Self-reported adherence is more predictive of virological treatment response among patients with a lower tendency towards socially desirable responding

Background: Self-report is the most commonly used measure of adherence to highly active antiretroviral therapy, but typically shows weaker associations with virological treatment outcome than more objective adherence assessment methods. Socially desirable responding might hamper the validity of self-reported adherence. We investigated whether stratifying patients according to their socially desirable response set might improve the prediction of virological treatment response by self-reported adherence. Methods: Patients enrolled in the focus group of the Dutch national cohort ATHENA completed a social desirability scale, four self-report adherence questions, and had their plasma HIV type-1 (HIV-1) RNA concentrations measured. We calculated odds ratios and 95% confidence intervals for self-reported non-adherence to predict HIV-1 RNA>50 copies/ml among patients with a lower or a higher tendency towards socially desirable responding. Results: A total of 331 patients were included. Self-reported non-adherence was significantly predictive of HIV-1 RNA>50 copies/ml on three out of four questions among patients with lower socially desirable responding (n=198). Self-reported non-adherence did not predict HIV-1 RNA>50 copies/ml among patients with higher socially desirable responding (n=132). Conclusions: Stratifying patients according to their socially desirable response set improved the prediction of virological treatment response by self-reported adherence. This finding emphasizes the importance of discussing medication adherence with patients in a non-threatening and non-judgemental way that normalizes non-adherence in order to reduce socially desirable responding.