Background:  Audio computer-assisted self-interview (ACASI), an information technology tool, has been shown to provide more valid data than face-to-face interviews on sensitive behaviors, such as medication adherence and illicit drug use. We conducted a randomized, controlled study in our inner-city clinic population to evaluate whether immediate ACASI feedback reports given to clinicians would improve HIV-1 RNA viral suppression in patients starting HAART.

Methods:  Patients starting HAART were randomized to have summary reports of ACASI responses relayed to clinicians just prior to all primary care visits (feedback) or remain confidential (control). Randomization was stratified by treatment history (naïve vs experienced). ACASIs were performed over a 12-month period at all primary care visits just prior to patients seeing their provider and included questions regarding adherence to antiretroviral drugs, presence of depression, illicit drug use, alcohol use, and participation in drug or alcohol treatment programs. Suppression of HIV RNA to ≤50 copies/mL was assessed at 6 and 12 months. All analyses were performed as intent-to-treat.

Results:  Between April 1, 2003 and May 30, 2005, 195 patients (43% female, 87% African American, median age 43 years) enrolled; 26% were naïve to therapy. Baseline HIV RNA was 68,066 copies/mL and 52,936 copies/mL, and baseline CD4 count was 177 cells/mm³ and 165 cells/mm³ for the feedback and control groups, respectively (both p >0.15). Overall, there was no difference between groups (feedback 45% vs 38% control; p = 0.32) in HIV RNA suppression with no difference between naïve and non-naive. However, in a sub-group analysis, African American women were more likely to suppress HIV RNA to <50 copies/mL at 6 months (feedback 49% vs control 26%; OR = 2.7; 95%CI 1.03 to 7.34) and 12 months (feedback 36% vs control 28%; OR = 1.40; 95%CI 0.52 to 3.74).

Conclusions:  Our randomized clinical trial suggests that ACASI summary reports may provide a tool to enhance discussions of adherence to HAART and barriers associated with adherence to HAART between patients and their primary care providers. In our inner-city clinic population, use of ACASI did not improve virologic outcomes overall, however, for African American women, use of ACASI resulted an increased likelihood of obtaining an undetectable (<50 copies/mL) HIV RNA 6 months after initiation of therapy.