There is evidence that persons with severe mental illness have higher rates of chronic health conditions and a reduced life expectancy compared with the general population (1,2,3,4,5). For example, between 58% and 74% of persons with severe mental illness report at least one chronic health problem, such as hypertension, diabetes, and pulmonary disease (1,6). Further, persons with severe mental illness, particularly those diagnosed as having schizophrenia, live fewer years on average and have higher relative mortality risks than the general population (5,7,8). The medical needs of persons with mental illness are of particular concern because our mental health and primary health care systems are not well integrated, and there is evidence that the health care needs of persons with mental illness often go unmet (9).

In this context, there is little information about how the general medical problems of persons with severe mental illness who have a history of incarceration—defined here as one or more incarcerations in a local jail during a five-year study period—compare with the medical problems of persons with mental illness who do not have a history of incarceration. This gap in our knowledge is of concern because incarcerated individuals, in general, have higher rates of chronic illness, such as diabetes, coronary artery disease, and infectious diseases (10,11). Compared with consumers with no history of incarceration, are persons with severe mental illness and a history of incarceration at even greater risk of general medical problems? The aim of this study was to contribute to our knowledge about the complex needs of persons with mental illness by examining the synergistic effects of severe mental illness and incarceration on the general medical problems of persons with mental illness.

A quasiexperimental design was used to compare the general medical problems of persons with severe mental illness who have an incarceration history and persons with severe mental illness with no such history. Administrative data from local Medicaid enrollment and claims files, a local community mental health authority, and a local jail were obtained in order to enumerate all Medicaid (N=168,993), jail (N=120,109), and community mental health users (N=29,831) in King County (Seattle), Washington, from 1999 through 2003. Then data from these three systems were linked in order to identify single- and multisystem user groups: Medicaid only (N=122,842); jail only (N=88,704); mental health only (N=6,452); Medicaid and jail (N=23,955); Medicaid and community mental health (N=15,929); jail and community mental health (N=1,183); and Medicaid, jail, and community mental health (N=6,267). For this study, 100% of the sample of Medicaid, jail, and community mental health users (N=6,267) and a random sample of 25% of the Medicaid and community mental health users (N=3,983) were used.

Next, a definition of severe and persistent mental illness was applied to these two samples. Here, severe and persistent mental illness was defined as the conjunction of diagnosis, disability, and duration (12). Diagnostic information came from community mental health records. All persons who had a diagnosis of schizophrenia, affective disorder (with some exceptions), delusional disorder, or psychotic disorder not otherwise specified and who were enrolled in Medicaid were classified as having a severe and persistent mental illness. Because of the lack of an independent measure of severity or chronicity of mental illness in the administrative data, Medicaid eligibility because of Supplemental Security Income or Social Security Disability Income was used as a proxy for duration and disability because it takes evidence of profound impairment of prolonged duration to become eligible for these entitlement programs.

After applying this definition, the Medicaid, jail, and community mental health group contained 3,690 individuals with severe mental illness, which was 59% of the original group (N=6,267) and the Medicaid and community mental health group contained 2,062 individuals with severe mental illness, which was 52% of the original group (N=3,983). These groups included all Medicaid-eligible mental health consumers with severe mental illness in the county during the study period.

A consumer was classified as having a history of incarceration if he or she was incarcerated in jail one or more times during the study period. Next, data from Medicaid claims, outpatient behavioral health providers, and local and state hospitals were used to identify diagnosis of or treatment for general medical problems from 11 broad diagnostic categories, which included infectious and parasitic diseases; cancer; endocrine, nutritional and metabolic disorders; diseases of the blood; diseases of the nervous system; diseases of the circulatory system; diseases of the respiratory system; diseases of the digestive system; diseases of the skin; diseases of the musculoskeletal system; and injury and poisoning. To restrict the data to confirmed diagnoses, codes for ruling out diagnoses and diagnostic testing were excluded.

This study was approved by the institutional review boards at the University of North Carolina at Chapel Hill and the Washington State Department of Social and Health Services Research and Data Analysis Division.

Chi square tests were used to examine the association between incarceration history (coded yes or no) and 11 dummy variables (coded yes for having a medical problem or no for not having a problem) representing each of the 11 categories described above. Also, dummy variables for any problem (coded yes or no) and multiple problems (coded yes or no) were created, and their associations with incarceration history were examined.

Next, a series of logistic regression models were used to examine the relationships between incarceration history and the probability of having the general medical problems that were significant in the bivariate analyses, controlling for gender, race, age, and co-occurring alcohol and substance use disorders. Also, separate logistic regression models were used to examine the associations between incarceration history and having any problem and multiple problems.

The demographic characteristics of the samples are shown in Table 1. Compared with consumers with no incarceration history, those with such a history were more likely to be male (60% versus 38%) (χe²=239.10, df=1, p<.001) and African American (27% versus 13%) (χe²=295.10, df=1, p< .001). Also, compared with consumers with no incarceration history, those with an incarceration history were more likely to be diagnosed as having schizophrenia (37% versus 32%) (χe²=15.59, df=1, p<.001) and were more likely to have co-occurring substance use disorders (49% versus 14%) (χe²=701.34, df=1, p<.001).

Also shown in Table 1, consumers with an incarceration history were more likely to have infectious diseases (34% versus 27%), blood disorders (13% versus 11%), skin disorders (23% versus 18%), and injuries (38% versus 29%). Consumers without an incarceration history were more likely to have medical problems related to cancer (11% versus 6%) and endocrine disorders (29% versus 25%).

Moreover, consumers with an incarceration history were 1.4 times more likely than those with no such history to have any general medical problem and were 1.3 times more likely to have multiple medical problems. The statistically significant bivariate relationships remained significant in the multivariate models, with the exception of endocrine disorders (Table 1). Thus, compared with those with no incarceration history, consumers with an incarceration history were more likely to have infectious diseases (odds ratio [OR]=1.4), blood disorders (OR=1.2), skin disorders (OR=1.4), and injuries (OR=1.5) when the analyses controlled for gender, race, age, and substance use disorders.

As shown in Table 2, when the analyses controlled for gender, race, age, and substance use disorders, consumers with an incarceration history were 1.4 times as likely to have any general medical problem and 1.3 times as likely to have multiple physical health problems. That is, persons with severe mental illness with an incarceration history were 40% more likely to have medical problems and 30% more likely to have multiple problems compared with those with no history of incarceration.

This population-based study of persons with severe mental illness expands our knowledge about the general medical problems of one of our country's most vulnerable populations. Findings suggest that the behaviors, chaotic lifestyles, and unsafe living environments that increase the probability of incarceration are likely associated with an increase in the probability of developing a number of medical problems for persons with severe mental illness. Also, to some extent these findings give credence to the notion that jails have become public health outposts such that a jail detention for some persons living with mental illness could be their first, only, or best opportunity to get screened, diagnosed, and treated for general medical problems (13).

This study has a number of important implications. First, information sharing between local public mental health, public health, and local jails is critical. Communication among the primary health, behavioral health, and jail systems is generally poor and information-sharing arrangements that would allow for continuity of care from the community to the jail (that is, community mental health agents communicating to jail staff about an individual's medical or prescription needs) and from jail back to the community (that is, discharge planning and timely notification of release to ensure aftercare appointments and follow-up for general medical and behavioral health care) are not typical in most communities. Poor communication among these systems is of serious concern for consumers because many medications prescribed for the symptoms of mental illness have metabolic and cardiovascular side effects that could exacerbate existing medical conditions (14). Consumers could be at risk if jail medical staff members do not have access to medical histories and consumers are unwilling or unable to share information about comorbid mental illness and general medical problems. Moreover, left untreated upon release to the community, infectious and blood disorders among persons with severe mental illness can have the potential to become serious and costly public health issues. Local barriers that prevent information sharing among public health, behavioral health, and jails must be overcome so that health care for persons with severe mental illness is not disrupted.

Second, the integration of primary care and behavioral health care is important for all consumers but may be especially so for consumers who have a history of incarceration because these consumers are more likely to have multiple medical problems. The integration of primary care and behavioral health care is a significant challenge for many communities around the country, and a number of integration models and strategies have been implemented, including colocating primary care, behavioral care, and pharmacy services; embedding nurse care managers within mental health treatment teams; offering wellness programs to educate consumers about diet and exercise, diabetes management, smoking cessation, disease management, and other topics; and providing evidence-based screening and education related to heart, vascular, and respiratory diseases; infectious diseases; injury and violence; alcohol, drug, and tobacco use; and nutritional issues (15,16,17,18,19). The extent to which these strategies are implemented in routine mental health settings is not clear, and the evidence base that these strategies improve the physical well-being of persons with severe mental illness needs to be developed. Nevertheless, these strategies do offer some hope and guidance for community mental health authorities struggling to integrate primary and behavioral health care.

Finally, local jails were never designed to be providers of general medical and behavioral health care, yet jails have seen their role as providers of such care expand greatly over the past decade or longer. Diverting persons with severe mental illness from jails and facilitating community reentry from jails and prisons are strategies that could lessen the burden of care on the criminal justice system and prevent disruptions in primary and behavioral health services among consumers. Research on evidence-based diversion and reentry programs for persons with severe mental illness must be accelerated in light of the fact that persons who have general medical and behavioral health problems have an increasingly difficult time with successful community reentry, including accessing safe and adequate housing, obtaining gainful employment, and accessing much-needed general health and behavioral health services (10).

The strengths of this study include unique data from multiple public systems and population-based samples of persons with severe mental illness over a five-year study period. Limitations of the study include the reliance on administrative data to enumerate the general medical problems of persons with severe mental illness and the generalizability of findings from one county. Moreover, data about tobacco use were not available for this study. Given the association between smoking and cardiovascular, respiratory, and neoplastic disease, future research should include tobacco use when examining comorbid general medical problems among persons with severe mental illness. Also, the sample in this study excludes consumers who were not in contact with the behavioral health service system long enough or consistently enough to become eligible for Medicaid. The data used in this study are over five years old, and changes or trends in the medical problems of consumers over the past several years, although unlikely, may not be reflected.

The findings of this study underscore the need for improved information sharing among mental health, public health, and local jail systems; encourage the implementation of strategies designed to better integrate mental health and primary care among mental health providers; and call for accelerating research on effective criminal justice diversion and reentry interventions.

This work was supported in part by grant MH-63883 from the National Institute of Mental Health. For assistance in assembling the administrative data for this study, the authors are indebted to the Mental Health, Chemical Abuse and Dependency Services Division of the King County Department of Community and Human Services, the King County Department of Adult and Juvenile Detention; the Divisions of Mental Health, Medical Assistance Administration, and Alcohol and Substance Abuse of the Washington State Department of Social and Health Services; and the Center for Health Statistics at the Washington State Department of Health. The authors are grateful for the data management assistance of John Bainbridge, M.S., and Shirley Richards.

The authors report no competing interests.