Economics of HIV Testing

4/12/2013 04:24:00 PM
Harold Pollack has a post on HIV prevention efforts in the US. One of the takeaways is that efforts to reduce HIV infections in the population of men who have sex with men (MSM) have stalled. It is not immediately clear why. Pollack points out a few reasons:
  • Social stigma prevents MSM from getting tested. Testing leads to treatment and safer sex practices, both of  which dramatically reduce infection rates
  • Unlike injection drug users who basically died out, the MSM population "demographically replenishes itself, with large numbers of individuals at risk for infection every year."
  • Low awareness about the need for condom use among MSM
I have also completed a study suggesting that the recent up-tick in HIV infections was possibly due in part to the introduction of geo-social networking apps targeted at gay men, such as Grindr, beginning in 2009. I found suggestive evidence that Grindr increased the propensity of unprotected anal sex among New York City MSM by around 1 to 2 percentage points.

That said, I take issue with Pollack's last point, since data from the Behavioral Risk Factor Surveillance System, the Community Health Survey, and (if I'm not mistaken) the National Health and Nutrition Examination Survey all suggest that MSM are more likely to use condoms during penetrative sex than straight couples. The rate for straight couples is only around 35% compared to close to 50% for MSM. That's still much lower levels of condom use than we'd like to see (surely those aren't all monogamous long-term partnerships), but nevertheless, apparently MSM have a much higher transmission rate in spite of their safer sex practices.

I do, however, agree with the general point that we need to overcome the social stigma of getting tested by making HIV testing a standard recommended procedure for all adults, regardless of risk factors like sexual orientation. Pollack mentions that around 20,000 americans die each year from HIV. It is not immediately clear what fraction of those would have been preventable with early diagnosis and treatment, but modern anti-retro-viral therapies are effective enough that most HIV related deaths can be prevented with early enough treatment. Lets suppose for the moment that they are all preventable with treatment. A recent survey found that on average HIV tests costs around \$48.07 for both the test and related counseling. The Census puts the 2012 adult population at around 240 million. That means that the cost of testing everyone is about \$11.5 billion. Furthermore, the annual cost of treatment is roughly  \$19,912 per patient, meaning it would have cost roughly $400 million annually to treat all 20,000 who die each year from HIV, largely due to their failure to get tested. So according to that math (which admittedly requires some dubious assumptions) the cost of annually testing everyone and treating people who would not otherwise have been diagnosed is about \$12 billion. By contrast, even the lowest possible estimate of the value saving 20,000 lives is \$60 billion (market-based estimates of the value of a life rage from \$3 million to \$7 million).

There are probably better ways to slice that data. But from my back-of-the-envelope calculation, encouraging universal annual HIV testing for all adults would generate net benefits of at least \$48 billion per year.
Szew 4/15/2013 02:35:00 PM
I would suggest caution when advocating population wide testing. Don't forget that medical tests can have false positives. That means that if you do population wide testing for a disease that occurs in a relatively small percentage of a population it can easily turn out that a significant portion of the people that test positive do not carry the disease. For instance lets assume some disease occurs in 10% of the population but there's a test that returns a positive 5% of the time when the subject does not have the disease but always returns a positive if the subject has the disease.(it's a really good test) This means 14.5 percent will test positive. So nearly 1/3 of those that test positive do not have the disease. This is not to say that population wide testing is a bad idea, but that if it is done there should most likely be multiple rounds of testing before anyone considers medical procedures to address the results.
Matthew Martin 4/15/2013 02:56:00 PM
You have a good point. Incidentally, while HIV tests are considerably more accurate than your example, it actually is the case that about 1/3 of the positives are false positives (based on a study I read almost 5 years ago, so maybe things have changed).
Matthew Martin 4/16/2013 07:29:00 PM
It occurs to me that I didn't really complete my thought. So, the risk of false positives isn't a huge deal because we can simply run a second test for all the ones that come out positive the first time. Since fewer than 1% of the tests will be positive, the second test does not significantly affect my cost calculation. Between the two rounds of tests, each individual's probability of inaccurate results are effectively zero.
Anonymous 8/07/2013 05:23:00 AM
Hi, I work in a communication company I want to share my testimony to the world, my name is maria katrina, i am from united kingdom in belfast north ireland, I was a stripper in a club I got effected with HIV due to the nature of my job, In april 15 2013 i was tested positive to HIV, This is not design to convince you but its just a personal health experience . I never taught doctor hakim could ever get my HIV-AIDS cured with his healing herb spell, i have tried almost everything but I couldn't find any solution on my disease, despite all these happening to me, i always spend a lot to buy a HIV drugs from hospital and taking some several medications but no relieve, until one day i was just browsing on the internet when i came across a great post of !Michelle! who truly said that she was been diagnose with HIV and was healed that very week through the help of this great powerful healing spell doctor ,I wonder why he is called the great papa hakim, i never knew it was all because of the great and perfect work that he has been doing that is causing all this. so I quickly contacted him, and he ask me some few questions and so i did all the things he asked me to do,He ask me to buy some herbs and which I did for my cure,only to see that at the very day which he said i will be healed, all the strength that has left me before rush back and i becomes very strong and healthy, this disease almost kills my life all because of me, so i went to hospital to give the final test to the disease and the doctor said i am HIV negative, i am very amazed and happy about the healing doctor hakim gave to me from the ancient part of africa, you can email him now for your own healing too on his email: thank you doctor for healing me from HIV. I am free and happy. here is my mobile number to really confirm from me +44702407844 and here is my

Matthew Martin 8/07/2013 07:31:00 AM
This is sad. Without aggressive anti-retroviral therapies, HIV kills in about 9 to 11 years. Moreover, this is not just a matter of personal choice: without anti-retroviral drugs, your probability of infecting someone else is much higher.

On a side note, HIV tests are usually administered twice, because they are calibrated so that around a third of all positives are actually false positives. Though there is a slim probability of a false negative, if the second HIV test is negative, it means you never had HIV in the first place.