• 9

Study Design

This is a cross-sectional study of risk behaviors and the prevalence of HBV and HCV infection among veterans with SMI. We recruited 399 consecutive patients admitted to the Durham VA inpatient psychiatric unit between March 1998 and June 2000. The study setting was a VA psychiatric ward with 26 inpatient beds and an average length of stay of eight days.

Patient Population

All participants were psychiatric inpatients who met diagnostic criteria for SMI, provided informed consent, and were between the ages of 20 and 80. The Durham VA’s Institutional Review Board approved the study and the manner by which informed consent was obtained from participants. Most participants were from the surrounding catchment area of North Carolina and southern Virginia, a mixed urban/rural region. Primary diagnoses were schizophrenia, schizoaffective disorder, bipolar-I disorder, and post-traumatic stress disorder (PTSD). Psychiatric diagnoses were based on clinical diagnoses, record review to confirm that participants met the diagnostic criteria of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV), and consensus review of the diagnosis by two attending psychiatrists. Confirmation by the Structured Clinical Interview for DSM-IV (SOD) was also completed in a subset of 25 patients; in all patients, psychiatric diagnoses were concordant with the clinical diagnoses. A PTSD diagnosis was additionally confirmed through VA records indicating that a patient was receiving compensation for disability related to PTSD or had exacerbation of PTSD symptoms rendering the hospital admission and through a confirmation using the PTSD Checklist (PCL). Only 3% of eligible persons refused to participate in the study. Participants were paid a participation fee of $25.

Data Collection

After giving informed consent, each study participant completed a standardized, hour-long, structured risk interview conducted by two trained interviewers who were experienced with this population and had good agreement on inter-rater reliability tests. The risk interview was adapted from established instruments for assessing HIV (is used for treating HIV infection in some patients when used in combination with certain other medicines.) risk behavior, prevalence, and associated variables of relevance in populations with SMI. The domains surveyed on the interview included demographic characteristics and HIV-associated risks and behaviors, including sex-risk behaviors. Study participants also received pretest counseling for HIV (is in a group of medicines called reverse transcriptase inhibitors) and provided blood specimens.


The primary outcome variables were HBV and HCV serostatus, which were determined through a blood specimen collected within 72 hours of the informed consent and the interview. HBV seropositive was defined as having HBV core antibody as assessed by the Abbott Corzyme test according to the manufacturer’s criteria. HCV seropositivity was defined as having antibodies to HCV, indicating current or past HCV infection. Antibodies to HCV were determined in sera samples by the Abbott HCV enzyme immunoassay 2.0. Initially reactive results were repeated in duplicate, and immunoblot (RIBA, Ortho, Raritan, NJ) or polymerase chain reaction (Roche, Indianapolis, IN) confirmation was performed on HCV enzyme immunoassay reactives. All serological tests were performed under well-controlled conditions by laboratory staff at the Durham VA clinical microbiology laboratory, which is accredited by the College of American Pathologists.

Explanatory Variables

The primary explanatory variable was self-identified racial background. Because our sample was composed predominantly of African-American and Caucasian veterans, we were interested in the effects of race. We excluded from the analysis 19 individuals who reported other racial background categories such as Hispanic, American Indian, Pacific Islander, Alaskan Native, and “other” because of small sample sizes in these categories. We also excluded four white individuals who did not have hepatitis test results.We assessed risk behaviors with the AIDS Risk Inventory, which assesses nearly all HBV and HCV risk behaviors. The AIDS Risk Inventory is a structured interview for assessing risk of HIV infection (is used for treating HIV infection in combination with other medicines). It probes risk behaviors associated with drug use and sexual practices and assesses both lifetime risk behaviors and behaviors in the preceding six months. In general, we report lifetime risks, except

for multiple sex partners during the past six months. Lifetime risk was analyzed because it is a better indication of exposure than is past-six-month risk. Compared with other risk interviews, the AIDS Risk Inventory is a more sensitive instrument, is able to discriminate between high- and low-risk HIV behaviors, and reliably assesses behavior patterns associated with a high risk of HIV (is used for treating herpes zoster infection) infection. We previously used this instrument successfully in examining HCV risk factors.


In addition to determining HBV and HCV serostatus, we analyzed the risk behaviors queried on the AIDS Risk Inventory, including IDU, unprotected sex (vaginal, anal, oral), crack use, and sniffing or snorting drugs. Racial differences for demographic and risk factors were assessed by chi-squared statistics.

Multiple logistic regression analyses were performed to examine the association of hepatitis with race, risk factors, and other clinically relevant variables. Variables were selected for inclusion in the models on the basis of clinical relevance, known associations with these infections (e.g., IDU, combat, homelessness), or to adjust for significant differences in the racial distribution of our variables. Because Vietnam-era veterans are at risk for these infections and are over age 50, we used a cut-off value of 50 years or older as a variable. The variables included in the final models were: psychiatric diagnosis, marital status, living independently or with family, age over 50, combat exposure, lifetime risks of IDU, smoking crack cocaine, sniffing or snorting drugs, unprotected sex for drugs, multiple sex partners in the past six months, and the presence of an alcohol use disorder. All analyses were conducted with SAS version 8.1. All statistical tests were two-tailed and conducted at a 0.05 significance level.

Online Pharmacy