The New HIV Drug

(Photo: Babak Fakhamzadeh)

An FDA panel just approved the first drug recommended for preventing infection by, rather than limiting the effects of the HIV virus.  Part of the discussion by panel members was classic economics, expressing concerns that the drug’s availability would reduce people’s willingness to take as much care, in particular that it might reduce condom use.  

The same issue has been mentioned and analyzed in various economic studies, including old ones about the effects of mandating car seat-belt use on automobile accidents, and about the impact of sex education on teenage sexual activity and pregnancy.  Any insurance or safety measure generates a moral hazard; the important issue is the net effect on the outcome of interest — in this case, HIV infection.


Currently treatment as prevention is hugely expensive, well beyond the reach of most individuals, but even if it reduces in price (unlikely given the potential returns for pharma) as the fidelity ring social experiment showed, you cannot control people's behaviour, but if you can help protect people against infection (with whatever means) then I believe there is a moral imperative to do so.
That's my personal view, but I have two questions: 1) How does the cost to the consumer affect the moral hazard? 2) How will this affect health insurance policies?


* p.s. There is substantial empirical evidence that people will have risky sex whether there is a drug or not: there are many people who, with full knowledge of the risks, will still expose themselves. Moreover, there is growing evidence that in certain communities people are increasingly exposing themselves which has led to a concerning rise in other STI cases (e.g. syphilis outbreaks). This is a concern not just for the individual but also for the sexual population as some of these infections can reduce the efficacy of ART.
With 96% efficacy the drug can prevent against acquisition of infection at an individual level, the question for public health policy then becomes: what is the effectiveness of the drug at a population level? As a recent opinion article states, the answer to this question is unclear and depends among other things, on the ability to retain people in care. In any case, it will need to be combined with other measures to prevent against infection. From an ethical perspective though: is it justifiable to deny access to a drug (once approved as safe) with such a high proven efficacy? Secondly, is it not better to use all weapons to prevent infection?
The same fears were voiced over male circumcision but there is a chance that adequate counseling accompanying any prescription can help reduce disinhibition.



*correction - 96% efficacy for early ART usage (this is not what the FDA is approving, though currently they recommend early ART usage as possible)


It's important to understand two things that a lot of coverage of this story have left out:

1. This isn't a new drug at all. This is a current treatment for HIV infection with a new approved use for reducing the infection rate among certain categories of high-risk populations (e.g., partners of HIV-infected persons).

2. Much of HIV transmission occurs at a point well before a person knows he or she is infected, making this a relatively ineffective preventative measure in general populations.

Seat belts are general prevention. Sex ed is general prevention. This is an effective preventative measure in only limited circumstances. These comparisons only compound the problem of the generally bad reporting around this story.


2) Much of HIV transmission occurs at a point well before a person knows he or she is infected, making this a relatively ineffective preventative measure in general

While the first part of this response is true, your conclusion is misleading. the idea would not be to use it in isolation from other preventative measures but in combination ( I refer you to an article on this in current opinion in HIV/AIDS: Treatment as prevention: translating efficacy trial results to population effectiveness). There is no panacea for HIV transmission prevention until a vaccine, but the contribution of treatment as prevention could be substantial given its high efficacy at reducing acquision (96%).


Actually sorry I want to revise my response that the first part was true: where is your evidence that much of transmission occurs before a person knows she is infected?

david holtz

I am a gynecologic oncologist and face this argument with HPV vaccination. Granted, the risk of "risky behavior" are orders of magnitude less catestrophic with HPV, but all data I've seen so far indicate most people are not driven to have sex or use condoms by fear of disease. At what point is a hazzard large enough to drive behavior?


Fear of disease was a substantial contributor to safer sex behavior globally especially in SSA where the high mortality rate meant that people were attending funerals almost on a weekly basis. It is easy to forget that for a long time HIV was a death sentence. Thankfully though, better treatment with fewer side-effects mean that this is no longer the case (at least in the West).
There is evidence though, that people are starting to become desensitized and that effective treatment which has made HIV a chronic infection in the West could have contributed to desensitization.
Surely this means that HIV counseling should be improved? Also how does counseling affect the moral hazard?

Alex Blaze

I find the moral hazard argument fairly regressive without any numbers - couldn't one argue that condoms are a moral hazard because people will then think they can have sex? Clearly, condoms are part of the solution, and that's because the amount of sex they encourage (there are probably some people who would abstain if condoms didn't exist) times the chance of getting HIV even with regular condom usage is orders of magnitude smaller than the number of infections condom usage prevents.

But that's only because we have a fair ballpark of the numbers. This pill is prohibitively expensive at $11K/year; the drug only reduced infection rates by 42%, much less than condoms do; and even if it were cheap not everyone would take the pill daily. On the plus side, the fact that it'll be so expensive will prevent widespread education on it, so the moral hazard will stay pretty low.


I don't really see how this creates a moral hazard.

First, as mentioned Truvada has been on the market for a while now. While the FDA approved it for dealing with post exposure issues (either supression or prophylaxis) the approval for usage to be a sort of pre-exposure prophylaxis isn't that much of a sea change. Off label perscriptions are fairly common, and it seems that this was a popular reason for prescribing Truvada.

The only reason that the mfg would really see approval for this indication would (as PrEP) would be so that the mfg can market it as a pre-exposure preventative. Which doesn't really change that much, in reality.

Second, being a pre-exposure issue, I'm not sure how its availability will generate much more risky behavior. For the new approved use to even matter, the individual has to take the drug before engaging in the high risk sexual encounter. That means they are actively taking the initiative to mitigate the risk, which seems antithetical to the idea of a moral hazard.

The only way I can really see this as being a moral hazard issue, is if it increases transmission and high-risk sex because a partner may lie about being on Truvada. Even then I'm not sure how truvada actually works, and it may be necessary to take the drug personally to prevent contracting the disease (as opposed to taking the drug to prevent transmission from yourself).



Interestingly, using Truvada for prevention only reduced transmission by 44%. If people rely on this in lieu of using condoms, transmission could actually go UP. Truvada is an addition to, not replacement for, traditional safe sex methods – that’s the real story.

Babak Fakhamzadeh

Such an excellent choice for a photo!