Schizophrenia and Major Affective Disorder Among Medicaid Recipients With HIV/AIDS in New Jersey James Walkup, PhD, Stephen Crystal, PhD, and Usha Sambamoorthi, PhD Studies of HIV seroprevalence among individuals with serious mental illness in psychiatric settings suggest that the co- occurrence of serious mental illness and HIV may be a significant clinical and public health problem. Of 2873 psychiatric patients tested in 11 (mainly inpatient) studies,"- 1 223 (7.8%) were HIV positive, with rates higher among those 39 years or younger, among male patients, and among African Americans and Latinos (vs Whites and Asians).'2 Little is known about rates of serious mental illness among those with HIV/AIDS, nor has any study focused on seriously mentally ill individuals with HIV infection in the community (where most such individuals reside).'3 Because most studies rely on the de-identified blood samples from inpatients (i.e., test results could not be asso- ciated with individual respondents), demo- graphic and clinical characteristics of men- tally ill individuals with HIV infection are not well understood. The social marginality of seriously men- tally ill individuals with HIV infection makes community studies exceedingly difficult. The existence of a research file of New Jersey Medicaid beneficiaries with HIV or AIDS, linked to New Jersey HIV/AIDS Registry information by state staff, provided an oppor- tunity to profile patient characteristics. We used this file to conduct statistical tabulations under a protocol approved by the Rutgers Institutional Review Board; claims were used to identify psychiatric conditions. The result is the first statewide report on community- dwelling patients with serious mental illness and confirmed HIV/AIDS. vices for at least 180 days during the period January 1988 to March 1996. A total of 8294 cases met the inclusion criteria. Interview-based studies can identify seri- ous mental illness on the basis of diagnosis, impairment, and duration,'4 but claims data provide information only on diagnosis. There- fore, we defined serious mental illness as schizophrenia or major affective disorder, con- ditions with a substantial impact on function- ing. International Classification ofDiseases (9th revision; ICD-9) primary diagnosis codes in the 295.xx range were used for schizophre- nia. Referred to as "affective psychoses" in the ICD-9, major affective disorder was defined with codes in the 296.xx range, which includes manic, major depressive, and bipolar disorders. Our goal was a lower-bound estimate with few false positives, even at the cost of some missed cases. Therefore, we took 2 steps. We relied on only "high-credibility" diagnoses (those assigned by hospitals, physi- cians, medical clinics, or mental health providers). Furthermore, we adapted an algo- rithm developed by Lurie and colleagues,'5 requiring at least 1 inpatient or 2 outpatient diagnoses for a given condition to assign an individual to that category. Those eligible for both schizophrenia and major affective disor- der were placed in the schizophrenia group. Individuals who did not meet criteria for either schizophrenia or major affective disor- der were classified as having no serious men- tal illness. Measures Injection drug use history was based on registry classification of route of infection. Methods Participants Full details on file creation, codes, cri- teria, and classification procedures are available in an appendix that can be obtained from the authors. A brief descrip- tion follows. Criteria for study inclusion included (1) diagnosis of AIDS or HIV by March 1996, (2) being 18 years or older at the time of diagnosis, and (3) receipt of Medicaid ser- The authors are with the Institute for Health, Health Care Policy, and Aging Research, Rutgers University, New Brunswick, NJ. Requests for reprints should be sent to James Walkup, PhD, Institute for Health, Health Care Pol- icy, and Aging Research, Rutgers University, 30 College Ave, New Brunswick, NJ 08901 (e-mail: walkupgrci.rutgers.edu). This paper was accepted December 23, 1998. Note. The findings and opinions reported here are those of the authors and do not necessarily represent the views of any other individuals or organizations. American Journal of Public Health 1 101