Abstract
HIV outcome inequities remain prevalent in the US. Medical providers are designated gatekeepers of PrEP, and thus understanding the dynamics of medical office PrEP assessments is of major interest for the public health field. We analyzed data derived from Together 5000, an internet-based U.S. national cohort of Sexual and Gender Minority (SGM) individuals aged 16–49 years and at risk for HIV. Among those eligible for PrEP uptake (n=6264), we modelled predictors of talking with a medical provider about PrEP. A third (31%) of participants had spoken to a medical provider about PrEP, which was more common among those older than 24, transgender/non-binary-identified or with annual income of ≥$20,000. Among those who spoke to a provider, 45% suggested they would initiate PrEP—this outcome was more common among participants older than 24. With a persistent stagnant uptake nationwide new opportunities to influence PrEP uptake must be explored, an attractive lesser targeted space is the medical office, specifically supporting the discussion and decision on PrEP initiation for medical providers and their patients.
Keywords: PrEP uptake, PrEP assessment, Provider communication, HIV prevention
INTRODUCTION
In the United States (US), Sexual and Gender Minority (SGM) individuals are disproportionately affected by HIV, with over two-thirds of all new HIV cases in 2018 attributed to male-to-male sexual contact.1 HIV incidence is disproportionally elevated among African-American or Black and Latinx men who have sex with men (MSM), who make up nearly 40% and 25%, respectively, of new HIV cases nationwide.2,3 Moreover, in the US, HIV is differentially distributed in urban vs. rural areas, with rates of new cases among Black and Latino MSM in selected urban northeastern areas equal to those of some predominantly rural southern states.4 However, among MSM, racial disparities in HIV outcomes distribution are more prominent than geographic ones.5
The first drug formulation of Pre-exposure Prophylaxis (PrEP) for HIV prevention was FDA-approved in 2012 and is highly efficacious in preventing HIV seroconversion.6,7 Unfortunately, despite robust dissemination efforts, PrEP remains underutilized by people who could benefit from it (i.e.. SGM people, with a specific emphasis in Black and Brown SGM). Prescribing and claims data reveal significant racial disparities in PrEP uptake and use,8 with Black and Latinx MSM making up only a quarter of all PrEP users.9 Modeling studies suggest that, under current PrEP usage levels, HIV incidence among White MSM could be reduced by half over the next decade but by only less than a quarter among Black and Latinx MSM.10 In addition, transgender and non-binary people (TGNB), who stand to significantly benefit from PrEP use, have also demonstrated poor awareness and uptake in PrEP. A sample of transgender women reported over 60% of indication for PrEP but only 5% uptake;11 similarly, a sample of transgender men found that over 50% had an indication for PrEP but less than one-third were currently using it overall.12
Currently, only clinicians can prescribe PrEP to people who could benefit from it, and typically this is achieved via some form of patient-provider interaction – a critical element of the PrEP uptake process. Although novel PrEP formulations and delivery systems, such as long-acting injectable or event-based dosing, may increase uptake, patient-provider communication will continue to be relevant to access these products.13 Understanding what factors relate to the outcomes of such patient-provider interactions (i.e. PrEP uptake, sex-positive experience, appropriate HIV prevention guidance) is an important component to inform provider and communication-related interventions, and increasing PrEP uptake among MSM overall. For example, understanding to whom a discussion did not occur, as well as what happen following a discussion to those who had one will inform solutions to support both the medical provider and patients. In this paper, we sought to examine sociodemographic and healthcare access correlates of patient-provider communications about PrEP initiation, additionally we explored the patient-reported outcome of these discussions related to uptake, deferral, and refusal of PrEP in a U.S. national sample of SGM people who met objective criteria for PrEP uptake (i.e., PrEP-eligible).
METHODS
Data used for this analysis were from the Together 5,000 study, an internet-based, U.S. national cohort of cisgender men and transgender men and women who have sex with men. The goal of Together 5,000 is to identify modifiable individual and structural factors associated with HIV risk and PrEP uptake. Participants were enrolled via ads on geosocial networking apps between October 2017 and June 2018. Participants clicking on our ads were routed to a secure online survey where eligibility was ascertained, and informed consent was obtained. Detailed enrollment procedures for the study have been described elsewhere.14,15 The study was approved by the City University of New York Institutional Review Board.
To be eligible for enrollment, participants had to self-identify as male, transgender male or transgender female; be aged 16 to 49; report at least 2 male sex partners in the past 90 days; not be currently enrolled in a PrEP clinical study or an HIV-vaccine trial; be living within the US or its territories; not be taking PrEP at the time of enrollment; self-report a negative or unknown HIV status; and meet at least one additional criterion related to clinically objective or self-reported HIV risk, such as recent (i.e., past 12 months) diagnosis of a sexually transmitted infection (STI), self-reported recent condomless anal sex (CAS) with a man (i.e., past 3 months), use of post-exposure prophylaxis (PEP) in the past 12 months, sharing of needles or use of methamphetamine in the past 3 months.
Participants determined to be eligible (i.e., meeting objective criteria for PrEP)16 were emailed a link to a secondary survey that included the measures of interest in the proposed study. Participants completing this survey (n = 6,264) were emailed an amazon.com gift card valued at $15.
Demographic and healthcare characteristics
Measures of interest for the present analysis included the following sociodemographic characteristics: age (coded as <24, 25–29, 30–39 or 40+ years); self-identified race or ethnicity (White, Black/African American, Latinx, Asian/Pacific Islander or Other); self-identified gender identity (coded dichotomously as cisgender men or TGNB); annual income (coded as <$20,000, $20,000 – $49,999, $50,000 – $99,999 or $100,000+); health insurance status (yes/no); and primary care provider (PCP) status (yes/no). Based on self-reported state of residence, participants were assigned to one of five US regions (Northeast, Midwest, South, West or US Territories). Where possible, we kept US territories as an individual category; however, when cell counts were null, these were included in the West region of the US.
Communication with a medical provider about PrEP
Participants were asked, “Have you ever spoken to a medical provider about starting PrEP?” and provided with the following options:
No, I have not ever spoken to a provider about starting PrEP
Yes, and we both decided it was a good option for me and I should start PrEP
Yes, and we both decided it might be a good option but to wait before beginning PrEP
Yes, and we both decided it was not a good option for me
Yes, and the provider was not comfortable prescribing PrEP for me
Yes, and the provider thought it was a good option, but I chose not to do it
Our first patient-provider communication outcome variable, speaking to a medical provider about PrEP, was coded dichotomously (yes/no), whereby the reference group contained those who reported not speaking to a provider about PrEP. Next, we coded each of the 5 “yes” categories as binary outcomes among the sample who reported ever speaking to a medical provider about PrEP. If the participant spoke to a provider about PrEP, whichever was the reported outcome of that discussion was assigned a “yes” value, and the remaining sample was used as reference (ref = no).
Analysis Plan
Although our parent study uses a prospective cohort design, the current cross-sectional analysis used only the data collected during the baseline phase of the study, following the eligibility assessment. We explored differences in the sample distribution between those who reported ever speaking to a medical provider about PrEP, and those who did not using chi-square confidence tests. All variables selected for analysis were a priori conceptualized as being significant to both PrEP uptake, and healthcare utilization. Next, we built a multivariable log-binomial regression model predicting speaking to a medical provider about PrEP. We then repeated this process of model building for each “yes” category, developing a total of 5 additional models, among the sample of individuals who reported speaking to a medical provider about PrEP. All our models were constructed with the relevant outcome variable and all a priori variables included in the model. We set p < 0.05 as the threshold for statistical significance. All analysis were conducted using SAS version 9.4 (Cary, NC).17
RESULTS
Sample description
Our sample (n = 6,264) was diverse in composition with 52% of participants being under 30 years old, 48% representing a race other than White, and 2% being a TGNB person. A full description of the sample is presented on Table 1. Three-quarters of participants reported annual incomes lower than $50,000, about half said they had a PCP, and 73% were insured. Nearly half (47%) of the sample lived in the South, 21% in the West, 15% in the Northeast, 14% in the Midwest, and 0.4% resided in a US territory, such as Puerto Rico.
Table 1.
Demographics and Healthcare variables of a national sample of SGM people in the United States meeting objective criteria for PrEP
| Sample | Has discussed PrEP with provider | Has not discussed PrEP with provider | ||||||
|---|---|---|---|---|---|---|---|---|
| 6264 | 1950 | 31% | 4314 | 69% | ||||
| Demographics | n | % | n | % | n | % | χ2 | p-value |
| Age group | 48.65 | <001 | ||||||
| <24 | 1532 | 24% | 368 | 19% | 1164 | 27% | ||
| 25–29 | 1728 | 28% | 578 | 30% | 1150 | 27% | ||
| 30–39 | 2041 | 33% | 693 | 36% | 1348 | 31% | ||
| 40+ | 963 | 15% | 311 | 16% | 652 | 15% | ||
| Race/Ethnicity | 8.72 | 0.069 | ||||||
| White | 3250 | 52% | 1031 | 53% | 2219 | 51% | ||
| Black | 694 | 11% | 192 | 10% | 502 | 12% | ||
| Latino | 1534 | 24% | 459 | 24% | 1075 | 25% | ||
| Asian/Pacific Islander | 226 | 4% | 78 | 4% | 148 | 3% | ||
| Other | 560 | 9% | 190 | 10% | 370 | 9% | ||
| Gender identity | 5.39 | 0.015 | ||||||
| Cisgender | 6112 | 98% | 1889 | 97% | 4223 | 98% | ||
| Transgender/non-binary | 152 | 2% | 61 | 3% | 91 | 2% | ||
| Income | 80.37 | <001 | ||||||
| < $20,000 | 2105 | 34% | 507 | 26% | 1598 | 37% | ||
| $20,000 – $50,000 | 2595 | 41% | 864 | 44% | 1731 | 40% | ||
| $50,000 – $100,000 | 1204 | 19% | 452 | 23% | 752 | 17% | ||
| $100,000+ | 360 | 6% | 127 | 7% | 233 | 5% | ||
| Primary care status | 151.21 | <001 | ||||||
| has PCP | 3218 | 51% | 1227 | 63% | 1991 | 46% | ||
| no PCP | 3046 | 49% | 723 | 37% | 2323 | 54% | ||
| Health insurance status | 29.44 | <001 | ||||||
| Insured | 4543 | 73% | 1503 | 77% | 3040 | 70% | ||
| Uninsured | 1721 | 27% | 447 | 23% | 1274 | 30% | ||
| Region | 91.85 | <001 | ||||||
| Northeast | 909 | 15% | 359 | 18% | 550 | 13% | ||
| South | 2924 | 47% | 761 | 39% | 2163 | 50% | ||
| Midwest | 887 | 14% | 236 | 15% | 601 | 14% | ||
| West | 1329 | 21% | 492 | 25% | 837 | 19% | ||
| US Territory | 22 | 0.4% | 2 | 0.1% | 20 | 0.5% |
Overall, 31% of participants said they had ever spoken to a medical provider about PrEP. The proportion of participants who reported having spoken to a provider was greater in individuals older than 30 years (52% v. 46%), individuals making more than $20,000 annually (74% v. 63%), those with a PCP (63% v. 46%), and those insured (77% v. 70%). Individuals living in the South (39% v. 50%) reported the least proportion of individuals speaking to providers about PrEP.
Factors associated with speaking to a provider about PrEP
Table 2 presents the prevalence ratio estimates obtained from our first log-binomial model of the prevalence of having spoken to a provider about PrEP (n = 1,950). Here we found that the prevalence was significantly higher in individuals 25–30 years (aPR 1.25, 95% CI 1.11 – 1.40), and those 30–39 years (aPR 1.18, 95% CI 1.05 – 1.33) compared to those under 24 years; those who were TGNB (aPR 1.36, 95% CI 1.13 – 1.64) compared to those who were cisgender; individuals with incomes greater than $20,000 compared to those reporting <$20,000; and those who had a PCP (aPR 1.58, 95% CI 1.45 – 1.72) compared to those who did not. Conversely, compared to those living in the Northeast, those who lived in the South (aPR 0.72, 95% CI 0.65 – 0.79), Midwest (aPR 0.85 95% CI 0.75 – 0.96), and US Territories (aPR 0.24, 95% CI 0.06 – 0.90) less commonly reported having spoken to a provider about PrEP.
Table 2.
Multivariable log-binomial regression predicting the prevalence ratio of discussing PrEP with a medical provider
| Prevalence of dicussing PrEP with medical provider (ref: no) | |||
|---|---|---|---|
| Characteristics | aPR | 95% Confidence | |
| Age group (ref: <24) | |||
| 24–29 | 1.25 | 1.11 | 1.40 |
| 30–39 | 1.13 | 1.05 | 1.33 |
| 40+ | 1.06 | 0.93 | 1.22 |
| Race/Etbuicity (ref: white) | |||
| Black | 0.99 | 0.87 | 1.13 |
| Latino | 1.04 | 0.95 | 1.14 |
| Asian/Pacific Islander | 1.05 | 0.88 | 1.26 |
| Other | 1.13 | 1.00 | 1.28 |
| Gentler identity (ref: cisgender) | |||
| TGNB | 1.36 | 1.13 | 1.64 |
| Income (ref: <$20,000) | |||
| $20,000 – $50,000 | 1.30 | 1.17 | 1.43 |
| $50000 – $100,000 | 1.35 | 1.20 | 1.52 |
| $100,000+ | 1.25 | 1.06 | 1.43 |
| Primary Care status (ref: no PCP) | |||
| Has PCP | 1.58 | 1.45 | 1.72 |
| Health insurance status (ref: uninsured) | |||
| Insured | 0.94 | 0.86 | 1.04 |
| Region (ref: Northeast) | |||
| South | 0.72 | 0.65 | 0.79 |
| Midwest | 0.85 | 0.75 | 0.96 |
| West | 0.99 | 0.89 | 1.10 |
| US Territorry | 0.24 | 0.06 | 0.90 |
Patient-provider decisions following a discussion about PrEP
As described on Table 3, of the 1,950 participants who reported having spoken to a provider about PrEP, just under half (45%) of this group said that “…both [patient and provider] decided it was a good option for me and I should start PrEP,” and 17% said “…the provider thought it was a good option but I chose not to do it.” Another 16% said “…both [patient and provider] decided it might be a good option but to wait before beginning PrEP,” and 15% reported that “…the provider was not comfortable prescribing PrEP for me.” Lastly, 7% said that “we both decided it was not a good option for me.” Participants across groups significantly differed regarding their age (p < 0.01), race (p < 0.01), annual income (p < 0.01), health insurance status (p < 0.001), and PCP status (p < 0.001). Table 3 provides a full description of each group, a well as results of chi-square independence tests.
Table 3.
Demographics and healthcare distribution across different participant reported outcomes of speaking to a provider about PrEP
| we both decided it was a good option for me and I should start PrEP | we both decided it might be a good option but to wait before start PrEP | we both decided it was not a good option for me PrEP | the provider was not comfortable prescribing PrEP for me | the provider thought it was a good option but I chose not to do it | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 877 | 45% | 314 | 16% | 129 | 7% | 299 | 15% | 331 | 17% | |||
| Characteristics | n | % | n | % | n | % | n | % | n | % | χ2 | p-value |
| Age group | 30.39 | 0.002 | ||||||||||
| <24 | 147 | 17% | 73 | 23% | 20 | 16% | 65 | 22% | 63 | 19% | ||
| 24–29 | 273 | 31% | 92 | 29% | 26 | 20% | 80 | 27% | 107 | 32% | ||
| 30–39 | 337 | 38% | 100 | 32% | 52 | 40% | 105 | 35% | 99 | 30% | ||
| 40+ | 120 | 14% | 49 | 16% | 31 | 24% | 49 | 16% | 62 | 19% | ||
| Race/Ethnicity | 33.61 | 0.006 | ||||||||||
| White | 456 | 52% | 162 | 52% | 77 | 60% | 187 | 63% | 149 | 45% | ||
| Black | 89 | 10% | 31 | 10% | 13 | 10% | 14 | 5% | 45 | 14% | ||
| Latino | 207 | 24% |
|
23% | 25 | 19% | 68 | 23% | 87 | 26% | ||
| Asian/Pacific Islander | 35 | 4% | 18 | 6% | 2 | 2% | 9 | 3% | 14 | 4% | ||
| Other | 90 | 10% | 31 | 10% | 12 | 9% | 21 | 7% | 36 | 11% | ||
| Gender identity | 5.43 | 0.246 | ||||||||||
| Cisgender | 851 | 97% | 301 | 96% | 128 | 99% | 292 | 98% | 317 | 96% | ||
| TGNB | 26 | 3% | 13 | 4% | 1 | 1% | 7 | 2% | 14 | 4% | ||
| Income | 30.9 | 0.002 | ||||||||||
| <$20,000 | 241 | 27% | 76 | 24% | 27 | 21% | 81 | 27% | 82 | 25% | ||
| $20,000 – $50,000 | 400 | 46% | 126 | 40% | 48 | 37% | 142 | 47% | 148 | 45% | ||
| $50,000 – $100,000 | 196 | 22% | 78 | 25% | 40 | 31% | 57 | 19% | 81 | 24% | ||
| $100,000+ | 40 | 5% | 34 | 11% | 14 | 11% | 19 | 6% | 20 | 6% | ||
| Primary care status | 26.54 | <.001 | ||||||||||
| has PCP | 515 | 59% | 221 | 70% | 99 | 77% | 194 | 65% | 198 | 60% | ||
| no PCP | 362 | 41% | 93 | 30% | 30 | 23% | 105 | 35% | 133 | 40% | ||
| Health insurance status | 20.19 | 0.001 | ||||||||||
| Insured | 640 | 73% | 259 | 82% | 110 | 85% | 240 | 80% | 254 | 77% | ||
| Uninsured | 237 | 27% | 55 | 18% | 19 | 15% | 59 | 20% | 77 | 23% | ||
| Region | 20.25 | 0.12 | ||||||||||
| Northeast | 164 | 19% | 65 | 21% | 25 | 19% | 33 | 11% | 72 | 22% | ||
| South | 346 | 39% | 115 | 37% | 52 | 40% | 131 | 44% | 117 | 35% | ||
| Midwest | 129 | 15% | 42 | 13% | 16 | 12% | 51 | 17% | 48 | 15% | ||
| West/Territories | 214 | 24% | 84 | 27% | 34 | 26% | 78 | 26% | 84 | 25% | ||
Correlates of patient-provider decisions outcomes
Table 4 provides the results of each of the five models predicting each participant reported decision among those who said they spoke to a provider about PrEP (n =1,950). For each variable the reference group is the remaining sample who did not answer the given question on the affirmative. A participant reported mutual decision to start PrEP was more common in individuals 25–30 years (aPR 1.23, 95% CI 1.05 – 1.44) and those 30–39 years (aPR 1.30, 95% CI 1.11 – 1.52), compared to those under 24 years. It was significantly lower in individuals who had a PCP (aPR 0.90, 95% CI 0.80 – 0.99) and those with health insurance (aPR 0.87, 95% CI 0.77 – 0.98). A participant reported mutual decision to wait to start PrEP was significantly less common in participants between 30 – 39 years (aPR 0.66, 95% CI 0.49 – 0.89), and those over 40 years (aPR 0.65, 95% CI 0.45 – 0.93) compared to those under 24 years. Mutually deciding that PrEP was not a good option for them was more prevalently reported by participants with a PCP (aPR 1.55, 95% CI 1.01 – 2.37). Answering that a provider was not comfortable prescribing PrEP for me was less commonly reported by Black participants (aPR 0.38, 95% CI 0.22 – 0.65) and those reporting “other” races (aPR 0.61, 95% CI 0.40 – 0.95). However, it was significantly more commonly reported by participants living in the South (aPR 1.84, 95% CI 1.28 – 2.64), Midwest (aPR 1.77, 95% CI 1.17 – 2.67), the West or US Territories (aPR 1.64, 95% CI 1.11 – 2.41) compared to those living in the Northeast. Lastly, the decision to decline a provider recommendation for PrEP was significantly more prevalently reported by Black (aPR 1.79, 95% CI 1.33 – 2.42) and Latino (aPR 1.38, 95% CI 1.07 – 1.78) than White participants.
Table 4.
Multivariable log-binomial regression models predicting the prevalence ratio of each participant reported outcome of a discussion about PrEP with a medical provider
| we both decided it was a good option for me and I should start PrEP (ref : No) | we both decided it might be a good option but to wait before starting PrEP (ref: no) | we both decided PrEP was not a good option for me (ref: no) | the provider was not comfortable prescribing PrEP for me (ref: no) | the provider thought it was a good option but I chose not to do it (ref: no) | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Characteristics | aPR | 95% Confidence | aPR | 95% Confidence | aPR | 95% Confidence | aPR | 95% Confidence | aPR | 95% Confidence | ||||||||||
| Age group (ref: <24) | ||||||||||||||||||||
| 24–29 | 1.23 | 1.05 | -- | 1.44 | 0.77 | 0.57 | -- | 1.03 | 0.79 | 0.44 | -- | 1.42 | 0.82 | 0.60 | -- | 1.12 | 1.05 | 0.78 | -- | 1.40 |
| 30–39 | 1.30 | 1.11 | -- | 1.52 | 0.66 | 0.49 | -- | 0.89 | 1.16 | 0.68 | -- | 1.99 | 0.86 | 0.63 | -- | 1.16 | 0.82 | 0.61 | -- | 1.12 |
| 40+ | 1.07 | 0.88 | -- | 1.31 | 0.65 | 0.45 | -- | 0.93 | 1.38 | 0.75 | -- | 2.51 | 0.88 | 0.61 | -- | 1.26 | 1.23 | 0.87 | -- | 1.76 |
| Race/Ethnicity (ref: white) | ||||||||||||||||||||
| Black | 0.96 | 0.81 | -- | 1.14 | 1.02 | 0.71 | -- | 1.47 | 1.10 | 0.62 | -- | 1.96 | 0.38 | 0.22 | -- | 0.65 | 1.79 | 1.33 | -- | 2.42 |
| Latino | 0.98 | 0.86 | -- | 1.11 | 1.06 | 0.81 | -- | 1.38 | 0.83 | 0.53 | -- | 1.31 | 0.81 | 0.62 | -- | 1.05 | 1.38 | 1.07 | -- | 1.78 |
| Asian/Pacific Islander | 1.01 | 0.79 | -- | 1.30 | 1.39 | 0.91 | -- | 2.12 | 0.38 | 0.09 | -- | 1.50 | 0.65 | 0.35 | -- | 1.21 | 1.21 | 0.73 | -- | 2.00 |
| Other | 1.05 | 0.89 | -- | 1.24 | 1.05 | 0.74 | -- | 1.50 | 0.94 | 0.53 | -- | 1.70 | 0.61 | 0.40 | -- | 0.95 | 1.35 | 0.96 | -- | 1.88 |
| Gender identity (ref: cisgender) | ||||||||||||||||||||
| TGNB | 0.94 | 0.69 | -- | 1.27 | 1.26 | 0.75 | -- | 2.13 | 0.29 | 0.04 | -- | 2.05 | 0.79 | 0.39 | -- | 1.60 | 1.36 | 0.83 | -- | 2.23 |
| Income (ref: <S20,000) | ||||||||||||||||||||
| $20,000 – $50,000 | 0.98 | 0.87 | -- | 1.11 | 1.01 | 0.77 | -- | 1.33 | 0.94 | 0.58 | -- | 1.51 | 0.92 | 0.71 | -- | 1.19 | 1.18 | 0.91 | -- | 1.53 |
| $50,000 – $100,000 | 0.94 | 0.80 | -- | 1.09 | 1.21 | 0.87 | -- | 1.67 | 1.22 | 0.72 | -- | 2.07 | 0.72 | 0.51 | -- | 1.02 | 1.24 | 0.91 | -- | 1.70 |
| $100,000+ | 0.69 | 0.52 | -- | 0.91 | 1.93 | 1.30 | -- | 2.85 | 1.38 | 0.71 | -- | 2.70 | 0.84 | 0.51 | -- | 1.37 | 1.09 | 0.68 | -- | 1.76 |
| Primary Care status (ref: no PCP) | ||||||||||||||||||||
| Has PCP | 0.90 | 0.80 | -- | 0.99 | 1.30 | 1.02 | -- | 1.65 | 1.55 | 1.01 | -- | 2.37 | 1.04 | 0.83 | -- | 1.32 | 1.12 | 0.72 | -- | 1.11 |
| Health insurance status (ref: uninsured) | ||||||||||||||||||||
| Insured | 0.87 | 0.77 | -- | 0.98 | 1.14 | 0.85 | -- | 1.53 | 1.33 | 0.80 | -- | 2.22 | 1.24 | 0.93 | -- | 1.65 | 1.00 | 0.78 | -- | 1.29 |
| Region (ref: Northeast) | ||||||||||||||||||||
| South | 0.97 | 0.85 | -- | 1.11 | 0.91 | 0.69 | -- | 1.20 | 1.04 | 0.66 | -- | 1.65 | 1.84 | 1.28 | -- | 2.64 | 0.76 | 0.58 | -- | 1.00 |
| Midwest | 0.96 | 0.81 | -- | 1.14 | 0.88 | 0.62 | -- | 1.26 | 0.83 | 0.45 | -- | 1.54 | 1.77 | 1.17 | -- | 2.67 | 0.88 | 0.63 | -- | 1.23 |
| West/Territories | 0.93 | 0.80 | -- | 1.08 | 1.04 | 0.78 | -- | 1.40 | 1.05 | 0.63 | -- | 1.74 | 1.64 | 1.11 | -- | 2.41 | 0.84 | 0.63 | -- | 1.13 |
DISCUSSION
We found that fewer than one-third of respondents in this US national cohort of SGM people—all of whom met clinical indications for PrEP initiation at the time of their survey—had ever discussed PrEP with a medical provider. Providers are designated gatekeepers for patients to access PrEP, and our findings point to the need for support interventions at the medical consultation level. Discussing PrEP with a medical provider was more common in older age groups, those with higher incomes, those with a PCP, and those living in the Northeast. These findings are consistent with the current literature on PrEP uptake in the US.9,18 In exploring five potential outcomes of such conversations, reflecting both the provider’s recommendations and patient’s needs, we have contributed unique information to the field.
Among participants who reported having discussed PrEP with a provider, fewer than half reported that they mutually agreed (patient and provider) that the participant should start PrEP. Further, over 50% reported that they either refused the offer of or were denied a prescription. Of concern, we found a meaningful number of participants (15%) reporting provider discomfort in prescribing PrEP, a disparity that was associated with geographic region in ways that may serve to further perpetuate regional disparities in HIV in, for example, the South. A report looking at US counties, for example, found that the US South housed only one-quarter of PrEP providing clinics in the country, despite comprising over half of all new HIV incidence 18. Our findings highlight that talking to a provider about PrEP is nuanced and that the outcome of such conversations and their influence on PrEP uptake may be particularly consequential to actual uptake. While our results do not delve into the dynamics and content of such conversations, they help better frame the structural complexities of providing and elevating PrEP in healthcare settings. Simply having a clinical indication is not enough to translate the need for PrEP into PrEP uptake, and additional resources are needed better support patients obtaining PrEP.
Patient-provider communication is an important pathway by which to increase uptake of PrEP in the US.19,20 Our results suggest that both increasing the frequency and changing the outcomes of PrEP discussions in health care practices are needed. Qualitative and exploratory research on provider communication related to PrEP has presented several barriers needed to be overcome to improve uptake including cultural competency, time and resource constraints, discussing sexual risk, and competing medical priorities.19,20 Time constraints are often cited as barrier to a more robust integration of PrEP in primary care visits, for example, with providers reporting lack of time to provide appropriate education and adherence counseling to PrEP candidates.21 Additionally, providing optimal PrEP care requires comprehensive care coordination and this can only be achieved by a multi-disciplinary approach,22 a significant gap in our current health care system.23 Both issues indicate the need for more services to occur outside of the medical office and be performed by a broader set of care team members, like patient navigators or case managers – as to reduce the burden on the medical provider. Supportive services, such as patient navigation and care coordination are not part of the standard of care for PrEP, but they also point to ways to diversify and increase the offering of and engagement in PrEP. For example, an assessment from the New York City Department of Health and Mental Hygiene found that non-white patients were more likely to accept PrEP navigation than white patients.24
In settings that cannot adopt a model of care with more team members providing and supporting PrEP services, the successful integration of telehealth in primary care settings during the COVID-19-era 25 present a potential pathway to support PrEP care, by making it client-centered, telehealth-based, and technology-infused. While pre-pandemic barriers to telehealth, such as reimbursement gaps in primary care practices providing PrEP follow-up via telehealth (i.e., via telephone or web-based) were prevalent,23 these issues appear to have been resolved. The use telehealth to connect patient with culturally competent providers, as well as to expand the network of agencies and providers caring for a patient may significantly improve gaps in communication related to discomfort, and lack of competency. Additionally, using technology through algorithms in electronic medical records (EMR) to identify PrEP-eligible patients26 and through Clinical Decision Support tools,27 may further provide clinicians with the support they need to successfully identify and support PrEP-eligible patients. For example, patient education and eligibility screening could be performed ahead of time via applications or patient portals if candidates were pre-evaluated in EMRs. Of course, the use of technology cannot be discussed without noting issues of privacy, and access to technology.
Our findings should be interpreted in light of their limitations. Although our sample is large and geographically diverse, it is not meant to be representative of all SGM individuals, let alone all at risk for HIV. Additionally, data were self-reported, which can raise concerns related to social desirability; however, self-administration of surveys (as opposed to face-to-face) can help reduce such bias.28 The timing of survey responses is also important, since PrEP messaging is active and ongoing; our surveys were collected in 2017 and 2018 and could not capture more recent changes in social norms. Although participants met the criteria for PrEP care at the time they participated in our survey, it is possible that they did not meet the criteria at the time they spoke with their provider; however, we note that in subsequent waves of data collection from this cohort, the vast majority of those eligible for PrEP at baseline remained eligible at follow up.29 It is also important to note that our data analysis did not include any confirmation on PrEP initiation, though participants answers may suggest they would potentially start PrEP. Prior research on PrEP barriers has pointed to several barriers which persist between the acquisition of a PrEP prescription and actual PrEP initiation.30 Nevertheless, future research should not only assess patient-provider communication around PrEP, but also explore the timing of when conversations occur and confirm the outcomes of such discussions. Similarly, studies on provider perspective on PrEP communication are necessary to comprehensively address this issue. Additionally, future studies are needed exploring the complexities of building and sustaining an effective patient-provider relationship, specifically as it relates to the role of patient trust and satisfaction in their provider.
Conclusion
Researchers have identified multiple points in the PrEP care cascade at which PrEP initiation and retention in care can be stymied.31 These include perceiving oneself as a good candidate for PrEP, speaking to a provider about PrEP, obtaining a prescription and renewing it, etc. Our findings indicate that additional patient-provider interactions may represent discrete sub-steps within the cascade that occurs, for example, within a medical provider’s office and that can serve to facilitate or dissuade PrEP uptake. In other words, in addition to overcoming barriers in the steps identified along the PrEP care cascade, we may also need strategies to optimize the point of contact with a provider. Our findings also indicate that the results of the discussion between a patient and provider differed geographically in such a way that could perpetuate existing regional disparities in HIV. Important next steps include looking more closely at how interactions between patients and providers and the broader clinical context, such as the use of technology-enabled tools and organizational structures to support PrEP initiation, may influence actual uptake and satisfaction with PrEP use.
Acknowledgements:
We are thankful to participants for their time as well as other members of the Together 5000 study team.
Funding statement:
This research was supported in part by the National Institutes for Health (UH3 AI 133675, Grov) and the Einstein-Rockefeller-CUNY Center for AIDS Research (ERC CFAR, P30 AI124414, Goldstein). The NIH does not necessarily endorse these study findings.
REFERENCES
- 1.Centers for Disease Control and Prevention. Diagnoses of HIV Infection in the United States and Dependent Areas, 2017. HIV Surveillance Reports Web site. https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-report-2017-vol-29.pdf. Published 2017. Accessed January 12 2019, 2019.
- 2.Prevention CfDCa. Centers for Disease Control and Prevention. HIV in the United States and Dependent Areas. Centers for Disease Control and Prevention. https://www.cdc.gov/hiv/statistics/overview/ataglance.html. Published 2018. Accessed January 12 2019. [Google Scholar]
- 3.Centers for Disease Control and Prevention. HIV Among African Americans. https://www.cdc.gov/hiv/group/racialethnic/africanamericans/index.html. Published 2018. Accessed January 12 2019, 2019.
- 4.Centers for Disease Control and Prevention. HIV in the United States by Region. Centers for Disease Control and Prevention. https://www.cdc.gov/hiv/statistics/overview/geographicdistribution.html. Published 2018. Accessed January 12 2019, 2019. [Google Scholar]
- 5.Rosenberg ES, Purcell DW, Grey JA, Hankin-Wei A, Hall E, Sullivan PS. Rates of prevalent and new HIV diagnoses by race and ethnicity among men who have sex with men, US states, 2013–2014. Annals of epidemiology. 2018. [DOI] [PubMed] [Google Scholar]
- 6.Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. New England Journal of Medicine. 2010;363(27):2587–2599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Baeten JM, Donnell D, Ndase P, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. New England Journal of Medicine. 2012;367(5):399–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Smith DK, Van Handel M, Grey J. Estimates of adults with indications for HIV pre-exposure prophylaxis by jurisdiction, transmission risk group, and race/ethnicity, United States, 2015. Annals of Epidemiology. 2018;28(12):850–857. e859. [DOI] [PubMed] [Google Scholar]
- 9.Ya-lin AH, Zhu W, Smith DK, Harris N, Hoover KW. HIV Preexposure Prophylaxis, by Race and Ethnicity—United States, 2014–2016. Morbidity and Mortality Weekly Report. 2018;67(41):1147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jenness SM, Maloney KM, Smith DK, et al. The PrEP care continuum and racial disparities in HIV incidence among men who have sex with men. BioRxiv. 2018:249540. [Google Scholar]
- 11.Kuhns LM, Reisner SL, Mimiaga MJ, Gayles T, Shelendich M, Garofalo R. Correlates of PrEP indication in a multi-site cohort of young HIV-uninfected transgender women. AIDS and Behavior. 2016;20(7):1470–1477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Reisner SL, Moore CS, Asquith A, Pardee DJ, Mayer KH. The pre-exposure prophylaxis cascade in at-risk transgender men who have sex with men in the United States. LGBT Health. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mayer KH, Allan-Blitz L-T. PrEP 1.0 and Beyond: Optimizing a Biobehavioral Intervention. J Acquir Immune Defic Syndr. 2019;82 Suppl 2(2):S113–S117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Grov C, Westmoreland DA, Carneiro PB, et al. Recruiting vulnerable populations to participate in HIV prevention research: findings from the Together 5000 cohort study. Annals of Epidemiology. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Nash D S M, MacCrate C, Mirzayi C, Patel VV, Hoover D Pantalone DW, Golub SA, Millett G, D’Angelo A, Westmoreland DA, & Grov C An online, PrEP-era observational study of vulnerable, HIV-negative gay and bisexual men and transmen and transwomen who have sex with men in the United States, Puerto Rico and Guam: Protocol for a Cohort Study. Journal of Medical Internet Research - Public Health & Surveillance. in press. [Google Scholar]
- 16.Walensky RP, Paltiel AD. New USPSTF Guidelines for HIV screening and preexposure prophylaxis (PrEP): straight A’s. JAMA network open. 2019;2(6):e195042–e195042. [DOI] [PubMed] [Google Scholar]
- 17.SAS [computer program]. Version 9.4. Cary, NC: 2013. [Google Scholar]
- 18.Siegler AJ, Bratcher A, Weiss KM, et al. Location Location Location: An Exploration of Disparities in Access to Publicly Listed PrEP Clinics in the United States. Annals of Epidemiology. 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wilson K, Bleasdale J, Przybyla SM. Provider-patient communication on pre-exposure prophylaxis (Prep) for HIV prevention: An exploration of healthcare provider challenges. Health Communication. 2020:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Schwartz J, Grimm J. Communication Strategies for Discussing PrEP with Men Who Have Sex with Men. Journal of Homosexuality. 2020:1–14. [DOI] [PubMed] [Google Scholar]
- 21.Krakower D, Oldenburg C, Mimiaga M, et al. Patient-provider communication about sexual behaviors and preexposure prophylaxis: results from a national online survey of men who have sex with men in the United States. Paper presented at: Poster presented at the 8th International AIDS Society Conference on HIV Pathogenesis, Treatment & Prevention, Vancouver, British Columbia, Canada2015. [Google Scholar]
- 22.Silapaswan A, Krakower D, Mayer KH. Pre-exposure prophylaxis: a narrative review of provider behavior and interventions to increase PrEP implementation in primary care. Journal of general internal medicine. 2017;32(2):192–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Marcus JL, Volk JE, Pinder J, et al. Successful implementation of HIV preexposure prophylaxis: lessons learned from three clinical settings. Current HIV/AIDS Reports. 2016;13(2):116–124. [DOI] [PubMed] [Google Scholar]
- 24.Pathela P, Jamison K, Blank S, Daskalakis D, Hedberg T, Borges C. The HIV pre-exposure prophylaxis (PrEP) cascade at NYC sexual health clinics: navigation is the key to uptake. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2020;83(4):357–364. [DOI] [PubMed] [Google Scholar]
- 25.Wosik J, Fudim M, Cameron B, et al. Telehealth transformation: COVID-19 and the rise of virtual care. Journal of the American Medical Informatics Association. 2020;27(6):957–962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gruber S, Krakower D, Menchaca JT, et al. Using electronic health records to identify candidates for human immunodeficiency virus pre‐exposure prophylaxis: An application of super learning to risk prediction when the outcome is rare. Statistics in medicine. 2020;39(23):3059–3073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sutton RT, Pincock D, Baumgart DC, Sadowski DC, Fedorak RN, Kroeker KI. An overview of clinical decision support systems: benefits, risks, and strategies for success. NPJ digital medicine. 2020;3(1):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.King MF, Bruner GC. Social desirability bias: A neglected aspect of validity testing. Psychology & Marketing. 2000;17(2):79–103. [Google Scholar]
- 29.Mehrotra ML, Westmoreland DA, Patel VV, Hojilla JC, Grov C. Breaking Inertia: Movement Along the PrEP Cascade in a Longitudinal US National Cohort of Sexual Minority Individuals at Risk for HIV. Journal of Acquired Immune Deficiency Syndromes (1999). 2021;86(5):e118–e125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mayer KH, Agwu A, Malebranche D. Barriers to the wider use of pre-exposure prophylaxis in the United States: a narrative review. Advances in Therapy. 2020;37(5):1778–1811. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Parsons JT, Rendina HJ, Lassiter JM, Whitfield TH, Starks TJ, Grov C. Uptake of HIV pre-exposure prophylaxis (PrEP) in a national cohort of gay and bisexual men in the United States: the motivational PrEP cascade. Journal of acquired immune deficiency syndromes (1999). 2017;74(3):285. [DOI] [PMC free article] [PubMed] [Google Scholar]