February 5, 2008
There's nothing like hearing the results of studies directly from those who actually conducted the research. It is these women and men who are transforming HIV treatment and care. In this interview, you'll meet one of these impressive HIV researchers and read an explanation of the study she is presenting at CROI 2008. Accompanying me on this interview is Dr. Gerald Pierone, a clinician and researcher who is the director of the AIDS Research and Treatment Center of the Treasure Coast in Fort Pierce, Florida.
Lisa Hirschhorn, M.D., M.P.H.
I'm Lisa Hirschhorn, M.D., M.P.H., and I'm with JSI [John Snow, Inc.] Research and Training at Harvard Medical School. This was a study that was funded by HRSA [Health Resources and Services Administration] through the SPNS [Special Projects of National Significance] program to explore whether you could actually replicate a positive prevention intervention [for HIV-infected individuals], which had been proven in a clinical trial at a single site, into real-life practice at 15 Ryan White CARE Act-supported sites.1
We collaborated with the original developers from the University of Connecticut and took their research protocol and the research instruments they had, and from that [we] developed a training and a manual that could then be rolled out to sites. Fifteen sites applied and were given small amounts of funding to help them implement the intervention. We did one centralized training, and then people were provided with resources to train other providers at their site and to implement what's called the "options intervention," which is a brief, clinician-based positive prevention intervention that's based on the whole idea of giving people information and then motivating them to decide what they want to do to either maintain low risk or decrease active risk.
Bonnie Goldman: What kind of information do they give?
Lisa Hirschhorn: They go through a risk assessment. We ended up with seven steps. The first thing [the clinicians] do is assess what kind of risk behavior [their patients are] actually having right now. Then they negotiate with the patient which area they want to discuss for risk behavior, either for reduction or maintenance of low risk. They focus then on the areas that might put somebody at risk. "When do you not use condoms? When you're drinking or using drugs?" Then they talk about if [the patient wants] to change this or maintain it. "How important is it to you?" The patient rates it from one to 10. "How confident are you that you can actually make the change?"
Based on that -- if [the patient says], for example, "I think it's only a three," -- they say, "Why is it a three? Why do you care at all? Why is it important at all? How can we make this more important for you?" Then they discuss strategies of how to reduce the specific risk or to maintain the low risk. Like, "What are you going to do the next time that you drink?"
They agree upon a goal, which is a progressive goal. So if you're never using condoms, could you use them some of the time? Then, in the next appointment, they follow up on the discussion and it's documented in what's called a patient record form so that it becomes a permanent part of the record.
What we found is that 13 out of the 15 clinics were able to implement it, and 90% of the patients who reported having the risk discussion after they had an option visit said most of them thought it was very positive. They thought that providers' comfort and knowledge seemed to improve. The median [discussion took] about five minutes, which was much less than people thought it was going to be to go through the whole thing. The providers selected, to some extent, who was going to be able to get [the intervention]. For example, they avoided people who had very significant mental illness, such as psychoses; people who were only coming in with a crisis; who were really very actively high. It had to be people who were engaging in primary care. This was really meant to be an intervention that capitalized on the relationship between a patient and their primary care provider.
Bonnie Goldman: Who typically did this intervention work?
Lisa Hirschhorn: It would have to be whoever the primary care provider was: the doctor, the nurse practitioner or the physician's assistant. One of the sites already had in place some very good prevention educators. They actually implemented it using the prevention educators, which was not within the context of this [study]. It was this balance between fidelity and adaptation; they adapted it more than sticking with the fidelity.
There was a lot of debate whether that was an OK thing to do. You could argue on two different levels, one of which is that it's the person you have the closest relationship to. But because this was meant to be a repetitive, chronic-care management approach, the feeling was that it was more appropriate to have it with somebody with whom you were going to have repeated, ongoing contacts.
Bonnie Goldman: Could you tell me a little bit about the demographics?
Lisa Hirschhorn: Sure. We, like the original study, had very, very few people who were self-reported injection drug users, and even fewer who had active drug use. About half of the people self-reported [that they were] MSM [men who have sex with men] as a risk factor, and a significant number of others were heterosexual. About a quarter of the people had less than a high-school graduation. Most of the people had been [HIV] positive for quite a long time: The median was about seven years, although the range was anywhere from less than a year to as high as 28 years by self-report. Very high rates of active mental illness: 42%.
What was interesting was, qualitatively, we went and talked to the providers, and even the patients that had been in care with the provider for many years. When they went through this discussion, there were all kinds of risk behaviors that [the providers] hadn't known were going on. Some of [the patients] said, "Oh, I'm so glad that we're talking about sex now." The providers clearly had been giving positive prevention messages all along. Somehow, it wasn't quite getting through.
The other thing that was fascinating was that, because this forces you to talk about strategies to reduce or maintain low risk, a number of the patients who reported abstinence were not abstinent by choice. [They] actually had medical conditions that they had been too embarrassed to talk about. So it invoked these discussions, and a number of patients had medical evaluations when [the providers] uncovered some medical problems about why people were having erectile dysfunction. It was interesting not only in the sense of opening up a discussion for positive prevention, but also probably improving the discussion that they were having with the patients in general.
Bonnie Goldman: It's kind of challenging, though, for most clinicians to find the time to do this. Even if it just takes five minutes, they often don't have the time. How feasible is this, really?
Lisa Hirschhorn: This was one of the main concerns. I think there was a fair amount of initial resistance, but when we went to the clinics, it really did not disrupt the ongoing clinic. You have to have at least 20 minutes -- ideally, a little bit more -- for a visit. If you have a 12-minute visit, you're not going to have time to do this.
But again, it has to be a priority, and given the CDC [U.S. Centers for Disease Control and Prevention] recommendations2 and issues about ongoing transmission, the feeling is this has to become a priority. The clinicians really felt like it improved the communication that they had with their patients overall.
You definitely have to undergo a certain amount of resistance, but we talked to the administrators as well as the clinicians, and people felt it really hadn't disrupted the patient flow. One of the key things is that you don't implement it all at once to every patient. The idea is that you do it for one or two patients at each session, because the first interaction takes longer than the follow-up ones. The idea is, you start off and you do two or three at each session, preferably a little bit towards the beginning, and then over time you actually end up interacting with all the patients.
Gerald Pierone: Hi, I'm Gerry Pierone from AIDS Research and Treatment Center of the Treasure Coast. You mentioned overcoming the initial resistance. How long did it actually take to get the clinicians, the administrative staff and the support staff of these clinics on board and trained to do this?
Lisa Hirschhorn: It's a little bit difficult to assess that because we had other delays. For example, IRB [institutional review board] approvals getting the training. Once training went into place, it didn't take very long, mostly because each site -- with the exception of one site -- had a very dedicated leader who came in. They started doing it, usually with one other person, and then they would do training and would spread out from the clinic.
One of the keys to success -- this is one of the reasons why one of the sites didn't work -- was you had to have somebody who was committed and who felt that this was going to be done. In general, they wanted it to be a physician. A couple of the sites were able to do it with nurse practitioners, but it had to be somebody with enough leadership and authority that they could say, "This is what's going to happen."
At some of the sites, it took a couple months for some of the providers to buy [into] it. Even some of the most resistant ones, by the end of it, really thought that it was very helpful. The majority of the sites are continuing to do it in some manner or another.
Bonnie Goldman: The 50% of sexually active patients who reported that they used condoms, was that considered low or high? And was that before the intervention?
Lisa Hirschhorn: That was considered to be low, because most of these were sites where they were already doing some sort of positive prevention. I think that people were very surprised that it was low. Everybody made the assumption that it was actually higher, although there was some serosorting going on in terms of condom use.
Gerald Pierone: From what you're describing, it seems that no matter what, positive things came out of these interventions: Medical conditions were uncovered, dialogue occurred which hadn't been happening, and clinicians became aware of their patients' sexual activities when perhaps they weren't before. So even if outcomes weren't particularly changed, those outcomes, from a clinician's standpoint, were positive in impact.
Lisa Hirschhorn: That was our impression. Obviously, it's qualitative. One of the limitations of the study was that it took 18 months to be proven effective in the research setting, and because of issues of delay with IRB approval, and also the duration of funding that was available. We were only able to assess [the intervention] at three months, although even at three months we began to see changes in terms of sexual risk reduction or maintaining low-risk behavior.
Bonnie Goldman: Could you talk a little bit about the three-month impact?
Lisa Hirschhorn: First, we said, "How many of these people actually achieved their risk reduction goal?" It was 68% -- two thirds of people, even at three months, were able to achieve their goal. Eighty-five percent either decreased risk or maintained low or no risk. It was interesting: One of the [most challenging groups of] people were the people who were having low or no risk. How do you actually discuss that with patients in terms that encourage them to continue to maintain that? [Clinicians] felt like this helped them do that.
We saw [a] slightly better [response] in terms of maintaining low risk or increasing condom use among [people with] HIV-negative or unknown-status partners than among [people with] HIV-positive partners. We saw a trend toward the decrease of injection drug use to get high, particularly in the context of sexual behavior. We had very, very little injection drug use, but fairly substantial amounts of other substances without a prescription or alcohol: it's 55%.
We did some exit interviews of patients. They reported that provider knowledge of sexual activities and recreational drugs improved, even though this was a group of patients who had fairly high baseline knowledge, as did the providers. Somehow, the communication, in some subtle ways, seemed to get better as well.
Gerald Pierone: Can you see taking this research one step further, where you could show a lower rate of sexually transmitted diseases among people, lower rates of hospitalization for psychiatric comorbidity, etc.? Perhaps better control of viral replication?
Lisa Hirschhorn: I think the sexually transmitted infections are going to be tough because the rates are very, very low [in this study]. We do some quality work in some other states and we found the same sort of thing, probably because some patients go and get treated elsewhere. In most other studies, it's been very difficult to show a decline in that. In terms of other biological outcomes, I think it would be interesting to see whether there is any impact in terms of adherence -- as a result, for example, of a decrease in recreational drug use.
Mental illness is an interesting question. Clearly, one of the main barriers that people felt was this very high rate of active mental illness, and some of the limitations in terms of getting adequate treatment for it. The providers did not do this [intervention] for patients who had very, very severe mental illness, people that were actively psychotic. You had to be grounded a little bit in reality to undergo this motivational behavior change, so I don't know if the highest-risk patients are going to necessarily be the ones for whom this is appropriate. Some of them aren't even showing up to get general medical care. So this is really meant to be much more within the context of patients that are coming into care. How do you actually make them safer in terms of what they do?
Female Researcher: Can I ask a question? I'm a clinician at Whitman-Walker Clinic [in Washington, D.C.]. Your interventions -- is there a pivotal question here that the nurse can ask?
Lisa Hirschhorn: It was actually developed more as something that is part of the conversation that [patients are] having with the primary care provider. It wasn't a risk assessment, which is what's next door to you, where the doctor says, "Oh, you told the nurse this and that." It forces the primary care provider -- doctor, nurse practitioner -- to actually have that discussion. [Patients] felt that was actually one of the things that helped, because, first of all, it validated [what they were discussing]. "Oh, it's important, my doctor, my nurse practitioner is talking about it." And it also allowed them to continue on as a conversation of, "Oh, so it says you're doing this and that." We felt that it was much more of a building conversation.
Female Researcher: I think I find some screening questions usually [required a] 20-minute appointment, and most of the time there is some other priority, though it's on the list. If someone was asking me -- before they get to us -- some screening questions, then you can focus on the area that needs to be. You don't tell them, "This is what you told the nurse." But you already identified by just glancing at it, "This is a risk area that I probably should spend more time on." Time is the ultimate factor.
Lisa Hirschhorn: I agree. That's why if it's less than 20 minutes, you can't do it. But the feeling was, for some patients it's going to need to be done every time, for other ones not necessarily. Although in the original protocol, it was. It's not a lot of time, and it is something that has major impact.
Which area you focus on is a negotiation. As opposed to saying, "Me as your doctor, this is what we're going to talk about," you say, "There's a bunch of different areas here. Which do you want to talk about?" Even though it may not be the one that you think is the most important, it's [part] of this whole idea that it's a gradual change. It's information, motivation, and then behavior change over time, as opposed to a one-time, "Here's a problem, let's fix it today," and then move on.
It's much more based in a chronic care management approach. The same way you would do with diabetes. It would be something that you would say, to get their A1c [glycated hemoglobin] down is going to be a long process. The idea is to look at this as a change that can take 18 months to get somebody there.
Female Researcher: How many visits?
Lisa Hirschhorn: Talking to the providers, it seemed if somebody was low risk, particularly if they were maintaining abstinence and they were abstinent by choice, that probably once or twice, and then yearly, was the general feeling about it. Otherwise you're beating somebody over the head when they're maintaining it. It's just a question of, "Have things changed? Is this a struggle for you?" Things like that. But for people that were actively engaging in higher-risk behavior, for them it would be something that would be a constant, ongoing conversation.
Female Researcher: Were the providers surprised by how much they learned in this training that you had?
Lisa Hirschhorn: The qualitative feedback was that they learned a lot. Even though some of them were very experienced in terms of positive prevention and things like that. How to talk about it was something that they thought was very helpful. And the patients reported that their knowledge got better.
Bonnie Goldman: Did you provide actual language for how people should talk to their patients about this?
Lisa Hirschhorn: Yes. Some of the training was language and, "When people say this, how do you respond to it?"
Bonnie Goldman: Real practice sessions?
Lisa Hirschhorn: Right, yes.
Gerald Pierone: Are you planning a Web site that we can direct readers to?
Lisa Hirschhorn: There's careacttarget.org, which is the HRSA Web site. University of Connecticut, I think, is going to have a Web site -- it used to be up, but it's come down, and will hopefully go back up again. But the information should be available through the CARE Act Target site. [You can also click here to view the study's Web site.]
Gerald Pierone: So a clinic could go to careacttarget.org and download this information?
Lisa Hirschhorn: Yes. It's not up yet, but it's going to be. There will be a manual.
Bonnie Goldman: Thank you very much.
This transcript has been lightly edited for clarity.