When considering PrEP for men who have sex with men in the UK, the most relevant data come from the PROUD study, which was conducted in England and reported its results in 2015. (See PROUD study graphic.) Men who took part in the PROUD study were at higher risk of HIV infection than many other gay men. They frequently reported multiple partners, inconsistent or irregular condom use, recreational drug use, sexually transmitted infections (STIs) and use of post-exposure prophylaxis (PEP). However, any man who does not always use a condom with male partners was eligible to take part.

In the control group of men not receiving PrEP, there were 20 HIV infections. In the group of men offered PrEP, there were 3 HIV infections, each in a man who was probably not actually taking PrEP at the time.

Comparing the two groups, PrEP reduced the number of infections by 86% (credible range:  64-96%). This surpasses the real-life effectiveness of consistent condom use. One HIV infection was prevented for every 13 men given PrEP.

A wide range of results have been reported in other PrEP studies, with some finding high effectiveness and some none at all. (See table Key PrEP studies.) The crucial factor determining effectiveness is adherence, in other words whether people actually use PrEP regularly, without missing too many doses.

In studies in which many participants had poor adherence, PrEP had no benefit at all. This was the case in several studies with young women in African countries.

But where adherence has generally been good, PrEP has been shown to be effective. This includes studies with men who have sex with men and with serodiscordant couples in east Africa. Even in studies whose headline findings were that PrEP was ineffective, individuals who used PrEP regularly had greater protection.

If programmes recruit individuals who are motivated to take PrEP and help them with adherence, PrEP is likely to be highly effective.

When HIV-positive people start HIV treatment, it reduces infections to heterosexual HIV-negative partners by 96%. Other than this, a reduction of HIV infections by 86% surpasses that of most other HIV prevention interventions that have been tested in randomised controlled trials, many of which have failed to demonstrate any benefit. Moreover, the evidence of effectiveness for many behavioural interventions (such as groupwork and social marketing campaigns) relates mostly to short-term changes in sexual behaviour which do not necessarily lead to reductions in infections. When HIV incidence has been assessed, behavioural interventions have on average reduced infections by 46%. 

Condom use has not been tested in randomised control trials. However, observational studies suggest that people who say they consistently use condoms have around 80% fewer HIV infections (in heterosexuals) or around 70% fewer infections (in men who have sex with men) than people who never use them.

Protective levels of Truvada are usually reached in rectal tissue and blood after between four and seven daily doses. These results are most relevant for gay men. Because of the lower concentrations of tenofovir in the vagina and cervix, it may take PrEP longer to protect women, perhaps requiring three weeks of daily doses.

While several studies have found oral PrEP to be effective for men who have sex with men, studies offering PrEP to heterosexual men and women have had mixed results. All these trials were conducted in African countries, with some testing vaginal gels rather than pills. (See table Key PrEP studies.) 

In three studies, heterosexual men and women given PrEP had between 39 and 75% fewer HIV infections. But in three others, women given PrEP had as many HIV infections as women receiving a placebo.

Individuals who used PrEP regularly were more likely to remain HIV negative than others.

But adherence has been very poor in some trials conducted in African countries. There were social barriers to taking PrEP, including women’s position in society, personal relationships, HIV stigma and ambivalence about the research process. The studies with the most disappointing results were done with young, mostly single women. Good adherence was achieved in a study recruiting heterosexual couples in which one partner was living with HIV.

This does not mean that PrEP can never be an appropriate prevention technology for women. The circumstances of heterosexual men and women in the UK who need to protect themselves from HIV are likely to be different from those of the people who took part in the African studies. PrEP may be a viable option for some heterosexuals in the UK.

But there may also be biological factors which could make PrEP less effective for women. Researchers have found that after a single dose of PrEP, concentrations of tenofovir are much lower in the vagina and cervix than they are in the rectum. The implication would be that women may need to maintain near-perfect adherence to have protection against HIV during vaginal sex, whereas a lower level of adherence may be protective during anal sex. More research on this topic is needed.

Recommendations on providing PrEP to heterosexual men and women are included in American guidelines. For a serodiscordant couple aiming to conceive a child, PrEP may be used alongside HIV treatment for the positive partner.