Pre-exposure prophylaxis ‘would be cost-effective’ if more than 50% efficacious

This article is more than 18 years old. Click here for more recent articles on this topic

One of the concerns expressed about pre-exposure prophylaxis (PrEP) is that it will never be affordable among the populations in which most of the trials are being conducted. The Helsinki Declaration says that “medical research is only justified if there is a reasonable likelihood that the population in which the research is carried out stands to benefit from that research.” Bob Grant of the University of California, San Francisco quoted the declaration before informing the Sixteenth International AIDS Conference that a cost-effectiveness model showed that PrEP based on tenofovir/FTC (Truvada) would be cost-effective if its efficacy was more than 50%, and PrEP using 3TC (lamivudine) could be cost-effective at efficacies as low as 30%.

Grant developed a model taking into account heterosexual and homosexual transmission rates, higher infectiousness in early and late infection, age and sex effects on susceptibility to infection, risk behaviour variation, known age-sex partner preferences, and primary and secondary drug resistance. He also introduced as a variable the degree of ‘condom replacement’ (i.e. people dropping condom use if PrEP became available) that introducing PrEP might cause.

Grant did three analyses based on drug and care costs in low, middle and high income countries but only presented the low-income case, based on Ghana, to the conference.



Cost-effectiveness analyses compare the financial cost of providing health interventions with their health benefit in order to assess whether interventions provide value for money. As well as the cost of providing medical care now, analyses may take into account savings on future health spending (because a person’s health has improved) and the economic contribution a healthy person could make to society.


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


A pill or liquid which looks and tastes exactly like a real drug, but contains no active substance.



statistical significance

Statistical tests are used to judge whether the results of a study could be due to chance and would not be confirmed if the study was repeated. If result is probably not due to chance, the results are ‘statistically significant’. 

His model was based on the likely local costs of, distribution, clinical services, education, counselling, laboratory work and treatment of seroconverters, especially those with resistant virus.

The cost per HIV infection averted of a 30% effective PrEP regimen would be $15,000 and of a 60% effective regimen $5,000, he said.

Taking into account future treatment costs this would mean that, based on the lowest drug prices currently obtainable in Ghana, PrEP regimens using 3TC would be cost-effective if their efficacy was as low as 30%, ones using tenofovir would be cost-effective at 40% efficacy, and ones using tenofovir/FTC cost-effective at 50% efficacy.

Varying the frequency of laboratory monitoring would change the cost-effectiveness: if recipients were only monitored every twelve months, a tenofovir-based regimen would be cost-effective if 35% effective, but it would need to be 50% effective if patients were monitored every three months.

In terms of behaviour change, if PrEP was only 40% effective, then a 50% decrease in condom use among people taking it would reduce its efficacy to zero. If PrEP was 60% effective, its effectiveness would fall to 40% if condom use was halved. But if PrEP was more than 80% effective, ‘condom migration’ would have relatively little negative effect.

Grant did not take everything into account in his model; one audience member pointed out that PrEP costs would rise if continued use led to long-term toxicity amongst users.

During the same session Leigh Peterson from Family Health International also presented the results of their trial of PrEP in three West African countries. Most of these have already been published in this report on Aidsmap. One factor that Peterson emphasized, however, was the extraordinarily high levels of condom use seen among trial participants.

These in turn led to the incidence among the trial population as a whole being half of what was expected, which was one of the reasons why the study did not produce a statistically significant difference between infections in participants taking tenofovir and ones taking placebo. Although six infections occurred in the placebo arm and two on tenofovir, the probability of this result was 0.24, meaning that there was a one-in-four chance that the results were due to random effects.

At baseline, when participants were screened, condom use (as measured by the proportion using condoms last time they had sex) was already fairly high at 52%. But at follow-up three months into the trial, condom use at last sex had increased to 94%; in other words, only one in 20 acts of intercourse were not condom-protected.

Other risk factors fell too: at screening the number of partners the women had had in the last 30 days was 21, at follow-up it was 14. The number of new partners at screening was eleven, and at follow-up was six. But the absolute amount of sex the women had did not change; they were just having it with fewer partners and using more condoms.

Peterson commented that “the one thing we cannot compromise on is the quality of the counselling we offer,” and said that the trial results were as much a testament to the effectiveness of intensive safer-sex counselling as they were to PrEP.


Grant R. Cost-effectiveness analysis of HIV chemoprophylaxis. Sixteenth International AIDS Conference, Toronto, abstract ThLb0102, 2006.

Peterson L. et al. Findings from a double-blind, randomized, placebo-controlled trial of tenofovir disoproxil fumarate (TDF) for prevention of HIV infection in women. Sixteenth International AIDS Conference, Toronto, abstract ThLb0103, 2006.