October 25, 2008
There's nothing like hearing the results of studies directly from those who actually conducted the research. In this interview, you'll meet one of these impressive HIV researchers and read his explanation of the study he presented at ICAAC/IDSA 2008.
My name is Benjamin Linas. I'm an instructor in medicine at Harvard Medical School and I work at Massachusetts General Hospital in the division of infectious diseases.
Benjamin Linas, M.D.
This is a project looking at eligibility criteria for state AIDS Drug Assistance Programs [ADAPs].1 ADAPs provide ART [antiretroviral therapy] to patients who have no other access to care. Historically, many ADAPs have not been able to meet all the demand for services due to budget constraints. In the past, when ADAPs have been forced to restrict care, they usually open up a wait list on a first-come, first-served basis. Any patient who is clinically eligible for ART is eligible for the program. If there's not enough room, people went to the wait list as they applied to the program.
We started working with the Massachusetts program a couple of years ago to plan for [the possibility that] if Massachusetts should need to have a wait list -- which, fortunately, it has not needed to date -- how they should do that, on the theory that maybe we'd have more success prioritizing patients based on their stage of disease as opposed to on a first-come, first-served basis.
For this project, we used the administrative data set from the Massachusetts ADAP and we built a mathematical model, or simulation, of an ADAP program. The model would go to the ADAP data set and create hypothetical patients that were based on real people with an age and a sex and a CD4 count. The model would check and say, based on your CD4 count, whether you are eligible for this program. If your CD4 was above the eligibility, then you would enter a simulation [in which you have HIV but are] not receiving ART. Your CD4 would fall and you would have a risk of opportunistic infections. When your CD4 reached the threshold level, then you could come back and apply to the program.
If your CD4 was below the threshold, then the model would see if there was room for you in the program. If there wasn't, then you had to enter a wait list of care until room became available and then you would move into the program.
Then people in the program moved through a simulation of their disease with ART, in which they had suppressed viral loads and their CD4s went up instead of down.
Doing that, we tracked what the outcomes would be over the course of five years.
What we found was that strategies that prioritize people based on their CD4 count consistently had better outcomes than the strategy that put people on a wait list on a first-come, first-served basis. There was a lower rate of mortality, there was a lower rate of first opportunistic infections and death in every permutation that we tried: if we had more demand; if we had lower CD4 counts in the applicant pool; more men; more women; faster rates of disease; slower rates of disease. Consistently, the CD4-based strategy had better outcomes. It [achieved this] by minimizing the time that people with very low CD4s waited for ART.
I think the take-home message from this is that the best way to improve outcomes is to fund your ADAP at 100% of capacity so you can provide ART to everybody who should be on ART. Fortunately, right now, ADAPs are able to do that, although there is some concern for the immediate future, with the current fiscal situation and budget cuts coming.
If programs are at the point where they're forced to open a wait list, then I think that they can have better outcomes if they go with the approach of prioritizing people based on their CD4 count instead of relying on this tried-and-true first-come, first-served approach.
Why isn't this just good medical sense? If the people most in dire need for it should get it, it just makes sense.
I agree. I think it makes medical sense, and it's a paradigm that's used in other resource allocation settings. For example, livers for transplant are definitely allocated based on need and stage of disease.
I think there are two reasons why it hasn't been done thus far: First, there's an administrative cost to doing this: Somebody's going to need to track CD4s. Programs are going to have to have validated, good mechanisms for collecting that information, which they don't have now. That is a cost and could be difficult.
I think there is, on the surface, a certain fairness to first-come, first-served. You show up and, if there's room, we'll take you, and if there's not, then you'll have to wait just like everybody else who has to wait.
But the truth is that you're not doing the most with your resources that you could be doing. Further, people from historically disadvantaged groups -- people of color and uninsured people -- often present to care later. While this project doesn't directly look at this in any sort of quantitative sense, one can imagine that people who are going to present to care later and have to wait longer are people who might be from disadvantaged groups historically, so I think that there's potentially an equity outcome to this as well.
Plus, people have died waiting on these lists in the South.
People have. Fortunately, as of today, there's no wait list [in the South]. The reason is because a few people did die on wait lists in the South, and that prompted some emergency legislation to [provide] stopgap funding. Any state that had a wait list on the day that legislation passed received excess funds.
That, in combination with Medicare Part D, has helped to unload some of the burden on ADAPs. But I think that, unfortunately, everyone's nervous about the coming couple of months.
Is your study being taken in consideration by other ADAPs that are in much more dire need?
I don't know. I can't say. I hope so. That's our goal.
We did it in conjunction with Massachusetts. Massachusetts feels that, should it need to open a wait list, it's going to go to a CD4-based approach as we demonstrated. I've talked with Rhode Island in the past, although it's been a little while since I've spoken to them, so I don't know what their plans are.
I hope that people will see this paper and that it will have some impact. But that's always the goal when you do policy research.
Has this paper been published?
It's currently in manuscript form, but it's not been published yet.
This transcript has been lightly edited for clarity.