High Levels of Prevention-effective Adherence to HIV PrEP

An Analysis of Substudy Data From the EPIC-NSW Trial

Benjamin R. Bavinton, PhD; Stefanie Vaccher, PhD; Fengyi Jin, PhD; Garrett P. Prestage, PhD; Martin Holt, PhD; Iryna B. Zablotska-Manos, PhD; Rebecca Guy, PhD; Janaki Amin, PhD; David J. Templeton, PhD; Barbara Yeung, MPH; Mohamed A. Hammoud, PhD; David Lewis, PhD; David Baker, MBBS; Nila Dharan, MD; Anna M. McNulty, MBBS; Andrew E. Grulich, PhD


J Acquir Immune Defic Syndr. 2021;87(4):1040-1047. 

In This Article

Abstract and Introduction


Background: Preexposure prophylaxis (PrEP) prevents HIV infection but relies on good adherence at times of risk, termed "prevention-effective adherence." Most studies assess adherence without reference to sexual behaviur, making it challenging to determine if poor adherence coincides with HIV risk.

Setting: We examined data from a behavioral substudy of a large-scale PrEP implementation trial in New South Wales, Australia.

Methods: Trial participants completed optional brief quarterly surveys, reporting the number of pills taken and sexual behavior with male partners for each day of the "last full week" before each survey. Condomless sex (CLS) was defined as "higher risk" for HIV when with HIV-positive men with detectable/unknown viral loads or unknown HIV status men. Adequate PrEP protection was defined as ≥4 pills for participants assigned male sex at birth and ≥6 pills for participants assigned female sex at birth (including transgender men).

Results: Of 9596 participants dispensed PrEP, 4401 completed baseline and ≥1 follow-up survey. Participants reported on 12,399 "last full weeks": 7485 weeks (60.4%) involved CLS and 2521 weeks (33.7% of CLS-weeks) involved higher risk CLS. There were 103 weeks in which participants did not have adequate PrEP protection and had higher risk CLS: 4.1% of higher-risk CLS weeks (n = 103/2521), 1.4% of all CLS weeks (n = 103/7485), and 0.8% of all observed weeks (n = 103/12,399).

Conclusions: In a large PrEP trial, prevention-effective adherence to PrEP was very high at 99%. Our findings illustrate the importance of measuring pill-taking and sexual behavior in the same period so that prevention-effective adherence can be better estimated.


Oral HIV preexposure prophylaxis (PrEP) prevents HIV infection in individuals at high risk but relies on good adherence.[1–3] For those taking daily PrEP, 4 or more pills per week are needed for adequate PrEP protection for anal intercourse, whereas 6 or more pills per week are needed for receptive vaginal/front hole intercourse in cisgender women and transgender men.[2,4] "Event-driven" dosing is also effective for cisgender gay, bisexual, and other men who have sex with men (GBM), which requires taking 2 PrEP pills 2 to 24 hours before a sexual encounter and then 1 pill per day for 2 days after the last sexual encounter.[5]

To prevent HIV infection, individuals need to take PrEP pills at times or "seasons"[6] when they are potentially at risk of HIV, a concept termed "prevention-effective adherence."[7] People may take fewer pills, or cease taking PrEP entirely, at times when they believe that they are at no or low risk. A challenge for accurate measurement of PrEP adherence and concomitant HIV risk is that unless pill taking is assessed alongside sexual behavior, it is unknown whether lower pill numbers reflect poor adherence at times of risk or if fewer pills are being taken during periods of low or no risk. Similarly, without first determining the dosing regimen being used by an individual, measures of adherence cannot ascertain whether a low number of used pills represents poor adherence to daily PrEP or effective use of a nondaily dosing regimen.[7,8] Most PrEP adherence studies, whether they use self-reported adherence or objective measures of adherence, such as drug concentrations in biological samples, medication event monitoring system (MEMS) caps, or medication possession ratio (MPR), have not typically compared adherence with sexual behavior in the same period; for example, 2 recent systematic reviews of PrEP adherence studies did not mention timing of sexual behavior in relation to pill taking.[9,10] However, it has been noted that measurement of prevention-effective adherence in research and programs is a challenge.[11]

In the Australian state of New South Wales (NSW), PrEP has been widely available since March 2016 through the large-scale implementation trial, Expanded PrEP Implementation in Communities–New South Wales (EPIC-NSW),[12] followed in April 2018 by availability at low cost through Australia's national program for subsidized medicines, whereby PrEP can be prescribed by any doctor and dispensed at community pharmacies. Analysis has demonstrated continuing low HIV incidence in 9596 EPIC-NSW participants who were dispensed PrEP (0.16 per 100 person-years) across the trial period and up to 12 months after.[13] Additionally, a PrEP-driven 25% decline in state-wide HIV diagnoses among GBM was reported after EPIC-NSW scale-up.[14] HIV incidence among participants with an MPR of 1.0 (ie, no days in which they did not possess PrEP pills) was zero [95% confidence interval (CI) = 0.0 to 0.08 per 100 person-years], and none of the 30 seroconverters identified in the trial appeared to be using daily PrEP at the time of HIV infection.[13] Further analysis of MPR during EPIC-NSW found that one-quarter of participants discontinued PrEP and did not recommence it during follow-up.[15] Additionally, by the final quarter of the study, mean MPR had dropped to 0.6.[15] HIV incidence remained very low, despite the levels of discontinuation increasing and adherence to daily PrEP apparently dropping, suggesting that those who ceased PrEP or appeared to have poor adherence may have stopped PrEP or modified how they took PrEP in line with decreased risk. To explore this, we examined the number of pills taken and sexual activity reported by participants in the same period in the behavioral substudy of EPIC-NSW.