Does Adherence Change When No One is Looking? Comparing Announced and Unannounced Tenofovir Levels in a PrEP Trial

Differences between unannounced and announced tenofovir levels as measures of PrEP adherence are not well understood. In an ancillary adherence study involving one urban site (Kampala) and two rural sites (Kabwohe and Tororo) from the Partners PrEP study, 268 specimen pairs from chronologically proximal clinic and home visits were tested for plasma tenofovir levels. Comparing clinic and home specimens, 89 versus 89 % were classified as detectable (>0.31 ng/ml; p = 0.77), 87 versus 86 % as recent dosing (>10 ng/ml; p = 0.80), and 82 versus 80 % as steady-state (>40 ng/ml; p = 0.44). Mean difference between announced and unannounced drug levels, adjusted for specimen collection time was 3.2 ng/ml (p = 0.50) for Kabwohe, 23.2 ng/ml (p = 0.003) for Kampala and -3.3 ng/ml p = 0.69) for Tororo. In the setting of high adherence, plasma tenofovir levels tested at the clinic were categorically similar as levels tested at home; however, differences were seen between urban and rural settings.