February 27, 2007
|Listen (4MB MP3, 9.5 min.)|
Over the last decade, antiretroviral therapy has produced striking survival gains in the United States, yet not everyone has shared equally in those gains. We know that HIV-infected patients from underserved populations in the United States receive suboptimal access to care. They are more likely to be late in starting therapy, and more likely to discontinue care prematurely. Imperfect access to care translates into shorter survival.
The goal of our study was to assess just how big the survival losses are. We examined how gender, ethnic and racial disparities in initiation and discontinuation of HIV treatment are reflected in differential years of life lost. We used the CPAC model of HIV disease funded by the National Institute of Allergy and Infectious Diseases to assess survival.
We found that an estimated 16 years of life are lost per person on average due to HIV in the United States, for patients who receive guideline (U.S. Treatment)-concordant care. An additional five years are lost per person due to late presentation and premature discontinuation of treatment. Worse still, these losses are up to 35% higher for racial minorities, and are greatest for women. The loss is greatest for Hispanic women, whose average survival is a full 6.4 years less than it would be if they were to receive care that conforms to national guidelines. White women had the lowest rate of late initiation and premature discontinuation. Even so, they have an additional 3.7 years of life lost.
We have made great strides in HIV care in the United States. The survival gap between HIV infected and normal life expectancy continues to narrow, but there is more work to do today. Survival losses due to suboptimal access to care are entirely preventable. We can and should aim to achieve early presentation and retention in care for all HIV-infected persons, with particular focus on women with ethnic and racial minority groups.
These findings of our study present true evidence in support of expanded routine voluntary HIV testing as a practical intervention to ensure early diagnosis, and reduction in disparities in HIV care in the United States.
Journalist: You said that there were five years of additional loss of life due to late start and premature discontinuation of ART [antiretroviral therapy]?
Sixteen years of life lost due to HIV disease -- it's human guideline-concordant care.
Journalist: That's HIV disease, irrespective of any other ...
It's compared to HIV-uninfected individuals of the same demographic composition.
Journalist: Given that, at least racially, there are some disparities of health care in the HIV-negative general population, do these sort of add to that? Are they in addition to that, or can they be reflecting pre-existing disparities?
I think these are maybe reflective of general disparities. It has been shown in several studies that racial and ethnic disparities are seen in cardiovascular care, [and] in overall medical care in the United States. So I think these are just the findings that parallel this in HIV care. But in the case of HIV, and in the case of cardiovascular care, there is a treatment that is available that can extend life, and that's why it's really imperative that we would take strides to make those treatments available for vulnerable populations.
Journalist: Can you comment on how this is specifically affecting the Hispanic population in the United States? And women as a subset?
The Hispanic population, especially women, is a group which has the highest rates of discontinuation of treatment, and they tend to present to care late, very late. In fact, about 50% of Hispanic women are presenting to care with low CD4+ [cell] counts. That's why the disparities, or the losses in life expectancies, are highest for that group of patients. I would like to add that Hispanics are the only ethnic group in the United States where the HIV epidemic is growing. So now it's about 20% of HIV-infected patients [that are] are Hispanic. So that's a group which really deserves special care.
Journalist: It's an eye opener when we realize that the median CD4+ cell count at which people initiate treatment in the United States, in general, is quite low.
Yes. I think the median was below 200.
Moderator: I think that, in general, there's a lot of work that needs to be done to get people earlier into care. But I think the message is, in some subpopulations where it's even more dramatic. There are lots of points of intervention potentially. Yes?
Journalist: Do females in the United States tend to be identified with infection earlier than males, in part because of prenatal screening?
Not all of them. White women are more likely to be identified early, but women from racial and ethnic minorities are not necessarily being identified early. Even if they identify, they [are] not necessarily linked to care, because identification does not necessarily mean that they will be treated in time.
Moderator: The issue is that they may be coming in late in HIV, pregnant. So I think the way to get around it is much more widespread availability of HIV testing. Yes.
Journalist: Just help me to understand what we're saying here. It's that if you start treatment promptly with these patients, you can predict that you are going to save five years of life.
Not quite. What we did is, we tried to estimate the life expectancy for those who initiate treatment on time, and stay on treatment, [patients] who are using all the regimens available. Because some regimens fail, but there is a next line [of treatment] that is available. So for those who stay perfectly on treatment, the life expectancy is much higher. So the life expectancy of HIV- uninfected people is 44 years. The life expectancy of people who are staying on guidelines-concordant care is 28 years. The life expectancy that we estimated, using actual access to care, is 23 years.
So these are the incremental differences. The first difference is 44 minus 28; it's due to HIV -- that's 16 years. The second -- 28 minus 23 - it's five years, which are due to suboptimal access to care.
Journalist: Is there any indication that you can recover any of these years? I understand we're talking about a group of people, not individuals. But is there any indication that treatment improves, [i.e.,] lowers, that risk?
That's a very good question. Obviously, our analysis did not take into account that in the future, hopefully, the treatment will have a better efficacy, and potentially may recover. But as of now, treatment actually was a very high efficacy for several lines of therapy. So potentially, yes; it may recover. But I think if we continue with the current disparities, those differences still will be present.
Moderator: When I think of it simplistically, there are probably four interventions one can do. Number one is, prevent HIV -- because, clearly, if you want people to have the life expectancy of everybody else, HIV plays a major factor, I would say. These populations are, of course, at higher risk. There's a disparity in terms of just the numbers of new infections in the U.S. occurring in these same populations. Then there's identifying people early with HIV, and then providing them with access to appropriate treatment and then keeping them on appropriate treatment for the long run. So I think of it as almost like a four-fold intervention.
I agree. Given the amount of money that we put into development of new drugs and improving the efficacy, which is absolutely necessary, there are very, very few interventions. I did a very thorough search on how to improve the retention to care. If a small proportion of money could be put into improved retention as case management, or many home health, home care visits, potentially the gap of five years may be decreased substantially.
To view this study abstract, click here.