The consistency of adherence to antiretroviral therapy predicts biologic outcomes for human immunodeficiency virus-infected persons in clinical trials.

We prospectively studied long-term antiretroviral adherence patterns and their impact on biologic outcomes for human immunodeficiency virus (HIV)-infected participants in 2 randomized, multicenter clinical trials. For the period from baseline to month 12 of the study, participants who reported adherence levels of 100%, 80%-99%, and 0%-79% had plasma HIV RNA levels that decreased by 2.77, 2.33, and 0.67 log(10) copies/mL, respectively (P<.001), whereas their CD4 counts increased by 179, 159, and 53 cells/mm(3), respectively (P<.001). Adherence predicted nondetectable HIV RNA levels (<50 copies/mL) at 12 months of follow-up (P<.001). The HIV RNA level was nondetectable in 72% of participants who reported 100% adherence at all 4 follow-up visits, compared with 66%, 41%, 35%, and 13% of participants who reported 100% adherence at 3, 2, 1, or 0 follow-up visits, respectively (P<.001). Nonwhite race was associated with poorer adherence (P<.001), and older age was associated with better adherence (P<.001).

Application of quality improvement theory and process in a national multicenter HIV/AIDS clinical trials network.

Effective clinical trials depend on the production of scientifically sound data. Clinical research coordinators monitor various activities to assure that data meet standards for timeliness and quality. Traditional methods of assuring data quality are less than optimal because they are based on correcting mistakes after they occur. Because they are focused on problem prevention, the techniques of continuous quality improvement represent a more effective means of maintaining high quality of data. This article describes the means by which principles of continuous quality improvement were incorporated into an HIV/AIDS clinical research network, as well as outcomes associated with these efforts.

Sustained benefit from a long-term antiretroviral adherence intervention. Results of a large randomized clinical trial.

OBJECTIVE: To assess the efficacy of 2 adherence interventions, medication managers (MM) and medication alarms (ALR), among antiretroviral (ARV)-naive persons with HIV initiating ARV therapy. METHODS: A multicenter, randomized, adherence intervention clinical trial was conducted among participants coenrolled in an HIV ARV strategy study for ARV-naive individuals. Sites were assigned by cluster randomization using a 2 x 2 factorial design to administer MM, ALR, MM + ALR, or neither (control). MM participants received individualized, structured, long-term adherence support from trained MMs. ALR participants received individually programmed ALR alarms for use throughout the study. RESULTS: The 928 participants, followed a median of 30 months, included 22% women and 75% nonwhites; the median baseline CD4 count was 155 cells/mm. First virologic failure was 13% lower in all MM versus no-MM groups (P = 0.13) and 28% lower in MM versus no-MM subgroups randomized to 2-class ARV arms in the parent ARV study (P = 0.01). MM (vs. no-MM) participants had significantly better CD4 cells count (P = 0.01) and adherence (P < 0.001) outcomes. ALR (vs. no-ALR) participants had worse virologic outcomes. CONCLUSION: This large randomized clinical trial demonstrated that interpersonal structured adherence support was associated with improved long-term medication adherence and virologic and immunologic HIV outcomes.