In the psychiatric and popular literature, it is commonly reported that the largest concentrations of institutionalized persons with mental illness in the United States are now found in correctional environments (1). In New York State the number of prison inmates receiving mental health services has grown steadily, almost doubling from 4,500 in 1990 to 8,600 today. More recently, from January 1, 2004, to January 1, 2008, there was a 12.7% increase in total number of inmates receiving mental health services in state prison, and the number of inmates being served in prison who have a diagnosis of serious mental illness seems to be growing at a faster rate (2). In the same four-year period there was a 21.7% increase in patients who had a diagnosis of serious mental illness. The number grew from 3,117 patients on January 1, 2004, to 3,795 on January 1, 2008 (2).

The upsurge in inmates needing or receiving mental health care in correctional facilities is also occurring nationally. In a 2001 National Institute of Corrections survey 25 of 30 states report growth (3). Recent Bureau of Justice Statistics studies corroborate this increase (4). The objective of the study presented here was to review the characteristics of patients with serious mental illness diagnoses newly incarcerated in New York State prisons to help plan for their service needs and to provide information for potential diversion policy initiatives.

In New York State the Central New York Psychiatric Center, under the auspices of the Office of Mental Health (OMH), is responsible for the mental health treatment of prison inmates. To serve their needs, there is a 206-bed inpatient hospital and there are mental health units at 23 of the 69 prisons. The mental health units have a range of programming, including clinic, residential, crisis observation, and discharge planning services.

Information for this study was extracted from the charts of inmates entering the New York State prison system who received a diagnosis of serious mental illness from OMH staff between May 15, 2007, and June 14, 2007. Diagnoses considered serious mental illness include schizophrenia or psychotic disorders, bipolar I or II disorder, bipolar disorder not otherwise specified, mood disorder not otherwise specified, and major depressive disorder. (The first author has used this definition of serious mental illness in a previous article in Hospital and Community Psychiatry [former name of Psychiatric Services] [5].) The New York State Office of Mental Health institutional review board has approved the use of deidentified data initially collected for quality assurance purposes. Variables available were age, gender, primary psychiatric diagnosis, year of first inpatient psychiatric hospitalization, year of first serious mental illness diagnosis, number of prior psychiatric hospitalizations, number of serious suicide attempts, number of prior inpatient substance abuse admissions, number of prior New York State prison incarcerations, and most serious current crime.

An initial psychiatric diagnosis is made by an OMH psychologist or licensed social worker within a few days after arrival in state prison. To determine the diagnosis the clinician reviews the records that accompany the inmate, such as the presentence investigation and county jail treatment records; reviews all the assessments completed by correctional staff; searches several databases for prior treatment and hospitalization information; and conducts a face-to-face interview. A few days later a psychiatrist reviews the chart material, conducts a second interview, and confirms or modifies the diagnosis.

In the analysis three subgroups were examined. They include patients committed for violent versus nonviolent offenses, repeat versus first commitment in New York State, and women versus men. The first two subanalyses were chosen because first offenders and nonviolent offenders with a diagnosis of serious mental illness could be the first targets for diversion. A subanalysis on women was conducted, as they are an often-overlooked subgroup in research studies.

There were 2,918 inmates received into New York State prison between May 15, 2007, and June 14, 2007. A total of 514 (18%) were admitted to mental health services, and 172 (6%) were given a diagnosis of serious mental illness. The mean±SD age of these 172 patients was 36±9.6.

Forty-eight (28%) patients received a diagnosis of schizophrenia; 14 (8%), psychotic disorder not otherwise specified; 22 (13%), major depressive disorder; 21 (12%), bipolar I or II disorder; 33 (19%), bipolar disorder not otherwise specified; and 33 (19%), mood disorder not otherwise specified.

A total of 167 (97%) had a prior inpatient hospitalization, and 48 (28%) had four or more. Seventy-nine (46%) had their first hospitalization ten or more years ago. Seventy-six (44%) were first diagnosed as having a diagnosis of serious mental illness ten or more years ago. A total of 107 (62%) had at least one serious suicide attempt, and 31 (18%) had three or more attempts.

A total of 101 (59%) had received inpatient substance abuse treatment, and 27 (16%) had three or more instances. Seventy-nine (46%) of the patients had been incarcerated in New York State prison before. Sixty-seven (39%) had committed a violent crime—15 (9%) committed assault, 41 (24%) committed robbery, five (3%) committed a sex crime, five (3%) committed murder or manslaughter—compared with 25% of all prison admissions (730 of 2,918 admissions) (χ²=4.85, df=1, p<.001).

A total of 212 women were received into prison during this study period. Compared with men, a higher percentage of women were admitted to mental health services (76 of 212 women, or 36%, versus 676 of 2,706 men, or 25%) (χ²=12.3, df=1, p<.001) and received a diagnosis of serious mental illness (32 of 212 women, or 15%, versus 135 of 2,706 men, or 5%) (χ²=31.4, df=1, p<.001). Of the variables available, female patients with a diagnosis of serious mental illness were significantly different from males only on psychiatric diagnosis. Compared with men, women were more likely to have major depressive disorder (11 of 32 women, or 34%, versus 11 of 135 men, or 8%) (χ²=16.0, df=1, p<.001) and bipolar I or II disorder (ten of 32 women, or 31%, versus 11 of 135 men, or 8%) (χ²=10.0, df=1, p<.01), and they were less likely to have mood disorder not otherwise specified (one of 32 women, or 3%, versus 32 of 135 men, or 24%) (χ²=6.2, df=1, p<.05).

Although the average age of patients with nonviolent crimes and those with violent crimes was similar, it was longer ago that the patients with nonviolent crimes had their first hospitalization (13 years versus nine years) (t=2.8, df=103, p<.05) and their first diagnosis of serious mental illness (12 years versus nine years) (t=2.2, df=122, p<.05). Also, nonviolent patients had more prior state prison incarcerations (1.0 versus .5) (t=2.8, df=168, p<.01).

Patients with a diagnosis of serious mental illness serving their first New York State prison incarceration were significantly younger than those with previous incarcerations (mean±SD of 32±10.1 versus 38±8.8) (t=3.8, df=94, p<.001).

The data clearly show that people entering prison who receive a diagnosis of serious mental illness have a significant and chronic psychiatric and substance abuse treatment history. Although treatment for these patients during their time in prison is absolutely necessary, a major policy question that should be asked is why were these patients sent to prison instead of being diverted to a community program? Areas to investigate include access and retention in assertive community treatment programs (6), the use of diversionary efforts such as mental health courts (7), involuntary outpatient commitment (8), and preconviction processes. Research studies have reported that inmates with serious mental illness can be successfully treated in the community (9), and the seminal MacArthur study (10) found that the risk of violence among formerly hospitalized patients with serious mental illness was equal to or lower than the risk in the general population.

This review collected information for only one month. The volume of patients arriving in New York State prison during this study period is similar to other periods, and although there is no reason to believe that our sample is unrepresentative, additional study is necessary. Also, this study included data only from New York State. Similar studies in other states are required.

Future studies should also focus on participation in community mental health services before and at the time of arrest, and they should examine whether the courts ordered forensic exams or considered alternatives to incarceration. Although the study presented here reviewed only patient charts, patients could be interviewed and databases such as Medicaid claims could be mined to investigate these questions.

No policy changes that affected diagnosis process and practice were initiated in the four years before this study that would account for the growth in number of patients receiving a diagnosis of serious mental illness. Therefore, we believe the increase is real and does not simply represent better detection.

Finally, a study of the reliability of the diagnosis for these 172 patients found that the initial diagnosis made by the psychologist or social worker upon entry to the prison system and the diagnosis made by the psychiatrist only a few days later was identical in about 95% of the cases (Stapholz B, Way B, Sawyer D, unpublished manuscript, 2008). At six months, 80% of the diagnoses were identical. Some change in diagnosis at six months would be appropriate as additional community records arrive and there have been numerous clinician sessions to further clarify diagnosis.

The authors report no competing interests.