and housing practices41–43, and widespread stigma that impedes access to healthcare and social services44, incarceration exacerbates multiple social determinants of health. The cumulative health impacts of incarceration on communities of color, who are disproportionately represented in the current carceral system, are increasingly being recognized as a public health crisis. Yet, despite this growing awareness, the differential effects on numerous groups are not well understood. The widening of pre-existing inequalities in chronic disease prevalence should enter discussion regarding policy changes to the US criminal legal system, expansion of post-release health services, and updated care guidelines. Further studies are needed to better evaluate the impacts of incarceration across different racial/ethnic groups.
Acknowledgements:
We appreciate the assistance of Arthur Fierman, MD and Mark D Schwartz, MD in editing initial drafts of this manuscript. This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by a grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due to Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining Data Files from Add Health should contact Add Health, The University of North Carolina at Chapel Hill, Carolina Population Center, Carolina Square, Suite 210, 123 W. Franklin Street, Chapel Hill, NC 27516 (addhealth_contracts@unc.edu). No direct support was received from grant P01-HD31921 for this analysis.
Contributors
M.K. and J.S. composed the research aims and M.K., R.E. and J.S. utilized statistical methods to analyze the data. N.I., L.T., and R.E. advised on study design. R.E. prepared the initial draft of the manuscript. All authors reviewed the findings and provided edits and revisions to the manuscript.
Funding
The efforts of LT, NI, JS, and MK are supported in part by the Centers for Disease Control and Prevention (CDC) Grant U48DP006396. NI’s time is partially supported by the Centers for Disease Control and Prevention (grant U48DP001904), the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (grants R01DK110048-01A1 and P30 DK111022R01DK11048), National Heart, Lung, and Blood Institute (grant 1UG3HL151310), National Institute on Minority Health and Health Disparities (grant U54MD00053817), and National Center for Advancing Translational Science (grant UL1TR001445). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The efforts of RE are supported by grant number T32HS026120 from the Agency for Healthcare Research and Quality. The project described was supported by the National Center for Advancing Translational Sciences (NCATS), National Institutes of Health, through Grant Award Number UL1TR001445. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding organizations.
Declarations:
Conflict of Interest:
All authors declare that they have no conflicts of interest.
Footnotes
Prior Presentations
An earlier version of these findings was presented as a poster at the Annual Society of General Internal Medicine Meeting in Orlando, FL April 7th, 2022.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Deere BP, Ferdinand KC. Hypertension and race/ethnicity. Curr Opin Cardiol. 2020;35(4):342-350. [DOI] [PubMed]
- 2.Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, et al. Heart disease and stroke statistics—2014 update: a report from the American Heart Association. Circulation. 2014;129(3):e28–e292. [DOI] [PMC free article] [PubMed]
- 3.Bennett A, Parto P, Krim SR. Hypertension and ethnicity. Curr Opin Cardiol. 2016;31(4):381-6. [DOI] [PubMed]
- 4.McBean AM, Li S, Gilbertson DT, Collins AJ. Differences in diabetes prevalence, incidence, and mortality among the elderly of four racial/ethnic groups: Whites, Blacks, Hispanics, and Asians. Diabetes Care. 2004;27(10):2317-24. [DOI] [PubMed]
- 5.Bancks MP, Kershaw K, Carson AP, Gordon-Larsen P, Schreiner PJ, Carnethon MR. Association of Modifiable Risk Factors in Young Adulthood With Racial Disparity in Incident Type 2 Diabetes During Middle Adulthood. JAMA. 2017;318(24):2457-2465. [DOI] [PMC free article] [PubMed]
- 6.Cheng YJ, Kanaya AM, Araneta MRG, et al. Prevalence of Diabetes by Race and Ethnicity in the United States, 2011-2016. JAMA. 2019;322(24):2389–2398. [DOI] [PMC free article] [PubMed]
- 7.Havranek EP, Mujahid MS, Barr DA, et al. Social Determinants of Risk and Outcomes for Cardiovascular Disease: A Scientific Statement From the American Heart Association. Circulation. 2015;132(9):873-898. [DOI] [PubMed]
- 8.Ahmad FB, Anderson RN. The Leading Causes of Death in the US for 2020. JAMA. 2021;325(18):1829–1830. [DOI] [PMC free article] [PubMed]
- 9.Bondolfi C, Taffe P, Augsburger A, Jaques C, Malebranche M, Clair C, Bodenmann P. Impact of incarceration on cardiovascular disease risk factors: a systematic review and meta-regression on weight and BMI change. BMJ Open. 2020;10(10):e039278. [DOI] [PMC free article] [PubMed]
- 10.Topel ML, Kelli HM, Lewis TT, et al. High neighborhood incarceration rate is associated with cardiometabolic disease in nonincarcerated black individuals. Ann Epidemiol. 2018;28(7):489-492. [DOI] [PMC free article] [PubMed]
- 11.Marcovecchio ML, Chiarelli F. The effects of acute and chronic stress on diabetes control. Sci Signal. 2012;5(247):pt10. [DOI] [PubMed]
- 12.Kyrou I, Tsigos C. Stress hormones: physiological stress and regulation of metabolism. Curr Opin Pharmacol. 2009;9(6):787-793. [DOI] [PubMed]
- 13.Wang EA, Pletcher M, Lin F, et al. Incarceration, incident hypertension, and access to health care: findings from the coronary artery risk development in young adults (CARDIA) study. Arch Intern Med. 2009;169(7):687-693. [DOI] [PMC free article] [PubMed]
- 14.Testa A, Jackson DB. Criminal justice system involvement and food insufficiency: findings from the 2018 New York City Community Health Survey. Ann Epidemiol. 2020;52:42-45. [DOI] [PubMed]
- 15.Jennings L, Branson CF, Maxwell AM, Winkelman TNA, Shlafer RJ. Physicians' perspectives on continuity of care for patients involved in the criminal justice system: A qualitative study. PLoS One. 2021;16(7):e0254578. [DOI] [PMC free article] [PubMed]
- 16.Wildeman C, Wang EA. Mass incarceration, public health, and widening inequality in the USA. Lancet. 2017;389(10077):1464-1474. [DOI] [PubMed]
- 17.Howell BA, Long JB, Edelman EJ, McGinnis KA, Rimland D, Fiellin DA, Justice AC, Wang EA. Incarceration History and Uncontrolled Blood Pressure in a Multi-Site Cohort. J Gen Intern Med. 2016;31(12):1496-1502. [DOI] [PMC free article] [PubMed]
- 18.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] [PubMed]
- 19.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] [PMC free article] [PubMed]
- 20.Nkansah-Amankra S, Agbanu SK, Miller RJ. Disparities in Health, Poverty, Incarceration, and Social Justice among Racial Groups in the United States: A Critical Review of Evidence of Close Links with Neoliberalism. Int J Health Serv. 2013;43(2):217-240. [DOI] [PubMed]
- 21.Carolina Population Center University of North Carolina at Chapel Hill. Study Design. n.d. Available from: http://www.cpc.unc.edu/projects/addhealth/design.
- 22.Harris KM. The Add Health Study: Design and Accomplishments. Available from: http://www.cpc.unc.edu/projects/addhealth/documentation/guides/DesignPaperWIIV.pdf.
- 23.Fatani N, Dixon DL, Van Tassell BW, Fanikos J, Buckley LF. Systolic Blood Pressure Time in Target Range and Cardiovascular Outcomes in Patients With Hypertension. J Am Coll Cardiol. 2021;77(10):1290-1299. [DOI] [PMC free article] [PubMed]
- 24.Nwabuo CC, Yano Y, Moreira HT, et al. Association Between Visit-to-Visit Blood Pressure Variability in Early Adulthood and Myocardial Structure and Function in Later Life. JAMA Cardiol. 2020;5(7):795-801. [DOI] [PMC free article] [PubMed]
- 25.Clark D 3rd, Nicholls SJ, St John J, et al. Visit-to-Visit Blood Pressure Variability, Coronary Atheroma Progression, and Clinical Outcomes. JAMA Cardiol. 2019;4(5):437-443. [DOI] [PMC free article] [PubMed]
- 26.Khan MR, Kaufman JS, Pence BW, et al. Depression, sexually transmitted infection, and sexual risk behavior among young adults in the United States. Arch Pediatr Adolesc Med. 2009;163(7):644-652. [DOI] [PMC free article] [PubMed]
- 27.Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401.
- 28.McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003;157:940–943. [DOI] [PubMed]
- 29.Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702–706. [DOI] [PubMed]
- 30.Agyapong NAF, Annan RA, Apprey C. Prevalence of risk factors of cardiovascular diseases among prisoners: a systematic review. Nutr Food Sci 2017;47:896–906.
- 31.Arries EJ, Maposa S. Cardiovascular risk factors among prisoners: an integrative review. J Forensic Nurs 2013;9:52–64. [DOI] [PubMed]
- 32.Binswanger IA, Merrill JO, Krueger PM, et al. Gender differences in chronic medical, psychiatric, and substance-dependence disorders among jail inmates. Am J Public Health 2010;100:476–82. [DOI] [PMC free article] [PubMed]
- 33.Yi SS, Kwon SC, Suss R, Ðoàn LN, John I, Islam NS, Trinh-Shevrin C. The Mutually Reinforcing Cycle Of Poor Data Quality And Racialized Stereotypes That Shapes Asian American Health. Health Aff (Millwood). 2022;41(2):296-303. [DOI] [PMC free article] [PubMed]
- 34.Yi SS, Kwon SC, Sacks R, Trinh-Shevrin C. Commentary: Persistence and Health-Related Consequences of the Model Minority Stereotype for Asian Americans. Ethn Dis. 2016;26(1):133-8. [DOI] [PMC free article] [PubMed]
- 35.Kanaya AM, Hsing AW, Panapasa SV, et al. Knowledge Gaps, Challenges, and Opportunities in Health and Prevention Research for Asian Americans, Native Hawaiians, and Pacific Islanders: A Report From the 2021 National Institutes of Health Workshop. Ann Intern Med. 2022;175(4):574-589. [DOI] [PMC free article] [PubMed]
- 36.Islam NS, Khan S, Kwon S, Jang D, Ro M, Trinh-Shevrin C. Methodological issues in the collection, analysis, and reporting of granular data in Asian American populations: historical challenges and potential solutions. J Health Care Poor Underserved. 2010;21(4):1354-81. [DOI] [PMC free article] [PubMed]
- 37.Aguayo-Mazzucato C, Diaque P, Hernandez S, Rosas S, Kostic A, Caballero AE. Understanding the growing epidemic of type 2 diabetes in the Hispanic population living in the United States. Diabetes Metab Res Rev. 2019;35(2):e3097. [DOI] [PMC free article] [PubMed]
- 38.Harris KM, Halpern CT, Whitsel EA, et al. Cohort Profile: The National Longitudinal Study of Adolescent to Adult Health (Add Health). Int J Epidemiol. 2019;48(5):1415-1415k. [DOI] [PMC free article] [PubMed]
- 39.Scheidell JD, Kapadia F, Turpin RE, et al. Incarceration, Social Support Networks, and Health among Black Sexual Minority Men and Transgender Women: Evidence from the HPTN 061 Study. Int J Environ Res Public Health. 2022;19(19):12064. [DOI] [PMC free article] [PubMed]
- 40.Volker B, De Cuyper R, Mollenhorst G, Dirkzwager A, van der Laan P, Nieuwbeerta P. Changes in the social networks of prisoners: A comparison of their networks before and after imprisonment. Soc Netw. 2016:47:47-58.
- 41.Myers SL. Employment and crime - an issue of race. Urban league review. 1981:9–24.
- 42.Augustine D, Zatz N, Sugie N. Why Do Employers Discriminate Against People With Records? Stigma and the Case for Ban the Box (p. 9). UCLA Institute for Research on Labor & Employment. 2020. https://irle.ucla.edu/wp-content/uploads/2020/07/Criminal-Records-Final-6.pdf
- 43.Geller A, Curtis MA. A Sort of Homecoming: Incarceration and the housing security of urban men. Soc Sci Res. 2011;40(4):1196-1213. [DOI] [PMC free article] [PubMed]
- 44.Howell BA, Earnshaw VA, Garcia M, Taylor A, Martin K, Fox AD. The Stigma of Criminal Legal Involvement and Health: a Conceptual Framework. J Urban Health. 2022;99(1):92-101. [DOI] [PMC free article] [PubMed]