Comfort discussing HIV pre-exposure prophylaxis with patients among physicians in an urban emergency department

Dear Letter to the Editor

Pre-exposure prophylaxis (PrEP), containing tenofovir disoproxil fumarate/emtricitabine, is effective to prevent HIV.^1–7 One million adults are estimated to be in need of PrEP and engaging providers that can link at-risk persons into the PrEP continuum of care remains a priority.^8–11 Emergency department (ED) physicians encounter at-risk individuals presenting an opportunity for PrEP education and linkage to care.^12–15 Studies of PrEP awareness have focused on infectious diseases, HIV and primary care physicians.^8,16–20 We assessed awareness and comfort in discussing PrEP with at-risk individuals among ED physicians in Missouri.

From February through March 2017, we conducted an anonymous online survey of physicians in the Division of Emergency Medicine at Washington University in St. Louis (See Supplemental Digital Content). Participants were recruited through the department listserve using Qualtrics software (Qualtrics, Provo, UT). Survey development was informed by prior studies regarding provider PrEP awareness.^16,21,22 At the time survey, the department had 88 physicians who were faculty, fellows and residents, and there had been no department PrEP education sessions. The study was approved by the Washington University in St. Louis Institutional Review Board.

The 26-item questionnaire included physician demographics (i.e., level of training and years having practiced medicine since obtaining a medical degree) and seven topic domains: comfort in discussing HIV risk factors and PrEP with patients, awareness of PrEP, concerns with prescribing PrEP, perceived barriers to discussing PrEP with patients, educational training interest, training modality preference, and department support for training. The majority of question responses used a 5-point Likert scale: 1) strongly agree, 2) somewhat agree, 3) neither agree nor disagree, 4) somewhat disagree, and 5) strongly disagree and were used to create a binary measure of agree (1-2) versus did not agree (3-5) for the analysis. The primary outcome was comfort in discussing PrEP with patients. Comfort was defined as “strongly agree” or “somewhat agree.”

We assessed having an interest in PrEP educational training using a 3-point scale: 1) not interested, 2) moderately interested, and 3) very interested and defined (1) as “not interested” versus (2-3) as “interested.” Training modality preferences included: 1) in-person training, 2) webinar, 3) self-guided online training, 4) written materials, and 5) other. Perceptions of having department support for PrEP training used a 3-point scale: 1) not supportive, 2) moderately supportive, and 3) very supportive and defined (1) as “no support” versus (2-3) as “support.”

Bivariate analyses were performed between the primary outcome and awareness- and concern-related factors. Multivariable logistic regression was performed to determine awareness- and concern-related predictors of comfort discussing PrEP with patients. Predictors that were significantly associated with comfort discussing PrEP in bivariate analyses were incorporated into the final regression model. Odds ratios (OR) and 95% confidence intervals (CI) are reported. All statistical tests were two-sided and the significance level was set at 0.05. Statistical analyses were performed using SPSS version 24 (IBM Corporation, Chicago, IL).

Of 88 adult-trained emergency department physicians contacted, 76.0% completed the survey and were included in analyses. Many (64.2%) were faculty and the median years of practicing medicine were 15 years (interquartile range 10–25 years).

Most (79.1%) had previously heard of PrEP to prevent HIV infection, but few (23.9%) reported awareness of the current Centers for Disease Control and Prevention (CDC) PrEP prescribing guidelines⁷ and only 34.3% were familiar with any current research regarding PrEP safety and efficacy. Providers reported being comfortable discussing intravenous drug use (100.0%) and sexual practices (95.5%) with their patients. However, under half (43.3%) of respondents reported being comfortable discussing PrEP with their patients. Twenty-three percent were aware of local PrEP care referral information for their patients.

Concerns with PrEP prescribing included potential medication side effects (89.6%) and the selection for antiretroviral resistance (70.1%). Many (67.2%) reported that non-PrEP strategies should be attempted prior to PrEP initiation and that PrEP prescribing would lead to riskier sexual practices (47.8%). Over half (53.7%) of the providers were concerned that PrEP is not effective in reducing HIV transmission.

In bivariate analyses, the comfort of providers discussing PrEP with their patients was associated with awareness of guidelines (P < 0.001),⁷ familiarity with current research on PrEP safety and efficacy (P < 0.01), awareness of local care referral information (P < 0.001), and the lack of a concern that PrEP was not effective (P < 0.01).

When adjusting for the concern that PrEP is not effective, awareness of the guidelines (OR: 5.49; 95% CI: 1.14-26.48) and awareness of care referral information (OR: 6.54; 95% CI: 1.45-29.56) were significantly associated with comfort in discussing PrEP with patients (Table 1).

Table 1:

Multivariable regression analysis of emergency department providers’ comfort discussing HIV pre-exposure prophylaxis (PrEP) with patients (N=67)

Variable	OR (95% CI)	P	Model 1 aOR* (95% CI)	P	Model 2 aOR** (95% CI)	P
Awareness of CDC PrEP clinical guidelines7
No	1.0	-	1.0	-	-	-
Yes	9.48 (2.37 – 37.97)	<0.001	5.49 (1.14 – 26.48)	0.03	-	-
Familiar with current research on PrEP safety and efficacy
No	1.0	-	-	-	1.0	-
Yes	5.45 (1.82 – 16.36)	<0.01	-	-	3.57 (1.04 – 12.32)	0.04
Awareness of local PrEP care referral information
No	1.0	-	1.0	-	1.0	-
Yes	8.75 (2.17 – 35.36)	<0.001	6.54 (1.45 – 29.56)	0.02	6.28 (1.39 – 28.39)	<0.01
Concern that PrEP is not effective
No	1.0	-	1.0	-	1.0	-
Yes	0.24 (0.09 – 0.68)	<0.01	0.43 (0.12 – 1.52)	0.19	0.33 (0.10 – 1.08)	0.07

Abbreviations: OR: odds ratio; 95% CI: 95% confidence interval; aOR: adjusted odds ratio; CDC: Centers for Disease Control and Prevention.

^*Model 1 controlled for aware of CDC guidelines, aware of local PrEP care referral information, and concern that PrEP is not effective.

^**Model 2 controlled for familiar with current research on PrEP safety and efficacy, aware of local PrEP care referral information, and concern that PrEP is not effective.

ED physicians’ perceived barriers to discussing PrEP with their patients included being unaware of where to refer patients (83.6%), their patients not asking about PrEP (73.1%), and not having PrEP provider training (71.6%). Most (86.6%) physicians noted their patients’ insurance coverage could be a barrier to obtaining PrEP care, yet only half (50.7%) felt the medication was too expensive and few (38.8%) felt the clinical and laboratory costs would be prohibitive. Some physicians felt they did not have time to discuss PrEP (23.9%) or that PrEP was outside the scope of emergency medicine (19.4%).

Most (94.0%) of the physicians were interested in and perceived having department support (94.0%) for PrEP training. Self-guided online training (47.8%) and in-person training (25.4%) were most frequently preferred by providers compared to webinar (11.9%), written material (9.0%), and other (6.0%) (i.e., any format, conference lecture with emailed resources).

This is among the first PrEP awareness assessments among ED physicians, which is an important group of medical doctors who could help facilitate PrEP implementation in the US. We found most physicians (79.1%) were aware of PrEP to prevent HIV infection, but were uncomfortable discussing PrEP with their patients (43.3%). Our findings suggest that when designing future provider training interventions to enhance ED physician discussions of PrEP with patients, there should be a focus on addressing local referral information; one of the main associations observed with discomfort. Awareness of the local referral information was independently associated with discomfort after adjusting for awareness of CDC PrEP clinical guidelines⁷ and knowledge of PrEP research on safety and efficacy. A critical component for future trainings should be increasing the awareness of local PrEP referral resources. Applying these insights will help institutions more effectively leverage the emergency department’s role within the PrEP continuum of care,^11,23 such as HIV testing, PrEP education and linkage to care for at-risk individuals.

No insurance coverage and high care costs were concerns for ED physicians. These concerns likely stem from provider experiences of practicing in a Medicaid non-expansion state, which means there is limited eligibility for individuals to acquire public insurance.^24,25 In response to these concerns and known patient insurance-related barriers to PrEP care,²⁶ the Missouri State Health Department created a PrEP provider directory and PrEP care referral algorithm based on an individual insurance coverage (e.g., uninsured and types of insurance).^27,28 This study highlights that better strategies for statewide dissemination of these resources is needed to reach providers in various practice settings.

We found high levels of PrEP awareness (79.1%) among ED physicians in our sample similarly seen within other non-HIV providers (66%−77%).^8,18 In contrast, Wood et al. demonstrated 54.7% of emergency medicine, compared to 75.2% of family medicine physicians located in Washington state were PrEP aware (i.e., had heard of PrEP).²⁹ High PrEP awareness in our sample did not equate to high levels of awareness of the current CDC PrEP prescribing guidelines (23.9%)⁷, which was similarly observed among non-HIV providers (32%) in Blumenthal et al.¹⁸ This discrepancy is concerning and implies robust efforts are needed to disseminate national PrEP prescribing guidelines among non-HIV providers.

More than half (53.7%) of ED physicians did not believe PrEP was effective in preventing HIV transmission, a major prescribing concern. Other concerns included medication side effects (89.6%) and selection for drug resistance (70.1%), which were higher than that reported by physicians in San Francisco (75% and 60%, respectively).²² These concerns define necessary topics for future trainings targeted at ED physicians.

Optimistically, ED physicians felt PrEP was within their practice scope (80.6%), had time to incorporate PrEP discussions into practice (76.1%), and were interested in training (94.0%). Such interest was higher than the 68% observed among non-HIV specialists in Bacon et al.²² A self-guided online training was preferred (47.8%) similarly by San Francisco physicians (47%), but less than in a national survey (83%).^8,22 Importantly, most (94.0%) ED physicians perceived having department support for PrEP initiatives; having support from higher levels of administration is proven to effect sustained organizational adoption of new practices.³⁰

Study limitations include a single site study at an urban academic center and a selection bias of those who filled out the survey online. Future studies should employ qualitative methodology to explore why ED physicians did not feel comfortable discussing PrEP, as well as how to alter the cultural (e.g., administrative support) and structural environment (e.g., electronic medical record reminders) of the ED to foster and re-enforce PrEP discussions by doctors and other staff with patients.

The ED offers an opportunity to connect at-risk individuals to PrEP care. Study findings can inform future trainings so that ED physicians can avoid missed opportunities for PrEP education and care referral. Application of these insights can strengthen the ED physician’s role within the PrEP continuum of care in an effort to reduce HIV incidence.