Prevalence of Chronic Health Conditions Among Adults Released From the North Carolina Prison System, 2015-2016

Discussion

Despite the growing body of academic literature focused on the health of people in the criminal justice system [19], there is a dearth of clinically informed epidemiologic data to enumerate the chronic health problems of incarcerated people returning to the community. The lack of these data can inhibit planning by Transitions Clinics and other groups providing health and social services to released people.

In our one-year study, we found that nearly 7,000—or one in three—people returning from prison had a chronic health condition. The prevalence of health problems increased with age, with greater than 70% of those aged 55 years or older having at least one chronic condition. These estimates reflect a refinement of national estimates based on self-reports and underscore the significant numbers of people who are returning to the community each year and require continuity of care and regular medication to address their chronic health needs.

In addition to the large overall number of returning prisoners with health problems and the high prevalence of health conditions among older people, we found that among the 15% with psychiatric conditions, more than half had a concurrent non-psychiatric chronic condition. This finding is particularly concerning given that people with severe mental illness often receive inadequate care for their comorbid conditions, which contributes to the shortening of their life expectancies by an average of 13 to 30 years [20].

Similar to the observed prevalence of mental health problems, we found that 15% of our study population had cardiovascular disease. Although the risk of cardiovascular morbidity and mortality among former prisoners has received less attention than that of other causes, existing studies suggest that (as in the general population), cardiovascular disease is among the leading causes of death for people released from prison [10, 11]. For example, a study of incarcerated people released from Washington state prisons between 1999 and 2009 [21] found that following overdose, cardiovascular disease was the second leading cause of death (13%), and in a North Carolina study, the risk of death from heart disease among formerly incarcerated people was 4.5 times that of the general population [22]. Despite these findings, less than 30% (N = 14) of state prison systems reportedly target cardiovascular disease for continuity of care, and the North Carolina prison system has been among those not supporting this continuity [12].

Although we used prescription data as a surrogate to estimate disease prevalence, in some instances low prevalence estimates simply reflect a lack of screening or treatment. The two most prominent examples are hepatitis C (HCV) and opioid use disorder (OUD). The low prevalence of HCV we observed is likely a result of low testing rates in prison. HCV testing is uncommon across many systems because of the high price of therapies, despite their effectiveness. Nationally the prevalence of HCV in the prison population has been estimated to be 18%, and we have found a similar prevalence among a small sample of people released from the North Carolina prison system (20%, N = 74) [23]. Similarly, the absence of prescription records corresponding to medication assisted treatment (MAT) for OUD is a reflection of treatment practices and not the underlying prevalence. The need for MAT in prison and at release was highlighted by a recent North Carolina study which found that in the first two weeks following prison release, there was a 74 and 40 times greater risk of death from heroin and opioid overdose, respectively, as compared to the risks among those in the general population [24]. Although the need for OUD treatment during and following incarceration has gained significant attention in North Carolina and across the country, like other chronic health conditions, the health care resources for people released from prison are inadequate.

In response to existing deficits in continuity of care, we are collaborating with the state prison system, county health departments, and federally qualified health centers to implement the TCN model at multiple sites across the state to foster greater continuity of care for people released from prison. The program, named the Formerly Incarcerated Transition (FIT) Program and led by coauthor Ashkin, has been initiated in four counties in North Carolina, three of which are located in Central North Carolina and a fourth in Mecklenburg County. Prompted in part by our findings around comorbidities among those with psychiatric conditions, we are initiating one of the FIT Program sites at an existing primary care clinic that is co-located within a behavioral health campus for people with serious mental illness (SMI). To our knowledge, this is the first time that a TCN clinic will be specifically focused on serving the health needs of those with SMI. Considering that prisons and jails have become the de facto safety net provider for people with SMI, creating and evaluating TCN models for those with SMI is a high public health priority.

At their full capacity, we anticipate that these four clinics will allow us to work with approximately 250 clients at one time. Yet with such large numbers of people with chronic health problems released from prison and such a geographically large and diverse state, scalability is clearly a challenge.

A principle component of this challenge is to obtain sufficient funding to provide health care for these patients. Our earlier analyses suggest that a substantial population of prisoners with chronic health problems in Medicaid non-expansion states fall into the Medicaid “gap,” in which they are too financially impoverished to purchase health insurance, but do not qualify for traditional Medicaid [25]. Indeed, opering in a non-expansion state, we are finding that the majority of our patients have no medical insurance and cannot afford even the low sliding-scale fees at community health centers and their discounted pharmacy services. Presently, we have raised private funds to support a voucher system that can cover these costs for our patients, but this funding is limited.

The most direct remedy for this challenge would be the expansion of Medicaid—or a regulatory provision granting recently released people temporary Medicaid coverage. However, given the state legislature's historical position on Medicaid expansion, the plausibility of either of these options is uncertain [26].

In this context of limited resources, we propose four near-term priorities, which we seek to achieve in the next two years (by 2021), to further address the health needs of those released from prison. First, we propose opening additional FIT Program clinics so that at least 800 people, or 10% of those annually released with chronic conditions, can receive continuous care between correctional and community settings. We aim to initiate these clinics in every region of the state. Second, we plan to support FIT Program sites' capacity in treating patients with MAT for OUD. Third, we plan to conduct outreach to non-FIT Program safety net clinics that likely interact with large numbers of released people (eg, community clinics, emergency departments) to promote culturally sensitive and non-stigmatizing care for justice-experienced people. And fourth, we plan to encourage schools of medicine and allied health professionals to incorporate these same skills and values into their curriculum so that they become ingrained in the health care workforce.

In addition to informing programs supporting continuity of care in North Carolina, our findings contribute to the wider epidemiologic literature examining disease prevalence among correctional populations. This contribution is notable considering that prevalence data from US prisons have been difficult to obtain historically. The most widely cited prevalence estimates, those stemming from a nationally representative survey conducted from the US Department of Justice's Bureau of Justice Statistics (BJS) in 2004 and 2011, relied on self-reports to estimate prevalence of 15 health care conditions [16, 17]. The validity of the BJS survey items remains unassessed, and may be subject to issues of recall bias, stigma, education, and perceived secondary gains. National data based on clinical diagnosis data do not exist, and to our knowledge, a 2006-2007 study from the Texas state prison system is the only statewide prevalence study to use diagnosis data [27]. This lack of prevalence data is corroborated by a recent nationwide survey of prison system administrators, which found that while many prison systems track the prevalence of the most expensive conditions amongst their populations, few comprehensively track the prevalence of all chronic conditions in their state [12]. More systematic efforts at tracking disease prevalence are needed both in North Carolina and across the country if prison systems and their surrounding communities are to adequately care for released people.