http://www.economist.com/science/displaystory.cfm?story_id=11893615

If money is the sinews of war, as Cicero suggested, then it is always good to have America on your side. America bankrolled the second world war, and it looks as though it is going to bankroll the war on AIDS, too. On July 30th George Bush signed an act reauthorising PEPFAR, the President's Emergency Plan for AIDS Relief. It provides for $39 billion to be spent on AIDS over the next five years, up from $15 billion for the past five. It was a welcome backdrop to the XVIIth International AIDS Conference, held this week in Mexico. Mr Bush has been on the receiving end of a lot of flak from AIDS activists for imposing conditions on the way PEPFAR's budget is disbursed—preferring chastity to condoms, for example—but, overall, he has been a force for good in the field of AIDS. PEPFAR was his own idea and, as the old saying has it, he who pays the piper calls the tune. In that context $39 billion is a positive symphony.

There is also good news elsewhere. The latest report from UNAIDS, the United Nations agency charged with combating the disease, was released on July 29th. It suggests that the death rate from AIDS is now falling. It peaked in 2005 at 2.2m a year. Now it is down to 2m, thanks to the torrent of money that is providing patients in poorer parts of the world with drugs that keep HIV, the virus that causes the disease, under control (see chart). Of course there are critics, too. The aim was to provide such drugs to all who need them by 2010. This was always going to be a tall order. The target has not been formally abandoned. But Stephen Lewis, a former UN special envoy for HIV/AIDS in Africa, says that a close reading of the report suggests its authors no longer believe in the deadline. Mr Lewis reckons the activists are being softened up for a slippage to 2015.

That would be a disappointment. And there have been several of those since the last conference, two years ago in Toronto. Then the talk was of new forms of prevention—in particular, of vaginal microbicides that would stop women getting infected; of the suppression of genital herpes, which is thought to help HIV get into the body; and even of pressing the good old contraceptive diaphragm into use to block the virus. But three trials of microbicides have now failed, and so have two herpes-suppression trials and one trial of diaphragms. As for vaccines, the shimmering hope on the horizon of every AIDS conference, the one big trial to report has more than just failed; for many of the participants the vaccine seemed to make things worse. The consensus among vaccine researchers now seems to be that future trials should be abandoned and the money saved spent on basic science.

On top of all this gloom, it is still the case that nobody has the faintest clue how to cure someone once they have been infected with HIV. The drugs, welcome as they are, merely stave off the immune-system crash that opens the victim to lethal secondary infections such as tuberculosis. That means you have to stay on them for life. As a result, taxpayers are accumulating an indefinite—and indefinitely growing—responsibility for keeping people alive. Somehow, somebody has to work out how to stop the disease spreading.

The pill or the snip?

Oddly, one possible way of making a big dent in the problem has been sitting under people's noses for years. Anti-retrovirals, or ARVs, as the drugs used to treat AIDS are known, work by stopping HIV from breeding. They do not cure, because there are places in the body that the virus can hide from them. But while someone is taking the drugs, the virus almost disappears from his blood stream—and, crucially, his seminal fluid. Moreover, if the infected individual is a she, rather than a he, the drugs have a tendency to accumulate in the tissue of the vagina.

Those observations are leading in three directions. The first is the idea of pre-exposure prophylaxis—in other words, giving ARVs to uninfected people to clobber any new infection before it can get going. Two trials of this approach will report next year.
The second is to use ARVs as microbicides. The unsuccessful microbicide trials used either long, complicated molecules intended to act as a physical barrier to the viruses or detergents to disrupt them. ARVs in the vagina might stop them breeding before they get into the body.

The third approach, though, is the most intriguing. This is to do nothing more than press ahead faster with the treatment programme. Since treatment reduces viral load, it should, in theory, make those being treated less infectious. Of course, theory is one thing and practice another. But studies in Taiwan and British Columbia (the latter by Julio Montaner, the incoming president of the International AIDS Society, which organises the conference) have shown big falls in transmission rates as ARVs have been rolled out. Meanwhile, Myron Cohen, of the University of North Carolina, Chapel Hill, is beginning a clinical trial involving 1,750 "discordant" couples (those in which one partner is infected and the other not) to see if ARVs really do reduce transmission. This trial will not report for seven years, but Dr Montaner hopes the mere likelihood that treatment also prevents transmission will spur governments to redouble their efforts to roll out ARVs and thus obtain a prevention programme at no additional cost.

It was this sort of careful science—starting with a scientific hypothesis, following it up with observations in the field and ending with clinical trials—which proved that circumcision protects against infection. Indeed, amid the gloom about microbicides and vaccines, circumcision is the one bright spot in the field of AIDS prevention. Most forms of prevention have to be pushed on to people. But there are already parts of Africa, including South Africa, Swaziland and Zambia, where men are queuing up to have their foreskins cut off. The scientific basis of all this is that foreskin tissue is rich in a particular sort of cell that HIV likes very much. The field observation is that, within Africa, one of the two best predictors of the intensity of the epidemic in any given place is the prevalence of circumcision. The clinical trials suggest that circumcision by itself reduces a man's chance of becoming infected by 50-60%. The upshot, according to Brian Williams of the World Health Organisation, is that if, in some ideal world, every sexually active man in sub-Saharan Africa were circumcised, 2m new infections would be avoided over the course of ten years, and 300,000 deaths prevented.

The demand for circumcision seems to have caught everybody off guard. Some surprising people disapprove of the whole idea. Yoweri Museveni, the president of Uganda, who is normally at the forefront of the anti-AIDS movement, has spoken out against it on the grounds that it might encourage men to become more promiscuous (although that does not seem to have been true of participants in clinical trials). And members of some groups who traditionally circumcise boys or teenagers as a mark of group membership seem equivocal about the idea of outsiders adopting the practice. But the main constraint is the lack of enough people trained to wield the knife cleanly and safely. Indeed, Tachi Yamada, the head of the Gates Foundation's Global Health Programme, suggests bringing in volunteers from the rich world on a temporary basis, to clear the backlog.

The other predictor of the scale of the epidemic in African countries, besides rates of circumcision, requires a rather different approach. It is the rate of what are known, delicately, as concurrent partnerships. That is, how many lovers the average individual has at any one time. In this case, changing people's behaviour involves pushing very hard indeed.

Every home should have one

Behavioural change is, in some ways, prevention's orphan. The chiding voice saying "use a condom" or, worse, "don't have sex with anyone other than your regular partner", is not what most people want to hear. But, according to Helene Gayle, co-chairman of the Global HIV Prevention Working Group, which has been investigating the matter, no national epidemic has been brought under control without it.

One reason for the orphan status of behaviour change is that it is seen as slightly unscientific. Dr Gayle and her colleagues are trying to change that. They have been looking at research in the field and discovered a lot of trials that attempt to borrow the methodology of medicine by using control groups. These trials quantify everyday experience and, in the round, suggest that campaigns to promote condom use, to curb promiscuity and to promote the use by drug takers of clean needles can reduce infection rates in the groups they are aimed at by 20-30%. Not, perhaps, as impressive as foreskin-amputation, but certainly a worthwhile improvement.

Peter Piot, the head of UNAIDS, thinks that much more could be done in this area. Social science, even when it emphasises the "science" over the "social", can only achieve so much. Sometimes, you just have to bring in the professionals. To that end, he and his colleagues have already had informal talks with marketing experts from Procter & Gamble about getting the message over in a subtle and effective manner. Marketing people also understand that markets vary, and tailor their message accordingly.

That is something that the AIDS establishment has been bad at until recently. Many countries whose epidemics have been concentrated among gay men or injecting drug users, or where prostitutes have been the centres of infection, have preferred to pretend such people did not exist and have spent money preaching to heterosexuals who are at low risk. Conversely, in places where most infected people are part of the wider population, blind eyes have been turned to the reality of the way the disease spreads heterosexually. Some countries have played down the question of concurrent partnerships, a particular cause of trouble in parts of Africa. There has also been a widespread assumption that, among married couples, the promiscuous partner is likely to be the man. Studies of discordant couples show that to be so, but much less than had been assumed. Though the data are limited, they suggest that in about a third of discordant African couples it is the woman, not the man, who is infected.

The other thing people who work on AIDS have been bad about is championing this or that prevention strategy over the rest. The pro-condom versus pro-abstinence argument is a good example of that. But if one lesson has come out of the meeting, it is that you need every weapon you can lay your hands on.

It is particularly odd that this was not noticed before, since the tremendous success of ARVs comes from using several of them in combination. The watchword now is "combination prevention" to go along with combination therapy. If AIDS is to be beaten back without a vaccine—which is still a big if—that is the only sensible way forward. Wars, however well financed, are won by disciplined bodies of troops, not rabbles.