Measures
This study assessed adverse childhood experiences, psychiatric problems, substance use disorders, physical health, and community functioning through chart review, structured interviews, and self-report instruments.
Psychiatric diagnoses. Psychiatric diagnoses were based on chart records for 222 participants (81%) and on the Structured Clinical Interview for DSM-IV (SCID) (
28 ) for 52 participants (19%). Four of the sites assessed the validity of the chart diagnoses by administering the SCID and found high concordance rates (
κ =.72). This supports the validity of chart diagnoses for the study participants.
Adverse childhood experiences. Standardized, self-report measures administered in an interview format were used to assess childhood abuse, household dysfunction, and losses that occurred during the first 16 years of participants' lives. Childhood sexual abuse was assessed with the Sexual Abuse Exposure Questionnaire (
29 ). This scale identifies ten categories of increasingly invasive sexual experiences via self-report and has good test-retest reliability among clients with severe mental illness (
30 ). Participants were considered to have experienced childhood sexual abuse if they responded affirmatively to any of the six items involving physical sexual contact.
Childhood physical abuse was assessed by the three most severe items from the violence subscale of the Conflict Tactics Scales (
31 ). An affirmative response on any of the three questions (being hit, being knocked down or thrown, and being burned or scalded on purpose) was used to indicate a history of physical abuse.
Parental separation or divorce was defined by an affirmative response to the question, "When you were growing up, did your parents/caretakers get a divorce or separation?" Domestic violence was assessed by the question, "When you were growing up, did you see or hear your parents/caretakers arguing or fighting a lot?" Foster/kinship care was assessed by two questions that asked whether the respondent was placed in an orphanage, foster home, boys' home, reformatory, detention, jail, or similar placement or sent to live with relatives, family friends, or other people. Parental mental illness was derived by an affirmative response to the question, "When you were growing up, did your parent(s)/caretaker(s) ever see a counselor, psychologist, or psychiatrist, or go to a mental hospital, or take medication for an emotional problem?" Parental death was based on client self-report of whether the client's father or mother died before the client turned 16.
Psychiatric problems. Suicidal ideation, self-injurious behavior, and recent suicide attempts were assessed by asking participants if they felt so low that they thought of suicide, if they tried to hurt themselves, and if they attempted suicide in the past six months. Hospitalization history was measured via self-report by age at first hospitalization, the total number of psychiatric hospitalizations, and the number of hospitalizations in the previous year.
PTSD was assessed with the PTSD Checklist (PCL) (
32 ), a self-report measure. The PCL includes 17 questions, one for each
DSM-IV PTSD symptom; it requires the respondent to rate the severity of each symptom over the past month on a 5-point Likert scale. A PTSD diagnosis is made if at least one criterion B (intrusive) symptom, three criterion C (avoidant) symptoms, and two criterion D (hyperarousal) symptoms are rated at 3 or above on the Likert scale or if the total PCL score is 45 or more. The PCL has strong test-retest reliability and convergent validity among persons with severe mental illness (
33 ).
Substance use disorders. Current alcohol and drug use disorders were identified with the Dartmouth Assessment of Lifestyle Instrument (DALI) (
34 ). The DALI is an 18-item screening tool for substance use disorders (abuse or dependence) that was specifically developed and validated for persons with severe mental illness. It has high classification accuracy for
DSM-IV current substance use disorders involving alcohol, cannabis, or cocaine. Cutoff scores have been developed by use of an empirically derived algorithm. Clients were categorized as having a substance use disorder if the DALI was positive.
Physical health. Physical health was assessed with items from the Piedmont Health Survey (
35 ), which includes questions on chronic medical problems, including asthma, diabetes, heart trouble, hypertension, arthritis, cancer, lung diseases, ulcers, stroke, epilepsy, head injury, or infectious diseases (for example, sexually transmitted diseases and hepatitis). For this study the number of problems endorsed was summed to form an overall measure of health problems. Participants were also asked to report the number of times in the past six months that they had received care for a physical health problem and the number of days hospitalized for physical health problems.
HIV, hepatitis B, and hepatitis C were assessed through laboratory analyses of blood specimens obtained through venipuncture or finger stick. Serologic tests for HIV antibodies in serum utilized the Genetic Systems HIV-1/HIV-2 enzyme-linked immunosorbent assay (ELISA), and results were confirmed by an HIV-Western blot (BioRad). Antibodies to hepatitis B core were assessed with the Abbott Corzyme test. Antibodies to hepatitis C were assessed in serum via the Abbott HCV-2 ELISA and confirmed by a recombinant immunoblot (Ortho). All serologic testing and procedures were licensed by the U.S. Food and Drug Administration and were performed in laboratories accredited by the College of American Pathologists. Details of the procedures have been described by Rosenberg and colleagues (
36 ).
Community functioning. Information about homelessness was obtained via self-report. Homelessness was defined as having no regular residence or living in a shelter or on the street for at least one day during the past six months.
Poverty status was assessed via self-report and was based on past-year income. The 1999 guidelines of the U.S. Department of Health and Human Services were used to define poverty status; the guidelines take into account income, marital status, and number of children.
Criminal justice involvement was assessed via self-report and defined as ever having been arrested for any offense.
Work functioning was assessed via self-report and defined as whether the client was currently working or had worked in the past year.
Trauma exposure in adulthood. Exposure to physical assault or sexual assault since age 17 and in the past year was measured by the physical assault and sexual assault subscales of the Revised Conflict Tactics Scales (
37 ). Physical assault was defined as any assault, ranging from grabbing, pushing, or shoving to using a knife or gun, that was perpetrated against the participant. Sexual assault was defined as oral, anal, or vaginal intercourse achieved through either physical force or threat.
High-risk behaviors. High-risk behaviors were those associated with increased risk of diseases transmitted via blood, such as HIV, hepatitis B, and hepatitis C (for example, unprotected sex and sharing needles for injection drug use). The AIDS Risk Inventory (
38,
39 ), a structured interview for assessing risk behaviors associated with acquiring and transmitting these infections, was used. The instrument was modified for this study so that it would be easily understood by respondents with severe mental illnesses. The selected risk variables include lifetime trading of sex for drugs, gifts, or money; lifetime injection of a drug; lifetime administration of a drug by sniffing; lifetime needle sharing; lifetime sex between men; and two or more sexual partners in the past six months.