AEGiS-15IAC: [LbOrB15] SUSTAINED BENEFIT FROM A LONG-TERM ANTIRETROVIRAL (AR) ADHERENCE INTERVENTION: RESULTS OF A LARGE RANDOMIZED CLINICAL TRIAL

15th International AIDS Conference

Bangkok, Thailand — July 11-July 16, 2004

[LbOrB15] SUSTAINED BENEFIT FROM A LONG-TERM ANTIRETROVIRAL (AR) ADHERENCE INTERVENTION: RESULTS OF A LARGE RANDOMIZED CLINICAL TRIAL

Int Conf AIDS. 2004 Jul 11-16;15:Abstract No. LbOrB15

S Mannheimer¹, E Morse², J Matts³, L Andrews⁴, C Miller³, B Schmetter⁵, G Friedland⁴, for the CPCRA 062 Study Team of the Terry Beirn Community Programs for Clinical Research on AIDS⁶
¹Harlem Hospital/Columbia University, New York, United States; ²LSU School of Medicine, New Orleans, United States; ³University of Minnesota, Minneapolis, United States; ⁴Yale University, New Haven, United States; ⁵CPCRA Operations Center, Silver Spring, United States; ⁶NIH/NIAID, Bethesda, United States

BACKGROUND: Despite the need for better antiretroviral (AR) adherence, few randomized controlled trials (RCTs) of effective adherence interventions have been performed.

METHODS: We conducted a RCT among AR-naïve persons with HIV/AIDS to assess 2 long-term adherence interventions: (1) a medication manager (MM) and (2) a medication alarm (ALR™), for their effects on HIV RNA, CD4 count and self-reported adherence. Participants (pts) were enrolled when co-enrolled into an AR strategy study for AR-naïve pts; adherence interventions began before or at AR initiation and continued throughout follow-up. MMs were trained staff members who provided individualized adherence support using standardized assessment instruments. The alarm was a small device programmed to sound at times of pts' medication doses. A 2x2 factorial design was used; sites were assigned by cluster randomization to either: MM, ALR, MM+ALR, or control. Study was powered to detect a 15% lower rate of first virologic failure (HIV RNA > 2000 copies occurring at or after 4 mos.) for each intervention. Outcomes were assessed by life table analyses, intent-to-treat.

RESULTS: 928 pts, enrolled 11/99-1/03 (median follow-up 30 mos.), included 22% women, 75% nonwhites, 15% injection drug users; baseline median CD4 155 cells/mm³, median log HIV RNA 5.2; 38% had AIDS. Baseline characteristics were similar across arms. The rate of first virologic failure was 14% lower in MM vs. no MM pts (p= 0.13). Throughout study, MM pts were significantly more likely than non-MM to have 1) HIV RNA ≤ 400 (rates 190.4 vs. 161.2 per 100 person-yrs, p= 0.02), 2) a greater CD4 increase (p= 0.01), and 3) a higher proportion reporting 100% adherence (p< 0.001). No benefit was seen with the ALR alarm.

CONCLUSION: The CPCRA Adherence Study, the largest RCT of an AR adherence intervention, demonstrated that long-term individualized adherence support from a trained medication manager was associated with important, durable biologic and behavioral benefits.

040711
LbOrB15