Paying Patients for HIV Control

For the video version, click here.

Improving the public health costs money. Sure, you can create a program that might give you a positive return-on-investment in the long term, but pretty much every program has some up-front cost associated with it. More and more, enterprising researchers are asking whether cutting out the middle-man, and giving this money directly to patients, can be as effective as more traditional interventions.

And that’s the case in this study, appearing in JAMA Internal Medicine: 

The big question – can paying patients with HIV to plug-in to care and keep their viral load suppressed improve outcomes?

Think of this as two separate trials. The first offered patients newly diagnosed with HIV $125 if they successfully got plugged in to the care network. This rate of “linkage to care” was pretty high to begin with – roughly 75% in the 34 sites that were randomized in this study.

As you can see in the table, rates improved over time. But they improved similarly regardless of the presence or absence of financial incentives. No statistically significant effect seen here.

Here’s the thing on that though. Since the baseline rate of linkage to care in both arms was pretty high – my quick and dirty statistics suggest that there would have to be a pretty profound effect of financial incentives – like an increase to 100% linkage to care – to have a positive result here. So consider trial #1 equivocal.

Trial #2 looked at 37 HIV care sites, and here the researchers found that offering a financial incentive –  $70 every three months of being virally suppressed – did significantly improve viral suppression rates.

The effect was modest – an improvement in viral suppression by about 3.8% in the intervention group, but this did meet that statistical significance threshold.

So we have a modest effect here. Let’s do the math to see if it’s worth it.  The researchers doled out 39,359 $70 gift cards – a total of 2.8 million dollars. That led to 366 more virally suppressed people than would have otherwise been suppressed. That works out to roughly $7500 per person. While this may seem a bit shocking – when one considers the current cost of medications, this may be a reasonable approach. Obviously, a formal cost-benefit analysis will need to take into account any costs saved by the reduced rate of HIV transmission and complications in those newly suppressed individuals.

The elephant in the room here though, is that these incentive programs rub some physicians the wrong way. Whether it’s a concern for fraud, or that efficacy will wain over time, or that the same incentive won’t work on individuals of different means, or simply that paying people to do what they should be doing anyway seems wrong. But in the end, studies like this suggest that maybe Adam Smith was right all along.