Utilization of addiction treatment among U.S. adults with history of incarceration and substance use disorders

The National Epidemiologic Survey on Alcohol and Related Conditions-III (NESARC-III) is a cross-sectional survey of a nationally representative sample of the civilian non-institutionalized population of the United States aged 18 years or older. The sample included residents living in a variety of housing settings, including apartments, houses, gated communities, group homes, and workers’ dormitories, but did not include residents in institutions such as jail or prisons, inpatient mental hospitals, and shelters. Data for the NESARC-III was collected between April 2012 to June 2013. Multi-stage probability sampling was employed to select respondents randomly at the county, Census, and household levels. Interviewers conducted in-person structured interviews with respondents to collect information about their personal history, social activities, mental health and substance use disorders, and other health conditions. All interviewers received extensive training on field methods, received ongoing supervision, and conducted random respondent callbacks to verify data. Other details about the methodology of the NESARC-III have been detailed elsewhere [18].

Informed consent was obtained and respondents received $90 for participation. Protocols were approved by the institutional review boards at the National Institutes of Health and Westat; use of the data was approved by the institutional review board at Yale University School of Medicine.

With an overall response rate of 60.1%, the total original sample included 36,309 adult respondents. This study focused on the 10,853 respondents (29.9% of original sample) who were determined to have any lifetime SUD by a structured diagnostic interview and included 2670 (24.6%) who reported a lifetime history of incarceration and 8183 (75.4%) who reported no lifetime history of incarceration. The data were weighted through poststratification analyses to represent the U.S. civilian population based on the 2012 American Community Survey [19].

Personal background information about respondents were collected in various domains, including demographic characteristics, finances, geographic region, military history, immigration status, and health insurance.

Lifetime incarceration was assessed with a question that asked “Since you were 18, were you ever in jail, prison, or a correctional facility?” Respondents who responded “yes” were further asked “about how long altogether were you in jail or a correctional facility since you were 18” by number of days, weeks, months, or years.

Substance use disorders were assessed with the Alcohol Use Disorder and Associated Disabilities Interview Schedule (AUDADIS-5). The AUDADIS-5 is a structured diagnostic interview developed by the National Institute of Alcohol Abuse and Alcoholism and was used to assess alcohol use disorder and specific drug use disorders, according to criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [20]. The AUDADIS-5 has been extensively tested and shown to have good validity and reliability [21‐23]. In this study, we used the AUDADIS-5 to examine lifetime substance use disorder diagnoses.

Lifetime homelessness was assessed with one question that asked: “Since you were 15, did you have a time that lasted at least 1 month when you had no regular place to live- like living on the street or in a car?”

The Interpersonal Support Evaluation List (ISEL) shortened 12-item version was used to measure social support [24]. Respondents were asked to rate a series of statements about emotional support (e.g., feel that there is no one to share worries and fears with) and instrumental support (e.g., would be able to find someone to help with chores if sick) on a 4-point scale from 1 (Definitely false) to 4 (Definitely true). The mean rating of these items was calculated for a social support score.

Substance abuse treatment utilization was assessed with a series of questions about any lifetime use of any self-group treatment (e.g., Alcoholics or Narcotics Anonymous meetings), detox/inpatient treatment (e.g., detoxification clinic or inpatient ward of hospital), outpatient treatment/rehabilitation (e.g., outpatient clinic, day/partial patient program, or rehabilitation program), or emergency room for alcohol and/or drug use problems.

Among the total NESARC sample who responded to the incarceration question (n = 36,121), a weighted 10.6% reported a history of incarceration. Among the 10,853 U.S. adult respondents with a lifetime SUD, 2670 (weighted 22.42%) reported a history of incarceration and a weighted 77.58% reported no history of incarceration. Respondents with a lifetime SUD and history of incarceration reported a mean total of 303.26 days incarcerated (sd = 15.75).

Table 1 shows bivariate differences in background characteristics between respondents with SUDs who did and did not have any history of incarceration. Among notable differences (d > .3 or OR < .5 or OR > 1.5), respondents with SUDs who had a history of incarceration were more likely to be male, to be a veteran, to have lower income, and to have been homeless than those with SUDs and no history of incarceration. Respondents with SUDs and an incarceration history were also more likely to have Medicaid and less likely to have private insurance than respondents with SUDs with no incarceration history.

Table 2 details the rates of lifetime SUD diagnoses between the two groups. In bivariate comparisons, respondents with SUDs and a history of incarceration had much higher rates of almost all SUDs, but especially heroin, inhalants, hallucinogens, and cocaine than those with SUDs and no history of incarceration (all OR > 3.0). Respondents with SUDs and an incarceration history were also more likely to have used any SUD treatment for alcohol and drugs, including self-help group treatment, inpatient treatment, outpatient treatment, and emergency room services than their counterparts with no incarceration history (37% vs. 11% for any alcohol use disorder treatment, 18% vs. 4% for any drug use disorder treatment). Among respondents with alcohol use disorder and no incarceration (n = 7549), 11.5% had used any alcohol use disorder treatment compared to 39.6% who used any alcohol use disorder treatment among respondents with alcohol use disorder and a history of incarceration (n = 2386). Among respondents with drug use disorder and no incarceration (n = 2176), 14.9% had used any drug use disorder treatment compared to 31.4% who had used any drug use disorder treatment among those with drug use disorder and history of incarceration (n = 1340). Among respondents who used any SUD treatment, irrespective of incarceration history, the majority used self-help group treatment for both alcohol and drug use disorders.

Multivariable analyses were then conducted to compare respondents with SUDs with and without history of incarceration including only notable differences in bivariate analyses (d > .3 or OR < .5 or OR > 1.5). As shown in Table 3, respondents with SUDs and a history of incarceration were significantly more likely to be male, to have Medicaid coverage, and to have experienced homelessness compared to those with SUDs and no history of incarceration. Respondents also had lower income and were less likely to have private health insurance. In terms of SUDs and SUD treatment, respondents with SUDs and a history of incarceration were significantly more likely to have tobacco use, cannabis use, cocaine use, and other stimulant use disorders and more likely to have used any alcohol and drug use disorder treatment than respondents with SUDs and no history of incarceration.

To identify characteristics associated with utilization of any SUD treatment, stepwise logistic regressions were conducted on respondents with SUDs, separately for those with and without any history of incarceration (Table 4). Among respondents with SUDs and history of incarceration, having a tobacco use disorder, alcohol use disorder, heroin or other opioid use disorder, cocaine use disorder, or other stimulant use disorder were all strongly associated with any use of SUD treatment (all OR > 1.70). Notably, number of days incarcerated was not strongly associated with utilization of SUD treatment. Among respondents with SUDs and no history of incarceration, having Medicaid coverage, any lifetime homelessness, a tobacco use disorder, alcohol use disorder, heroin use, other opioid use disorder, or cocaine use disorder were all strongly associated with any use of SUD treatment (all OR > 1.70).

Among a nationally representative sample of U.S. adults with SUDs, 22% had been incarcerated before in their lifetime, which was twice the rate found in the total NESARC sample and is consistent with the vast literature documenting the strong links between SUDs, criminal justice involvement, and incarceration [1, 2]. This finding is unique in that it reports on the proportion with incarceration histories among the population of adults with SUDs as the denominator, as most past studies have reported on the proportion of SUDs among the population of adults with incarceration histories [8, 15]. The finding also brings up questions about differences between specific SUDs and use of SUD treatment among adults with incarceration histories in relation to those without such histories. As a result, we posed three additional research questions in this study and discuss our findings below.

One research question was: What proportion of adults with incarceration histories have used SUD treatment? We found that only a minority of adults with SUDs and incarceration histories have used SUD treatment. More specifically, 37% have used any alcohol use disorder treatment and 18% had used any drug use disorder treatment. Most commonly, the type of SUD treatment that was used for both alcohol and drug use problems was self-help groups like Alcoholics Anonymous or Narcotics Anonymous, although there were also relatively high proportions who used outpatient SUD treatment. A Cochrane Review of Twelve-Step programs like Alcoholics Anonymous concluded there was lack of experimental studies demonstrating the effectiveness of these programs [26], but a more recent systematic review found that there is substantial causal evidence of their effectiveness in reducing SUDs and SUD-related outcomes [27]. These self-help SUD treatment groups have become widespread and easily accessible [28]. Perhaps, the effectiveness of these programs may not be due to its specific structure, but because they are freely available, long-term, and easy accessible [29].

A second research question was: Among adults with SUDs, do those with incarceration histories use SUD treatment more than those without incarceration histories? Our findings showed that adults with SUDs and incarceration histories were more likely to use SUD treatment than those with no incarceration histories. Those with incarceration histories had 4.8 times the odds of using alcohol use disorder treatment and 4.7 times the odds of using drug use disorder treatment as compared to those without incarceration histories. Controlling for differences in background characteristics and SUDs, those with incarceration histories still had 3.1 times the odds of using alcohol use disorder treatment and 1.6 times the odds of using drug use disorder treatment. From our data we cannot determine whether the SUD treatment was evidence-based and whether it was accessed during the incarceration period or outside of that time. However, we can say that despite those with SUDs and incarceration histories being more likely to use SUD treatment, the utilization numbers among those with SUDs overall are low, regardless of incarceration history. Thus, the larger implication of our finding is that there continue to be barriers to care for SUD treatment among people with SUDs including those with incarceration histories. This is consistent with various other studies, such as a previous national study that found only one-third of those with SUDs and criminal justice involvement in the past year used SUD treatment in the past year [8]. Some studies have shown that mental health and SUD treatment along with social services can reduce recidivism among people who were formerly incarcerated [30, 31].

Our last research question was: What individual characteristics are associated with utilization of SUD treatment among adults with incarceration histories? Our regression analyses revealed no sociodemographic factors that were strongly associated with use of SUD treatment, but having an opioid disorder or stimulant use disorder was very strongly associated with use of SUD treatment. This was true both for adults with SUDs and incarceration histories as well as those with no incarceration histories and characteristics associated with SUD treatment utilization were largely similar between the two groups. Thus, it seems SUDs involving the “hard drugs” was more associated with SUD treatment than the more prevalent SUDs involving alcohol or cannabis. This finding is entirely consistent with a previous study that examined multiple international epidemiological surveys including the U.S. National Comorbidity Survey and found that cocaine and heroin use significantly predicted SUD treatment-seeking behaviors in the general population [32]. We agree with the study authors’ interpretation that this finding may be due to the possibility that opioids and stimulants are more likely than other substance to lead to impairments or symptoms that promote treatment seeking and often occurs later in the progression of drug use after other “gateway drugs” like alcohol and cannabis. This may also be important in the context of a recent study that found increased criminal justice involvement with increased opioid use [33]. Moreover, there is cause for concern that there is wide variability in quality, type, and intensity of treatment particularly for opioid use disorder. For example, one national study found that only 4.6% of justice-referred clients with opioid use disorder received agonist treatment compared to 40.9% of those referred from some other entity although agonist treatment can be highly effective for opioid use disorder [34]. This has spurred programs like the ones launched by the Rhode Island Department of Corrections [35] and by Rikers Island in New York City [36] to provide medication-assisted treatments for people with opioid use disorder in correctional facilities.

Together, the findings of this study highlight the need for public health interventions to address the high rates of SUDs among U.S. adults who have been involved in the criminal justice system. While we found higher utilization rates of SUD treatment among those with incarceration histories including those who have problems with opioids and stimulants, there is still much opportunity for increasing treatment utilization and ensuring use of evidenced-based practices. Importantly, a focus on prevention of SUDs would reduce the numbers who need SUD treatment and may possible curb criminal justice involvement [37, 38]. At the same time, the high incarceration rate among those with SUDs is not simply due to substance abuse, but a host of social determinants, such as unstable housing, poverty, and social networks; in fact, SUDs and criminal justice involvement share many of these same risk factors [39, 40]. Thus, comprehensive models of care that address these factors should be encouraged and evaluated for their effectiveness at population-based levels.

Our study had several limitations of note. First, this was a large-scale epidemiological examination of associations and lifetime incarceration and SUDs were assessed. The directionality of associations cannot be determined and it is likely many associations we found are bi-directional [41]. For example, we do now know whether the SUD treatment occurred before, during, and/or after incarceration. Second, the data were based on respondent self-report and reports about substance use and histories of incarceration may have been subject to various response biases. Third, the NESARC-III only sampled non-institutionalized adults so adults who are currently incarcerated or hospitalized were not included and so we may have missed an important segment of the population for our study. Fourth, we did not have detailed data on SUD treatment utilization so information about the intensity of SUD treatment services received and length of time were missing, which would be important for future study. These limitations notwithstanding, we believe the study provides important population data on SUDs and SUD treatment in the U.S. and our focus on adults with SUDs and incarceration histories highlights the need for more public health approaches to increase SUD treatment utilization in this population.

The strong association between SUDs and incarceration in the U.S. adult population suggests it is important to increase access and incentives to access SUD treatment during and after incarceration. This epidemiological study provides contemporary data on national rates of SUD treatment utilization among adults with SUDs and incarceration histories. While adults with SUDs and incarceration histories reported greater utilization of SUD treatment, there were generally low SUD treatment utilization rates overall regardless of incarceration history. These results underscore the importance of increasing access to and motivation for SUD treatment, and there may be both common and different barriers for individuals with SUDs depending on incarceration history. Greater need for prevention and intervention for the nexus between criminal justice and substance use problems is needed.