‘Near perfect’ adherence in early stages of Partners PrEP study

This article is more than 11 years old. Click here for more recent articles on this topic

Very high levels of adherence have been achieved in one of the ongoing randomised controlled trials of oral pre-exposure prophylaxis (PrEP), according to a poster presentation at last month’s Conference on Retroviruses and Opportunistic Infections (CROI 2011).

According to investigators in the Partners PrEP study, adherence levels of 99% were achieved by participants during an average four-month period during the two-year study.

Partners PrEP is a study amongst 4700 heterosexual couples of differing (serodiscordant) HIV status at nine sites in Kenya and Uganda. The HIV-negative member of each couple will be randomised into three groups to take a daily pill containing tenofovir, or tenofovir plus FTC, or a placebo. All pills will look identical. At the end of the study HIV incidence rates in the three trial arms will be compared. Results are expected by early 2013.



How well something works (in a research study). See also ‘effectiveness’.


A product (such as a gel or cream) that is being tested in HIV prevention research. It could be applied topically to genital surfaces to prevent or reduce the transmission of HIV during sexual intercourse. Microbicides might also take other forms, including films, suppositories, and slow-releasing sponges or vaginal rings.


Refers to the mouth, for example a medicine taken by mouth.

intent to treat analysis

All participants in a clinical trial are included in the final analysis, in the groups they were originally assigned to, whether or not they actually completed their course of treatment. This method provides a better estimate of the real-world effect of a treatment than an ‘on treatment’ analysis.

opportunistic infection (OI)

An infection that occurs more frequently or is more severe in people with weakened immune systems, such as people with low CD4 counts, than in people with healthy immune systems. Opportunistic infections common in people with advanced HIV disease include Pneumocystis jiroveci pneumonia; Kaposi sarcoma; cryptosporidiosis; histoplasmosis; other parasitic, viral, and fungal infections; and some types of cancer. 

Achieving high adherence to the study medication in trials of new biomedical prevention methods such as microbicides and oral PrEP has been a challenge in studies so far, even those achieving a successful result.

Conventional methods of adherence monitoring such as self-report have proved to be unreliable, as drug level monitoring (in the iPrEx PrEP trial) and the use of microbicide applicators sensitive to vaginal mucus (in the Carraguard microbicide trial) have shown that participants’ actual use of trial interventions is considerably lower than their reported use.

This not only compromises the potential efficacy of the intervention, it makes it very difficult to interpret trial data, as ‘intent-to-treat’ efficacy (based on which arm participants were allocated to) will be very different from ‘as-treated’ efficacy (based on whether they actually used the treatment or not), and this does not help to answer the question of what adherence and efficacy would be in ‘real world’ situations.

To try to get round this, investigators from the University of Washington, Seattle, who are co-ordinating the Partners PrEP study, used three different ways of measuring adherence:

  • Clinic-based pill counts, based on subtracting the number of pills actually prescribed from the number that would have been if every pill had been taken:
  • MEMS (Medical Event Monitoring System) caps, an electronic device which sends a radio signal to clinic monitors when a pill bottle is opened
  • Unannounced pill counts, in which trial assistants visited participants at random times once every month and counted the number of pills they had at home. There were an average 3.3 visits per trial participant.

None of the adherence measures excluded the possibility that participants are simply throwing their pills away rather than taking them, but the MEMS and home pill count measures at least means they are not being left in unopened bottles.

One in seven trial participants were recruited at an ancillary adherence study of whom 544 (11.6%) of participants contributed adherence data to this study. Thirty-six per cent entered the adherence study at the time they started the main study while others entered it at varied times during the study. Their adherence so far has been followed for a median of four months.

During this time, adherence as measured by all three measures exceeded 99% - it was 99.6% by clinic count, 99.1% by unannounced pill count, and 101.9% by MEMS. This ‘over 100%’ figure means that either participants are taking more than the prescribed dose or that they are opening medicine bottles without taking doses at times.

Adherence below 80% was found in 35 participants (6.4%) by unannounced pill count. The only characteristic related to poor adherence was youth: poor adherence was 10% less likely for every ten years older in participants.

Adherence as measured by clinic count and self-report in the iPrEx trial was 93%, yet a substudy of drug levels showed that only 50% of participants were actually swallowing their pills. Although the present study cannot predict what participants’ adherence will actually turn out to be when Partners PrEP reports, its additional two measures and very high reported levels so far suggest adherence may turn out to be better than in iPrEx.


Haberer J et al. Near perfect early adherence to antiretroviral pre-exposure prophylaxis (PrEP) against HIV infection among HIV-serodiscordant couples as determined by multiple measures: preliminary data from the Partners PrEP study. Eighteenth Conference on Retroviruses and Opportunistic Infections, Boston. Abstract 488. 2011. See here for conference poster.