A total of 2,499 men at high risk of HIV infection participated in the study, which was conducted at 11 sites in Brazil, Ecuador, Peru, South Africa, Thailand and the United States. Half of study participants received the PrEP pill, while the other half were given the placebo.
In all, 64 HIV infections were recorded among the 1,248 study participants chosen at random to receive the dummy pill, while 36 HIV infections were recorded among the 1,251 participants who got the drug.
These are the last two paragraphs from an article from CTV News, the Canadian Press.
I've also cut and pasted their headline and used it as my own. The headline indicates that there is a risk of HIV and the drug "cuts" that risk for gay men. What the last paragraph shows is that less gay men in the drug group of the trial contracted HIV than in the placebo group. The "risk" in the headline is about the risk of being infected by the virus. This shouldn't be confused with the risk of dying of AIDS because you are HIV positive.
A true test would be unethical. 2500 HIV negative people are split in two groups just as they were in this trial (placebo vs drug). They are allowed only to have sex with men from a group of HIV positive men. The sexual encounters must be done without condoms.
The truth about the trial however is that the men were allowed to randomly have sex as they saw fit. Condom use was not measured. This issue appears to have been dealt with by providing both groups with safe sex education. The measurement of whether or not each participant took his daily pill was done on the honor system. Daily use was not true to the design of experiment but dealt with after the fact to the advantage of the pharmaceutical company paying for the trial.
And I could go on. What was the diet between groups. What were their incomes. Where did they go for sex? Then you take just the ones who contracted HIV. The list of variables is beyond our comprehension. Yet a fair trial would be unethical. Science is up against a whole lot when it comes to this kind of research. In the end all you have are two groups. In one, 5% get HIV. In the next, 3% get HIV. Surprising?