Effect of audio computer-assisted self-interview (ACASI) mode on report of drug use and risky health behaviors

Background: Baltimore has the second highest incidence of AIDS cases of any major metropolitan area, with 40.4 cases per 100,000. Fifty percent of all reported living HIV and AIDS cases in Maryland were residents of Baltimore City at the time of diagnosis. Moreover, the fastest growing subgroup with AIDS is heterosexual men and women who are linked sexually to injection drug users. This study sought to determine whether audio computer assisted self-interviewing (audio-CASI) is more effective than standard interview methods in eliciting reliable information regarding stigmatized and sensitive behaviors, specifically, risky needle use and sex behaviors.

Methods: Study participants were drawn from enrollees in a NIH-funded research project (R01DA011402) entitled, “Engaging Detoxification Clients into Long-Term Treatment,” conducted at a treatment center in Baltimore, MD. All individuals recruited into the study were administered the Texas Christian University (TCU) AIDS Risk Assessment—a 19-item instrument that contains both drug use and sex risk items based on the past six months and past 30 days, in both face-to-face interview and audio-CASI format.

Results: While both formats had good reliability, participants were significantly more likely to report the more stigmatized activities of male/male sex, sex trade, and sex with an injection drug user on the audio-CASI than in the face-to-face interview.

Conclusion: The audio-CASI format was successful in identifying risk-taking behaviors that were not captured with a face-to-face interview. Thus, it has the potential to provide more accurate assessment of risky health behaviors than conventional interview methods.