Since the 1980s, the “War on Drugs” has contributed to drastic increases in incarceration in the United States, including a threefold increase in drug-related arrests and an eightfold increase in the prison population [1–3]. Consequently, the US has the highest documented incarceration rate in the world at 716 inmates per 100,000 residents [4], and an astonishing 4,751,400 people on probation or parole [5]. Because law enforcement policies target people who use drugs, 64.5% of people who are incarcerated experience a substance use disorder; however, only 11% receive any type of substance use treatment while incarcerated [6].

People who smoke crack cocaine (PWSC) and people who inject drugs (PWID) are not only at high risk for incarceration, but they are also at high risk for HIV and hepatitis C virus (HCV) infection. Thus, the estimated HIV prevalence is nearly three times higher among incarcerated populations than the general population (1.5 vs. 0.6%) and HCV prevalence is 9–27 times higher (12–35% compared to 1.3%) [7–9].

Access to care and adherence to treatment are key to reducing morbidity and mortality from HIV and HCV. For HIV, antiretroviral therapy (ART) is associated with improved clinical outcomes, longer survival and secondary prevention of infection, including reduced HIV transmission risk at the community level [10–12]. For HCV, the advent of direct acting antiretroviral (DAA) medications, with a relatively short course of treatment and minimal side effects, has led to cure rates of 90%, and the possibility of virtually eliminating HCV transmission. Benefitting from these therapies, however, requires participation in a series of sequential steps, often referred to as the HIV or HCV treatment cascade [13, 14]. These include diagnosis of HIV/HCV, linkage to care, clinical evaluation, treatment initiation, retention in care, and treatment adherence, with the ultimate goal of making viral load undetectable [13–15].

The lives of people using illicit drugs can be chaotic due to stigma, severe poverty, probation/parole requirements, comorbidities, serious mental illness, and the psychological and clinical effects of the substances they ingest, making it difficult to access and adhere to treatment for HIV or HCV in community settings. Incarceration can provide a point of access for HIV/HCV services [16–20], but this does not always translate to successful navigation of the care continuum [21–25]. Community supervision can provide another opportunity to facilitate access to HIV/HCV services, but this opportunity is often not realized [26, 27].

Research has suggested that substance use treatment (SUT) can effectively address the underlying reasons why many people who use drugs become engaged in drug-related offenses [28]. Further evidence suggests that SUT is cost-effective in reducing drug use and its associated health and social costs, as compared to incarceration [29]. SUT has been shown to help people lower their risk of HIV acquisition and transmission, improve their access and adherence to HIV treatment and reduce their viral load [30–33].

In this cross-sectional study of PWSC and PWID, we examine the association of a history of incarceration, community supervision and substance use treatment with access to HIV/HCV testing, components of the HIV treatment cascade, social and physical vulnerability and risk behavior. We additionally assess for predictors of HIV and HCV status.

With data from 2072 PWSC and PWID, people who reported a history of SUT were more likely to report higher levels of HIV and HCV screening, HIV care and ART initiation and were not more likely to report increased vulnerability or risky behavior. Other research has also suggested that people with a history of SUT have improved access to HIV care and treatment [30–33]. Researchers have also suggested that SUT can attend to the root causes of drug-related offenses among drug-involved offenders [28]. Yet as of 2011, 21.6 million people aged 12 or older had a substance use disorder, but only 2.3 (11%) million people received SUT [51]. There remains a great opportunity to invest and expand the use of SUT for drug-involved people with criminal justice histories in order to attend to their substance use disorder [28] and facilitate access to other essential health care such as HIV care and treatment.

In addition, our results suggested that people with a history of criminal justice involvement reported greater access to HIV and HCV testing, but did not report greater access to HIV care or treatment. Researchers and practitioners have urged medical and public health professionals to take advantage of the opportunities provided by the criminal justice system in providing care for a highly underserved population [16, 52, 53]. Both the California state prison and Alameda county jail systems have an opt-out model for HIV testing, and the community supervision system provides testing referrals as indicated by the court or probation/parole officer. While our study did not evaluate these specific policies, it is encouraging that a history of criminal justice involvement was associated with higher reported levels of testing.

Accessing HIV services can quickly become complicated and disjointed for people with a history of criminal justice involvement. People incarcerated for longer sentences might be able to reap the benefits of HIV care and treatment during incarceration, but lose them once released [54]. Further, people who are going in and out of jail for low-level drug use or ‘quality of life’ violations are likely to miss any stabilizing benefit as they are caught in a criminal justice cycle that continually changes their environment. Our findings suggest that people with a history of criminal justice system involvement do not have greater access to HIV care or treatment.

Our findings are also consistent with previous research suggesting that involvement with the criminal justice system is associated with increased social and physical vulnerability [55, 56]. Prior studies have found an increased likelihood of homelessness [57], unemployment [58], lack of educational opportunities [55] and lack of health insurance [59] among people who have been involved in the criminal justice system. Furthermore, our findings showed increased injection- and sexual-related risk behavior among people with a history of criminal justice involvement is consistent with previous research that suggests this dynamic can be driven by factors such as poverty, mental health disorders and concurrent sexual partnerships, that are a by-product of involvement with the criminal justice system [55, 60].

The principal limitation of the study is the observational nature of the research design. Although we adjusted for measured participant characteristics to address concerns of confounding, the potential for unmeasured or mismeasured factors to bias our results existed. In addition, concerns regarding the temporality between exposures and outcomes given the cross-sectional nature of our study exist. We attempted to ameliorate this concern by defining exposure periods prior to outcomes when possible, but this does not convey the same level of rigor as a longitudinal study. Our study had the largest sample and most power for the earlier stage of the HIV care continuum (testing), and included a small number of people living with HIV, which may have impacted the study’s power to detect associations for the following stages of the continuum (receiving HIV care and initiating ART). Furthermore, this study was not designed to look for relationships between different types of substance use treatment for different substance use disorders on study outcomes.

Another limitation is the self-report of behaviors and HCV status for the survey, and as a result, recall and social desirability are potential biases impacting the metrics collected as part for this analysis. Specifically, metrics for vulnerability (e.g., homelessness), sexual risk factors (e.g., multiple partners), injection risk factors (e.g., sharing needles) may be prone to social desirability bias. Due to stigmatization of behaviors and conditions, these types of biases are common when studying people who use drugs [61]. However, the resulting misclassification would be non-differential with regards to our primary exposures of interest and would generally bias our results toward the null [62]. On the other hand, prior research has illustrated that self-reported behaviors from drug users are both reliable and valid for epidemiological purposes [63].

Strengths of the study included the community-recruited sample of study participants. Most studies of HIV care engagement typically use service utilization data, while our sample provides a snapshot of access to services among people recruited from a community setting. Interviewing participants in community settings that are independent of any criminal justice or substance abuse treatment institution may increase participants’ comfort with disclosing stigmatizing behavior compared to participants who are interviewed within institutional settings. In addition, our study included a large sample of criminal justice involved individuals who would not be captured in prison or jail populations, including people who were on probation or parole, who had absconded, or who had warrants out for their arrest. Furthermore, we had data of high quality that were collected specifically for the purposes of research.

Challenges remain in improving access to components of the HIV continuum and reducing vulnerabilities and risk among PWSC and PWID who have a history of the criminal justice system involvement. Innovative models that improve access to the HIV care continuum for these populations is vital. In addition, utilizing SUT programs, as appropriate, that are integrated with other services for HIV, HCV and mental health disorders will be critical to simultaneously address poor access to HIV and HCV services and the underlying reasons people who use drugs engage in drug-related offenses.