To the Editor:
Self-reported rates of antiretroviral adherence typically exceed those derived from electronic monitoring (MEMS). Because their ability to explain virologic outcomes is inferior to MEMS, it is believed that self-reports overestimate adherence. It is generally assumed that this inaccuracy is, at least in part, a result of the patients' tendency to provide socially desirable responses to questions about adherence.
In their interesting study, Wagner and Miller 1 investigated the influence of social desirability on the accuracy of patients' self-reported adherence. They randomly assigned patients to an intervention comprising a discussion of challenges and difficulties related to adherence with an interviewer versus a control group. The intervention was assumed to reduce discomfort in disclosing nonadherence (ie, reduce socially desirable responding), thereby increasing the accuracy of self-report. Subsequently, they monitored adherence during 2 weeks using MEMS in both groups. Patients in both groups completed a self-report adherence measure at the study end point. They found no clear difference in accuracy of self-reported adherence between the experimental and control groups, suggesting that the influence of social desirability on the accuracy of self-reported adherence may not be as significant as suspected.
Unfortunately, the authors did not mention whether the patients were blinded or not to the purpose of MEMS (ie, monitoring of adherence). If patients were not blinded to the purpose of MEMS, this may have biased the results. If the patients were aware that their medication-taking behavior was registered by MEMS, this could have diminished their tendency to provide socially desirable responses, because the patients knew that their actual medication taking would be disclosed at the end of the observation period anyway.
There is sound evidence that patients provide less socially desirable responses when they are aware that inaccurate answers can be detected by a recording device. 2–4 It is likely that such a situation will arise if self-reported adherence is assessed when patients are aware that their adherence is being monitored with MEMS at the same time. Thus, the results of the study by Wagner and Miller 1 may not be representative for situations in which patients are asked to provide self-reports of adherence without concurrent monitoring with MEMS or when patients are blinded to the purpose of MEMS.
1. Wagner G, Miller LG. Is the influence of social desirability on patients' self-reported adherence overrated?J Acquir Immune Defic Syndr
2. Roese NJ, Jamieson DW. 20 years of bogus pipeline research: a critical review and meta-analysis. Psych Bull
3. Alexander MG, Fisher TD. Truth and consequences: using the bogus pipeline to examine sex differences in self-reported sexuality. J Sex Res
4. Tourangeau R, Smith TW, Rasinski KA. Motivation to report sensitive behaviors on surveys: evidence from a bogus pipeline experiment. J Appl Soc Psychol