Discrepancies between self-reported adherence and a biomarker of adherence in real-world settings.

Background Pre-exposure prophylaxis (PrEP) is only effective in preventing new HIV infections when taken consistently. In clinical practice, asking a patient about their adherence (self-report) is the predominant method of assessing adherence to PrEP. Though inexpensive and noninvasive, self-report is subject to social desirability and recall biases. Several clinical trials demonstrate a discrepancy between self-reported adherence and biomarker-based recent adherence. Less is known about the accuracy of self-report in real-world clinical settings. This brief report addresses this knowledge gap and describes the concordance between self-reported adherence and biomarker-based adherence in real-world clinical settings. Methods A Liquid Chromatography-Mass Spectrometry (LC-MS/MS) urine test for Tenofovir was developed and used clinically to detect recent non-adherence (no dose in at least 48 hours) for each individual. Two clinics' standard operating procedures recommend utilization of the urine-based adherence test for patients who self-report that they are not struggling with adherence. Those who self-report struggling with adherence receive enhanced adherence support without the need for additional testing. The number of results indicating recent non-adherence from these two clinics were analyzed to assess the concordance between self-reported adherence and biomarker-based adherence. Results Across two clinics, 3,987 tests were conducted from patients self-reporting as "adherent", and 564 (14.1%; 95% CI: 13.1%-15.2%) demonstrated recent non-adherence with the LC-MS/MS test. At Clinic #1 in Florida, 3,200 tests were conducted, and 465 (14.5%; 95% CI: 13.3%-15.8%) demonstrated recent non-adherence. At Clinic #2 in Texas, 787 tests were conducted, and 99 (12.6%; 95% CI: 10.4%-14.9%) demonstrated recent non-adherence. Conclusions Utilization of biomarker-based adherence monitoring at these two clinics resulted in 564 additional patients receiving enhanced adherence support who otherwise would not have been identified as non-adherent to their prescribed PrEP regimen. These findings suggest that objective adherence monitoring can be used clinically to enable providers to identify non-adherent patients and allocate support services accordingly.