Five themes emerged during data analysis that have implications for setting clinical protocols and informing future public health programming regarding PrEP. See for a summary of our results. The results and illustrative quotes are described in more detail below.
1) Little Consensus on PrEP Target Populations
There was no real consensus about who should receive PrEP, and often accessibility to the prophylactic medication depended on where patients sought out their care, in private or public clinical settings. Most providers agreed that serodiscordant couples were seen as ideal candidates for PrEP. Providers in public clinics felt it should be only given to those at highest risk, and only if it could be at low to no cost; those in private practice said they would prescribe for any patient that wanted PrEP and whose insurance would cover the costs. One San Francisco primary care physician (PCP) in private practice felt it was appropriate to prescribe it to the “worried well,” those gay men who, even after being assessed as low risk through counseling, may still experience significant worry about their risk for HIV infection. In contrast, those at sites where Truvada® was going to be provided at low to no cost to patients felt strongly that a protocol to determine eligibility for PrEP by risk for HIV acquisition should be put in place:
“This is going to be such a limited resource, that we want to make sure that it’s not necessarily going to all the worried well… I think some of the categories like sero-discordant couples, clearly, would be in a higher-risk category, and the rectal infection history, and PEP use in the past. And receptive unprotected anal intercourse. We’d probably figure out what the risk stratification would be, … and then what we’d consider sort of lower-risk that probably wouldn’t’ qualify for PrEP.” (Michelle, SF provider, Community/STI Clinic).
Other physicians, particularly those who were HIV specialists and had experience treating patients who had developed resistant strains, thought that in addition to risk assessment, a provider’s assessment on who might be adherent was an important eligibility consideration.
You kind of have to be in that sweet spot where I think you’re risky enough to merit taking a pill every day, but not so risky that I think I’m gonna give you this and undertreat your HIV that you’re about to get tomorrow. (Shawn, Oakland provider, HIV Specialist).
2) Current Models of Care and Skill Sets were not Always Well Suited for Prescribing PrEP
Most providers felt that PrEP is best provided in primary care settings by providers who are comfortable working with gay men and TG women, able to discuss sexual behaviors in a non-stigmatizing manner, and are informed about HIV. Most agreed that it was important that PrEP be offered in neutral clinical locations, i.e. those whose primary purpose was not HIV treatment. Practitioners that saw positive and negative patients felt they were at an advantage for being able to provide PrEP.
In my office stigma is not so much an issue because of my blended practice. And I do get patients who do not want to go back to the AIDS Center, they do not want to go to [local clinic] … where the clinics are kind of segregated. (Bill, Oakland provider, PCP).
However, most providers reported that their current models of care were going to need to change to accommodate the needs of patients on PrEP, particularly with respect to adherence counseling. For those working in clinics that primarily did STI and HIV testing, this involved switching to a more longitudinal model of care.
“[W]e are not used to having people that come back for check-ins on a regular basis… [With PrEP], we’re responsible for monitoring someone, and to make sure their kidney function is okay. That really does move into the realm of primary care. …[W]e’re going to have to do a lot of training with our own providers to make them more comfortable with doing some of that.” (Geeta, SF provider, Community/STI Clinic).
For those in HIV clinics, providing PrEP would entail additional training to provide preventive care to uninfected individuals. One San Francisco HIV provider had serious reservations about his clinic becoming a place that offers PrEP services:
[O]ur approach has always been to be a primary care center for people with HIV, and it would be an easy thing for us to extend that to people at risk for HIV, and assessing and prescribing, monitoring PrEP in that context.[But] I think it would be very difficult for us to wear the other hat of sort of a focused intervention where we would have like a PrEP clinic, … that would be a significant departure from our model of care. (John, SF provider, HIV specialist).
In community-based clinics, irregular access to a medical doctor for lab monitoring and follow up was an additional concern: “it kind of depends on whether it was something where there’s a protocol set up, where a registered nurse could provide – furnish this … we wouldn’t be able to operate it if RN’s were excluded from providing.” (Lawrence, SF provider, Community/STI Clinic).
3) Providers Expressed a Need to Build Capacity to Prescribe PrEP– training, Referrals, and Establishing Reimbursement Levels for Care and Drugs
All providers noted a need to increase their clinics’ operational capacity to provide PrEP. Although low demand for PrEP was noted among the different providers’ client populations currently, all anticipated that increasing demand for the intervention would necessitate the development of screening and eligibility protocols, clarifying insurance reimbursement rates, training existing staff, and perhaps hiring additional staff. As one provider noted, “If we wanted our medical assistants or anyone to provide PrEP, they would require some counseling training.” (James, SF provider, Community/STI Clinic).
Public clinics with limited resources would have a more challenging time developing the billing capacity, training, and staffing infrastructure necessary to provide PrEP to their patients, and providers worried that PrEP would amplify current disparities between the public and private health systems. One provider noted:
I think this [PrEP] is a tool for individual patients in individual circumstances. And clinics that have favorable circumstances as well, to use - kind of like – what’s the effect of bone marrow transplant gonna be on mortality rates. Not high, but if you’re one of the people who needs a bone marrow transplant, it’s awesome, and it’s there. PrEP is like that. It’s likely to exacerbate disparities rather than improve them. It’s definitely not a disparity-reducing type of intervention, because it requires a lot of infrastructure and a lot of readiness, on the part of both the patient and the clinic, to do it. (Shawn, Oakland, HIV Specialist).
Similarly, among providers seeing patients without insurance, other support services needed to be identified and in place before doctors would consider prescribing PrEP.
We take care of people who have histories of substance abuse, homelessness, mental illness, poverty, so getting them to be able to do something like this would require [a] belief in its importance by the providers who are struggling to get people to take meds for normal things like hypertension, diabetes, and schizophrenia. So it would require some supports, both around education, around adherence, and also financially, because we have as many uninsured as we do – we’d have to have access to meds and I’m sure it’s expensive. (Carla, Oakland provider, Community/STI Clinic).
Public health providers also recognized the need for a community-focused education component to let people know about PrEP and its availability. This would require balancing strategies for publicizing PrEP availability widely with the need to communicate that the intervention is not intended for everyone. One idea was to try to work with social networks and popular opinion leaders to get the word out about PrEP, particularly to African American communities which are disproportionately impacted by the epidemic.
We’ve done some work, thinking about what’s the right sort of community education part that needs to happen, who are the right kind of providers that it would be important, good to engage, what’s the right… opinion leader who has social clout…You also need to be able to educate and make this – the extent to which you can look at how to normalize some of the seeking of these types of services, or not at least have them be highly stigmatized, and how the fact that they’re related to sex, or HIV, or something… we don’t want to just put billboards up everywhere, and have everyone come and flood the system with people who are not at risk and really don’t need this intervention. (Mary, LA provider, Public Health Official).
4) Concerns About Monitoring Adherence, Side Effects and Toxicities, Resistance, and Risk Compensation Among PrEP Patients
Adherence. Providers noted that monitoring adherence would be a challenge. Current adherence monitoring practices for HIV-positive patients, such as monitoring refills, interviewing and counseling about adherence, would need to be extended to those patients on PrEP. Several providers noted the additional burden that would create on themselves and their staff if PrEP demand were to increase dramatically. In one practice, it was envisioned that a dedicated case worker would provide the intensive adherence counseling and monitoring of PrEP patients.
Aside from demands on staff time, providers whose patient populations included young MSM and TG women expressed the need for more tailored adherence counseling to address these groups’ particular adherence challenges:
I think a lot of young people tend to have less stable schedules… they wake up at different times during the day, they eat at different times during the day, they skip meals, their lives are just less regimented and structured and stable, and I think because of that, it’s easy to forget meds, it’s easy to - oh, I slept at my boyfriend’s house last night, so I didn’t bring my pills. Plus, they’re just less - I guess reliable in that way. They may not have their medications with them. They’re not taking them at the same time every day. They have this shame around it so they don’t want to show their friends they’re taking medication. (Karen, Oakland provider, Community/STI Clinic).
Providers in private practice, whose patient populations were better educated and privately insured gay men, felt that episodic use of Truvada® was likely to be common among their patients. “They might take it just 3 days before an event, if they know that they’re going to have a party or something special,” (Wayne, LA provider, PCP). This provider felt that providers may not be able to prevent this type of intermittent use and would need to develop adherence counseling that took this practice into account.
Side effects and toxicities
Most providers felt that Truvada® was “generally well tolerated” and had minimal concerns about side effects. However, all providers recognized the need to carefully monitor PrEP patients for toxicities. During our discussion of Truvada® side effects, one Oakland PCP who had experience prescribing Truvada® to HIV-positive patients recalled a recent case where a patient who had been taking Truvada® for years for HIV had recently experienced acute and chronic renal failure. Another provider in Los Angeles expressed concern about the interpretation of the iPrEx trial results with respect to toxicities in particular sub-populations of patients:
I think the fairly benign toxicity profile seen in iPrEx was at first reassuring, and then on reflection, in light of the adherence data, depressing. In that this is the toxicity that we saw during one year of treatment for people who basically took the drug less than half the time. What would the toxicity profile look like for somebody who is taking it actually according to the way it was prescribed, for perhaps a longer period of time? What would happen if we saw a less baseline healthy population, with perhaps more predisposition to renal dysfunction, like African-Americans in general? (Hector, LA provider, HIV specialist).
Resistance and frequency of HIV testing
Some providers were concerned about the development of viral resistance in the event that a patient seroconverted while taking PrEP. Overall, most providers felt that quarterly follow-up with PrEP patients would be required, although some felt that monthly follow-up was required, and others that semi-annually was sufficient. Providers whose patient populations included young MSM, subustance-using MSM, and MSM of color noted that these populations generally had difficulty keeping medical appointments and might require special care in any patient tracking protocol.
This is a pretty young MSM of color, very transient, kind of unstable population, who kind of slip in and out of care, who generally are not in care anyway. Those would be the people I would be most likely to give PrEP to, the party-hard, young MSM of color, who have high high-risk exposures and are just not able to cut down. The concern with resistance would be, let’s say if they get the PrEP, and they know how to get refills, and they fail to show up every six months for their HIV test. And then they’re still taking it, they’re still taking the PrEP off and on, not knowing they have HIV infection and not coming back to get their HIV tests. (Karen, Oakland provider, Community/STI Clinic).
None of the providers we interviewed were concerned about risk compensation by patients on PrEP. Overall, most providers tried to remain non-judgmental and pragmatic, while helping their patients have satisfying, yet safe, sex lives.
I think that’s kind of paternalistic, and I try to tell my patients, these are all the strategies we have. My goal is really to make sure that people feel comfortable with that information, and that they use it in the way they can use it. I try very, very hard not to even change my facial expression when people tell me about really, really risky behaviors that they’re engaged in, and be more helpful in terms of how people navigate their lives with the tools that I can make available to them, like medications. (Bill, Oakland, PCP).
5. Providers Believed in the Public Health Benefits of PrEP, Even in Light of the HPTN 052 Results
The results of the HPTN 052 trial were announced just before we went into the field for this study, and it was something that several clinicians brought up during interviews. That trial found that among serodiscordant couples where the positive partner was virally suppressed, there was a 96 percent reduction in HIV transmission to the uninfected partner 
. Given that public resources for PrEP will be limited and that HPTN 052 also showed efficacy in preventing HIV acquisition, it is important to understand provider perspectives on both biomedical prevention interventions 
. In all interviews, we found that providers saw merits in each approach, and that both should be considered essential aspects of a comprehensive HIV prevention program. One clinician researcher in San Francisco noted because some participants in HPTN 052 did become infected during the course of the study, the results of PrEP and HPTN 052 are best viewed as complementary and synergistic, and that additional modeling research could inform public health approaches that effectively utilize both interventions.
[W]e know that there’s a sizable proportion of infections that are still occurring within partnerships, and so – and that happened within the 052 study as well, that there are a number of unlinked infections that occurred. And so only treating the positive partner isn’t going to eliminate all the infections, and so finding the right balance between treatment and PrEP I think is important to have as a target. (Michael, San Francisco provider, Clinician Researcher).
Given the high prevalence of undiagnosed HIV infection in urban populations 
, one public health official described targeting PrEP towards higher-risk individuals as adding value to current efforts to control the epidemic through a test-and-treat approach:
I think we’ve all recognized that if we could have an impact, there’s a real need and responsibility to really focus on making sure that we get everyone we possibly can who is living with HIV, to have an undetectable level, from a prevention standpoint… Looking at our surveillance data for 2009, 35 and 40 percent of people who are cases for L.A. County had no CD4 viral load count at all in the surveillance system, meaning they did not access care at all. And so when we know we’re dealing with that, and then we think we have about 21 percent undiagnosed, if we apply the CDC estimates to our local population. We have this pool of virus…Given that that is the setting, to the extent to which we can really focus PrEP in on the riskiest group, the tip of the iceberg, the ones that really are having just a lot of unprotected sex and a lot of risk, I think that it’s probably still needed. (Mary, Los Angeles provider, Public Health Official).
This concern was echoed by providers working with particularly hard-hit populations, such as African American MSM 
, who saw PrEP as an essential ingredient to breaking the cycle of HIV infection.
So if we’re aggressively treating the people who have the illness – and even a half – three-quarters – I’m sure there’s mathematical modeling that would tell us what it is – but half or three-quarters of the people at highest risk, are on – half the time, taking PrEP, then that may be the breaking point at which, when you do mess up and don’t take your pill, the person that you’re sleeping with is not going to be virally productive on that day. So there’s a point at which it starts to have an impact over a community of people. Because that’s really the way this works, is that the reason why certain communities are so impacted is because people all have sex with each other… But that’s why you can have 46 percent of the African-American gay men infected, is because there’s that hothouse concentration. And that means that if you could get enough of just that one population protected, then you could break that cycle within that community, potentially. (Jessica, Oakland provider, HIV Specialist).