Who stops taking PrEP, and why?

Kathleen Ryan of Monash University. Photo by Gus Cairns.

Several studies presented at the 10th International AIDS Society Conference on HIV Science (IAS 2019) in Mexico City looked at who discontinued HIV pre-exposure prophylaxis (PrEP) after starting it as part of a demonstration project.

The most consistent finding was that young people have difficulty maintaining PrEP use: in all the studies presented, youth was the single most significant predictor of PrEP discontinuation.

An Australian study showed that, rather worryingly, people’s risk of STIs was just as great after they stopped PrEP as while they were on it, suggesting that they are at as much risk of HIV after stopping PrEP as they would be if they had never started it.


cisgender (cis)

A person whose gender identity and expression matches the biological sex they were assigned when they were born. A cisgender person is not transgender.

pilot study

Small-scale, preliminary study, conducted to evaluate feasibility, time, cost, adverse events, and improve upon the design of a future full-scale research project.


acute infection

The very first few weeks of infection, until the body has created antibodies against the infection. During acute HIV infection, HIV is highly infectious because the virus is multiplying at a very rapid rate. The symptoms of acute HIV infection can include fever, rash, chills, headache, fatigue, nausea, diarrhoea, sore throat, night sweats, appetite loss, mouth ulcers, swollen lymph nodes, muscle and joint aches – all of them symptoms of an acute inflammation (immune reaction).

equivalence trial

A clinical trial which aims to demonstrate that a new treatment is no better or worse than an existing treatment. While the two drugs may have similar results in terms of virological response, the new drug may have fewer side-effects, be cheaper or have other advantages. 


A doctor, nurse or other healthcare professional who is active in looking after patients.

The PrEPX study was the demonstration roll-out undertaken by the state of Victoria, in Australia. It was set up with specific targets in mind: to reduce HIV incidence by 25% in the general population and 30% in gay and bisexual men. It enrolled 4275 people between 26 July 2016 and 31 March 2018, which is 14 times the annual number of diagnoses in Victoria, which had already been falling before PrEPX was started – the highest number of people taking PrEP relative to HIV diagnoses anywhere in the world. They received 16,689 quarterly PrEP prescriptions – an average of 3.9 per person.

The current study analysed the 3489 people who enrolled before October 2017. Some metrics, such as sexual risk behaviour, came only from the 2900 people attending study clinics that were also in the Australian Collaboration for Co-ordinated Enhanced Sentinel Surveillance (ACCESS) programme.

The average age of people starting PrEP in PrEPX was 34, with 25% aged below 29. Nearly all were cisgender gay or bisexual men, apart from 39 transgender women (1.1%). A quarter of participants had taken PrEP before, many in the Vic-PrEP pilot study. Nearly three-quarters (73%) had had condomless anal sex in the three months before enrolling, 13% had used methamphetamine, and 5% had injected drugs.

During the study period, just 85 people (2.4%) officially withdrew from the study, but 877 people discontinued participation without notice (25%). The definition of discontinuation was a gap between the last PrEP prescription and either any subsequent prescription or the end of the study of more than 210 days. Of the 877, 275 (31%, or 7.8% of all participants) never returned after their first prescription.

People who discontinued could restart and in fact 197 people – 22% of those who discontinued, and 5.5% of all participants – restarted PrEP. 

Certain groups of people were more likely to discontinue PrEP. As noted above, young people were especially likely to discontinue, with people under 29 being 75% more likely to discontinue than those aged 40-plus. People who had reported injecting drugs were 64% more likely to discontinue and people who had reported using methamphetamine were 34% more likely. People who were referred for PrEP by their clinician were 27% more likely to discontinue than people who came to the clinic asking for PrEP.

People who consistently used condoms with casual partners (in this case, just the attendees of the ACCESS-participating clinics) were 52% more likely to discontinue PrEP than others.

Discontinuing PrEP did not lead to discontinuation from sexual health services. Of the 743 people from the ACCESS clinics who discontinued, 440 subsequently had an HIV test (59% of those who stopped PrEP) and of these, 158 (34% who had an HIV test) restarted PrEP, meaning that 80% of re-starters did so after having an HIV test.

There were ten HIV diagnoses among people who stopped PrEP, equivalent to an annual incidence of 2.3%. Of these, two people were diagnosed after their second visit and may have had an acute HIV infection when they entered the study. Four others never returned after receiving their first prescription. Of the other four, two had two PrEP prescriptions (one discontinued, then started again) and two had three, but all were diagnosed well after their last prescription ran out – an average of 6.5 months or 199 days afterwards.

Diagnoses of sexually transmitted infections (STIs) were just as common for people who discontinued PrEP as people who kept taking it. There were diagnoses of gonorrhoea in 7.6% of those who remained on PrEP and 8.2% of those who stopped it. In the case of chlamydia the figures were 7.8% and 8.4% respectively, and for syphilis, 1.6% and 2.6%.

Although this does appear to show that taking PrEP does not in itself lead to a rise in STI risk (or rather, that stopping it does not lead to a fall in that risk), it also implies that the behavioural HIV risk for people who discontinue PrEP is no lower than in people who stay on it, and that by discontinuing they are returning to the same degree of risk they had before starting PrEP.

Dr Kathleen Ryan of Monash University, presenting the study, said that it was a concern that people’s “perceived risk may be lower than their actual risk”. Suggestions for improving the drop-out rate included educating people about how to take intermittent PrEP if their risk was also intermittent; informing people better about acute side-effects that might be encountered in the first month on PrEP; and providing more support, including financial support to get to clinics. 

Low rate of PrEP discontinuations in Brazil

Another presentation examined the risks of discontinuation among people participating in Brazil’s national PrEP programme. Since January 2018, 6079 people have enrolled in the national programme. Of these, 5388 attended follow-up appointments and 691 people (11.4%) did not attend – a low drop-out rate compared with some other programmes.

Some types of people were more likely to discontinue than others. As in the Australian study, young people (in this case 18-24) were more than twice as likely to stop PrEP than those in their thirties. Cisgender women were more than twice as likely to stop as gay and bisexual cisgender men.

Homeless people were also at twice the risk of discontinuation. Other groups more likely to stop PrEP included people in the sparsely populated north of Brazil, sex workers, and people refereed to PrEP by their doctors rather than those who came seeking it – as in the Australian study. People who had already taken PrEP before were 30% less likely to discontinue. 


Ryan K et al. Results from a large Australian PrEP demonstration study: discontinuation and subsequent HIV and other sexually transmitted infection risk. Tenth International AIDS Society Conference on HIV Science, Mexico City, abstract MOAD0303, 2019.

View the abstract on the conference website.

Pereira IO et al. Predictors of discontinuation in Brazil's free-of-charge PrEP program. Tenth International AIDS Society Conference on HIV Science, Mexico City, abstract MOAD0304. 2019.

View the abstract on the conference website.