Assessing the effect of HAART on change in quality of life among HIV-infected women

Abstract

Background: The impact of highly active antiretroviral therapy (HAART) on health-related quality of life (QOL) of HIV-1 infected individuals in large prospective cohorts has not been well studied.

Objective: To assess the effect of HAART on QOL by comparing HIV-infected women using HAART with HIV-infected women remaining HAART naïve in the Women's Interagency HIV Study (WIHS), a multicenter prospective cohort study begun in 1994 in the US.

Methods: A 1:1 matching with equivalent (<or= 0.1%) propensity scores for predicting HAART initiation was implemented and 458 pairs were obtained. HAART effects were assessed using pattern mixture models. The changes of nine QOL domain scores and one summary score derived from a shortened version of the MOS-HIV from initial values were used as study outcomes.

Results: The background covariates of the treatment groups were well-balanced after propensity score matching. The 916 matched subjects had a mean age of 38.5 years and 42% had a history of AIDS diagnosis. The participants contributed a total of 4,292 person visits with a median follow-up time of 4 years. In the bivariate analyses with only HAART use and time as covariates, HAART was associated with short-term improvements of 4 QOL domains: role functioning, social functioning, pain and perceived health index. After adjusting for demographic, socioeconomic, biological and clinical variables, HAART had small but significant short-term improvements on changes in summary QOL (mean change: 3.25; P = 0.02), role functioning (6.99; P < 0.01), social functioning (5.74; P < 0.01), cognitive functioning (3.59; P = 0.03), pain (6.73; P < 0.01), health perception (3.67; P = 0.03) and perceived health index (4.87; P < 0.01). These QOL scores typically remained stable or declined over additional follow-up and there was no indication that HAART modified these trends.

Conclusion: Our study demonstrated significant short-term HAART effects on most QOL domains, but additional use of HAART did not modify long-term trends. These changes could be attributed to the direct effect of HAART and indirect HAART effect mediated through clinical changes.

Figure 1

Boxplots of QOL summary score between HAART users and HAART naïve groups before and after propensity-score matching. Box widths are proportional to the number of observations in each group.