Atlanta AIDS study

The dataset provided is a small subset of a larger file. The following description comes from the Master’s thesis of Kimberly R. Jacob Arriola (with slight modification). Run a PROC CONTENTS on the dataset to see which variables are included in your subset.

Study Design

Participants in this study are part of a larger study that explores clinical and psychosocial influences on HIV therapy. In this prospective cohort study, participants underwent a structured interview at two points in time: at baseline and approximately six months later. Also at six-month intervals, clinical and pharmacy information was extracted from medical and pharmacy records.

Participants were recruited on a daily basis. A quota sampling approach was used in an effort to recruit equal numbers of men who have sex with men (MSM), male IDUs, and women to fulfill the needs of the larger study. However, for the current study, the determination of those who engage in drug abuse was made based on self-report during the baseline interview. For each working day, the specialty nurse used the appointment log to identify potential participants. Then, the medical record was used to classify individuals as MSM, IDUs, or women, and individuals were recruited into the study based on their membership in one of these groups. However, because the groups are not mutually exclusive, a hierarchy was used that classified individuals as those who engage in IDU first, followed by MSM and women.

Individuals were invited to participate in the study, and if they agreed to participate, an interviewer was immediately contacted and asked to conduct the baseline interview. For those patients who were not able to complete the interview at that time, an interview was arranged for the near future. Once the interviewer arrived, the baseline interview was conducted in a small private area and took approximately 45 minutes to complete. For each interview, the respondent underwent the consent process before the actual interview took place and was offered $20 in cash for his or her participation.

Each month, a computer print out was used to identify individuals who had completed their baseline interview six months prior. An attempt was made to conduct the follow-up interview in the clinic either before or after an appointment. However, for some individuals this was not possible, so the follow-up interview was conducted in the community. The follow-up interview also took approximately 45 minutes, and individuals were again offered $20 in cash.

Participants

Participants are 119 HIV-infected patients of a large urban AIDS clinic who had both baseline and follow-up data available. Fifty-six percent were male, and 44% were female. Respondents ranged in age from 26 to 61 years (M = 40.31, SD = 7.43). The respondents’ self-identified race/ethnicity is as follows: 83% Black or African American, 15% White, and 1% Hispanic (either Black or White). Participants’ highest grade completed in school is as follows: 30% completed elementary or some high school, 39% graduated from high school or met high school equivalency requirements, 21% had some college or technical training, and 10% graduated college or had begun some graduate work. Only 4% were currently homeless. Sixty-seven percent of respondents were single and had never been married; 22% were divorced, separated, or widowed; and 11% were married. Twenty-five percent of respondents had children currently living with them. All participants were on antiretroviral therapy: 6% were on monotherapy, and 94% were on combination therapy.

Exposure Variables

The baseline interview consisted of seven sections (see Appendix A): (a) General information, (b) Jail/Prison experiences, (c) Medical history, (d) the Brief Symptom Inventory (BSI), (e) Tobacco and drug abuse, (f) Sexual behavior, and (g) Social network information. The current subset includes information from the following sections: General information, BSI, and drug abuse.

General information. This section asked the respondent basic demographic questions such as birth date, age, race/ethnicity, birthplace, current and previous living situation, highest grade completed in school, marital status, child status, and current and previous work situation.

Brief Symptom Inventory(48). This 53-item scale is a brief psychological self-report symptom scale. This scale was designed to be a short form of the Symptom Checklist-90-Revised (SCL-90-R). Each item is rated on a 5-point scale ranging from not at all (0) to extremely (4). The BSI consists of nine dimensions: (a) Somatization (7 items), (b) Obsessive-Compulsive (6 items), (c) Interpersonal Sensitivity (4 items), (d) Depression (6 items), (e) Anxiety (6 items), (f) Hostility (5 items), (g) Phobic Anxiety (5 items), (h) Paranoid Ideation (5 items), and (i) Psychoticism (5 items). Additionally, four filler items are included in the scale. The 9 subscales have demonstrated good test-retest and internal consistency reliability. Additionally, high correlations with the SCL-90-R and the Minnesota Multiphasic Personality Inventory support the validity of the scale, and a factor analysis supports the dimensional structure of the scale (48).

Clinical Outcomes. In the current study, CD4 count ranged from 2 to 762 (M = 271.11, SD = 199.58). Viral load ranged from 81 to 750,000 (M = 52,872.68, SD = 122,323. Both pieces of information were extracted from medical charts.