Abstract
Cabotegravir long-acting injectable HIV pre-exposure prophylaxis (LA PrEP) is efficacious, with a good safety profile, and was approved by the US Food and Drug Administration in December 2021. Understanding variations in potential user preferences for LA PrEP may inform implementation and subsequently improve uptake and community-level effectiveness. HIV-negative, sexually active men who have sex with men (MSM) aged ≥15 years were recruited online for the 2019 American Men's Internet Survey, before LA PrEP approval. Respondents completed a discrete-choice experiment (DCE) with hypothetical LA PrEP attributes (out-of-pocket cost, perceived side effects, injection frequency, perceived stigma, service location). Latent class analysis segmented respondents into groups based on their preferences for the attributes presented, and relative importance of preference weights and willingness-to-pay were calculated. While the majority had never used daily oral PrEP, 73% of the 2489 respondents were very or somewhat likely to use LA PrEP. Three latent classes were identified from 2241 respondents in the DCE. The “side effects-averse” class was the largest group (64% of respondents) and placed 61% relative importance on side effects. The “ambivalent” class (20% of respondents) placed higher importance on stigma (17% of relative importance) than other classes. The “cost-conscious” class (16% of respondents) placed higher relative importance (62%) on cost compared with other attributes and classes. Perceived side effects were an important hypothetical barrier for LA PrEP uptake among a large proportion of potential MSM users. Minimizing out-of-pocket costs is likely to increase uptake and may be important to equitable access. Tailored communication strategies are recommended for the different groups of potential LA PrEP users.
Keywords: HIV/AIDS, prevention, discrete-choice experiment, PrEP
Introduction
Daily oral pre-exposure prophylaxis (DO PrEP) has been approved for adult use in the United States since 2012, but only 363,000 people in the United States were taking PrEP as of 2022, most of them men.1,2 Only 25% of ∼1.2 million people who were eligible for PrEP in the United States had been prescribed PrEP as of 2020.3 Previous work shows large gaps between eligibility, PrEP-to-need ratio, and use.4 Between 2013 and 2017, only 44% of the PrEP-indicated men from the American Men's Internet Survey (AMIS) had ever initiated DO PrEP, 69% who started DO PrEP discontinued it, and 40% never restarted.4
Barriers to DO PrEP adherence and persistence (long-term use while indicated) among men who have sex with men (MSM) in the United States include fear of social stigma, concerns about side effects, low perception of HIV risk, distrust in the health care system, cost, lack of insurance, and lack of access to medical care.5,6 Low decision-making power, particularly among MSM engaging in transactional sex or experiencing abuse, also presents a barrier to PrEP uptake and persistence.7 Among young Black and Latinx MSM and transgender women aged 17–24 years, additional barriers included fear of disclosure and low desire to take daily medications.6
Better uptake of current and new PrEP regimens is essential to achieve PrEP's overall public health promise of reducing annual HIV incidence in the United States. Cabotegravir long-acting injectable HIV pre-exposure prophylaxis (LA PrEP) may overcome some of the challenges of DO PrEP since it is injected only once every 2 months, rather than a daily pill. Cabotegravir LA PrEP was shown to be efficacious with a good safety profile in clinical trials for cisgender men and cisgender and transgender women.8–10 It received Food and Drug Administration's (FDA) approval in December 2021.11 In trials, participants spoke favorably of LA PrEP over oral PrEP, citing the convenience of the injection and not having to worry about daily pill-taking.12
LA PrEP will bring its own implementation challenges to uptake and persistence, potentially including cost; fear of side effects, including injection site pain; worries about drug effectiveness waning between injections; and general distrust of injections.13–15 A study conducted with MSM in Washington, DC, found that if given a choice, two thirds of MSM would pick LA PrEP compared with less than one quarter choosing DO PrEP.16 A 2018 qualitative study conducted with MSM finishing 1 year of DO PrEP as part of a demonstration project found that two thirds would want to switch to LA PrEP and that factors related to logistics of LA PrEP (e.g., getting to regular clinic visits) would feature more heavily than physical factors (e.g., injection site pain).17
In a qualitative study of young cisgender and transgender sexual minority men aged 17–24 years who were eligible for PrEP expressed nuanced thinking around LA PrEP. They had concerns related to adherence to dosing and clinic appointments and some expressed discomfort with needles. In contrast, others liked the idea of an 8-week injection better than the potential of missing daily pills or being “outed” by the pills, but also wondered about a medicine staying in their system that long.18 Although the implementation of LA PrEP is still relatively nascent, a therapeutic review of injectable HIV treatment (cabotegravir/rilpivirine) summarized available data and found such barriers as nonadherence, procurement, and challenges with reimbursement.19
There will likely be variations in preferences across subgroups of MSM. For example, in a preference and acceptability study of LA PrEP, youth had greater concern about cost than older men; in addition, Black MSM and those with lower incomes were less likely to prefer LA PrEP to oral PrEP.16,20 Further, one study found willingness to use LA PrEP to be higher among Hispanic men and those who were previously engaged with oral PrEP.21 Conversely, another study found that MSM at highest risk of HIV acquisition had greatest interest in LA PrEP compared with DO PrEP.14 A study of young MSM found that they preferred DO PrEP over other modalities (injectable, implant, event-driven, and infusions) but found three classes of decision-making processes among them (provider-led, patient-driven, and shared).22 Those in greatest need of PrEP may be the most socioeconomically disadvantaged, lack health insurance, and are unsuccessfully engaged in available health services.23 Given these differences across identities and factors, audience-segmented, tailored strategies are necessary to maximize uptake.
With the December 2021 FDA approval of cabotegravir LA PrEP, it is useful to understand potential users' preferences and how they may regard facilitators and barriers to uptake, adherence, and persistence. Discrete-choice experiments (DCE) are one method to quantitatively elicit potential users' preferences for products and services, including hypothetical or pre-market products24 and have been used widely to measure preferences for HIV prevention products such as LA PrEP.25 Further, through analytic methods such as latent class analysis (LCA), we can segment the sample into subgroups based on their patterns of responses to preference questions.26,27 This heterogeneity in the preferences of latent subgroups could then be used to inform social marketing strategies or tailor implementation, for example. In turn, these informed strategies may improve uptake and ultimately community-level effectiveness. In this study, we assessed likelihood of using LA PrEP and measured preference heterogeneity using LCA in a 2019 cross-sectional US nationwide online sample of MSM.
Methods
This study analyzed data collected through the 2019 round of the AMIS, an annual online cross-sectional survey that collects information from 10,000 US MSM on demographics, sexual health, mental health, and prevention service utilization. The protocol and data cleaning (e.g., de-duplication) are described in detail elsewhere.28
Participants and procedures
Participants are recruited each year through convenience sampling from a variety of websites and through social media applications using banner ads and email blasts to members. Eligible individuals were ≥15 years old, assigned male at birth, lived in the United States, and reported anal or oral sex with a man at least once in the past. Those eligible were asked to provide informed consent and completed the survey immediately. There was no compensation for participation. From September through December in the 2019 round of the AMIS, we added a module on LA PrEP to the AMIS survey instrument for all participants who did not report a previous HIV diagnosis. Those who reported at least some willingness to use LA PrEP were then eligible to participate in the DCE to measure preferences for various LA PrEP implementation scenarios. Participants with incomplete DCE data were dropped from analysis. The analytical sample was further restricted to MSM reporting oral or anal sex with men in the past 12 months.
Measures
The AMIS participants who did not report a prior HIV diagnosis were shown the following: “Pre-exposure prophylaxis, or PrEP, has been shown to reduce the risk of acquisition of HIV when taken as a daily pill. Researchers are working on an injectable form of PrEP that would be prescribed by a doctor and given as a shot. The shot would have to be given every few months to reduce the risk of getting HIV. How likely would you be to use this injectable form of PrEP to reduce the risk of getting HIV?” with a 5-point response scale (very likely, somewhat likely, neither likely or unlikely, somewhat unlikely, very unlikely).
The independent measures for these analyses included demographics, sexual behaviors, and DO PrEP use. In addition to standard age and race/ethnicity variables, we also classified participants' county of residency population density using the US National Center for Health Statistics (NCHS) Rural–Urban classification scheme.29 Sexual behavior included condomless anal sex with a male partner in the past 12 months (yes/no), number of male sex partners in the past 12 months (two or more, one), and type of male sex partner (main, casual, or both). Diagnosis with a bacterial sexually transmitted infection (STI) (chlamydia, gonorrhea, syphilis) in the past 12 months was also asked. The use of DO PrEP was also assessed for each participant as no use, current use, or prior use.
Discrete-choice experiment
The DCE used a 3-block, D-efficient design,30 with 27 tasks, created using Ngene software.31 Each block consisted of nine paired profiles with five hypothetical LA PrEP attributes with three levels each: out-of-pocket cost ($0, $25, $100), perceived side effects (mild, moderate, severe), injection frequency (every 6 months, every 3 months, every 2 months), perceived negative judgment (low-, medium-, and high-level risk), and service location (private doctor's office, sexual health clinic, pharmacy). As this DCE was designed before clinical trial results and FDA approval, the attributes and levels represent hypothetical profiles of LA PrEP. The attributes and levels were primarily based on review of the literature, including the findings from a qualitative substudy of the ÉCLAIR trial on LA PrEP among men in the United States.12
Analyses
Analyses were conducted only with participants who reported having oral or anal sex with another man in the past 12 months, who did not report a prior HIV diagnosis, and who answered the LA PrEP interest question. Likelihood to use LA PrEP was examined descriptively and compared by independent variables using multivariable logistic regression modeling and predicted margins estimation of adjusted prevalence ratios and 95% confidence intervals (CIs). Only variables that were significant in bivariate analyses using crude prevalence ratios and 95% CIs were used in the multivariable model. Willingness analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC, USA).
Discrete-choice experiment/latent class analysis
We used conditional logit and LCA (clogit and lclogit in Stata 16) to segment respondents into groups based on their preferences for the hypothetical attributes presented and calculated preference weights within each class.26,32 We estimated models with two through eight classes and chose a three-class solution based on Akaike's information criterion (AIC), Bayesian information criterion (BIC), and qualitative interpretation of classes.
Relative attribute importance
Using preference weights, we calculated relative utility of attribute levels by dividing the difference between the maximum and minimum preference weight for each attribute by the sum of all differences of all attributes and then multiplying by 100.33,34 This indicates the importance of each attribute on the participants' choices between alternatives, relative to other attributes.
Willingness-to-pay
We calculated willingness-to-pay (WTP) (Stata 16) by class, interpreted as WTP × dollars to avoid a one-unit change in each attribute level, for example, from severe to moderate perceived side effects.
Ethical statement
The study was conducted in compliance with federal regulations governing protection of human subjects and reviewed and exempted by the Emory University Institutional Review Board.
Results
There were 2489 MSM participants, the majority of whom were younger than 30 years, White, privately insured, and living in urban or suburban counties (Table 1). Most participants had also engaged in condomless anal sex, had two or more male sex partners, and had a casual male partner (only or also with main partner) in the past 12 months. Nearly one in seven (13.5%) had been diagnosed with a bacterial STI in the past 12 months. Although the majority had never used DO PrEP, 15.2% were current users and 3.9% had used it in the past. Almost three quarters (72.8%) were very or somewhat likely to use LA PrEP.
Table 1.
Likelihood of Using Long-Acting Injectable HIV Pre-Exposure Prophylaxis Among 2489 US Men Who Have Sex with Men, 2019
| Total sample |
Very likely |
Somewhat likely |
Neither likely or unlikely |
Somewhat unlikely |
Very unlikely |
PRa
(95% CI) |
aPRa (95% CI) |
|
|---|---|---|---|---|---|---|---|---|
| N (%) | n (%) | n (%) | n (%) | n (%) | n (%) | |||
| 2489 (100) | ||||||||
| Age, years | ||||||||
| 15–24 | 1081 (43.4) | 458 (42.4) | 324 (30.0) | 110 (10.2) | 89 (8.2) | 100 (9.3) | 0.96 (0.80–1.17) | |
| 25–29 | 517 (20.8) | 242 (46.8) | 142 (27.5) | 41 (7.9) | 36 (7.0) | 56 (10.8) | 1.20 (0.96–1.50) | |
| 30–39 | 365 (14.7) | 187 (51.2) | 81 (22.2) | 40 (11.0) | 19 (5.2) | 38 (10.4) | 1.27 (0.99–1.62) | |
| 40+ | 526 (21.1) | 231 (43.9) | 149 (28.3) | 63 (12.0) | 29 (5.5) | 54 (10.3) | REF | |
| Race/ethnicity | ||||||||
| Black, non-Hispanic | 278 (11.2) | 149 (53.6) | 65 (23.4) | 24 (8.6) | 13 (4.7) | 27 (9.7) | 1.30 (1.03–1.65) | 1.27 (1.00–1.62) |
| Hispanic | 400 (16.1) | 194 (53.6) | 106 (26.5) | 38 (9.5) | 32 (8.0) | 30 (7.5) | 1.04 (0.85–1.27) | 1.03 (0.84–1.27) |
| Other/multiple races | 206 (8.3) | 88 (42.7) | 60 (29.1) | 28 (13.6) | 28 (13.6) | 17 (8.3) | 0.84 (0.64–1.10) | 0.85 (0.65–1.11) |
| White, non-Hispanic | 1567 (63.0) | 667 (42.6) | 458 (29.2) | 159 (10.1) | 159 (10.1) | 169 (10.8) | REF | REF |
| Urbanicity | ||||||||
| Urban | 966 (38.8) | 451 (46.7) | 262 (27.1) | 97 (10.0) | 72 (7.5) | 84 (8.7) | REF | |
| Suburban | 530 (21.3) | 234 (44.2) | 151 (28.5) | 57 (10.8) | 36 (6.8) | 52 (9.8) | 0.96 (0.79–1.17) | |
| Small/Med metro | 764 (30.7) | 335 (43.8) | 211 (27.6) | 84 (11.0) | 48 (6.3) | 86 (11.3) | 1.01 (0.85–1.21) | |
| Rural | 223 (9.0) | 94 (42.2) | 71 (31.8) | 16 (7.2) | 16 (7.2) | 26 (11.7) | 1.04 (0.80–1.36) | |
| Education | ||||||||
| Less than HS diploma | 100 (4.0) | 37 (37.0) | 32 (32.0) | 12 (12.0) | 9 (9.0) | 10 (10.0) | REF | |
| HS diploma/equivalent | 362 (14.5) | 170 (47.0) | 96 (26.5) | 32 (8.8) | 26 (7.2) | 38 (10.5) | 1.40 (0.93–2.10) | |
| Some college/technical degree | 857 (34.4) | 392 (45.7) | 241 (28.1) | 94 (11.0) | 46 (5.4) | 84 (9.8) | 1.24 (0.85–1.82) | |
| College degree + | 1161 (46.7) | 516 (44.4) | 325 (28.0) | 114 (9.8) | 91 (7.8) | 115 (9.9) | 1.26 (0.87,1.93) | |
| Health insurance | ||||||||
| Private only | 1824 (73.3) | 807 (44.2) | 514 (28.2) | 186 (10.2) | 137 (7.5) | 180 (9.9) | REF | |
| Public only | 286 (11.5) | 148 (51.7) | 73 (25.5) | 27 (9.4) | 11 (3.8) | 27 (9.4) | 1.18 (0.94–1.49) | |
| Other/multiple | 118 (4.7) | 49 (41.5) | 42 (35.6) | 8 (6.8) | 6 (5.1) | 13 (11.0) | 0.99 (0.70–1.40) | |
| None | 218 (8.6) | 95 (43.6) | 58 (26.6) | 27 (12.4) | 14 (6.4) | 24 (11.0) | 1.05 (0.81–1.36) | |
| Sexual and drug use, past 12 months | ||||||||
| Condomless anal sex | ||||||||
| Yes | 1813 (72.8) | 871 (48.0) | 490 (27.0) | 156 (8.6) | 114 (6.3) | 182 (10.0) | 1.50 (1.27–1.76) | 1.43 (1.21–1.69) |
| No | 676 (27.2) | 247 (36.5) | 206 (30.5) | 98 (14.5) | 59 (8.7) | 66 (9.8) | REF | REF |
| Male sex partners | ||||||||
| Two or more | 1741 (69.9) | 848 (48.7) | 502 (28.8) | 137 (7.9) | 106 (6.1) | 148 (8.5) | 1.21 (0.99–1.49) | |
| One | 374 (15.0) | 124 (33.2) | 87 (23.3) | 69 (18.4) | 32 (8.6) | 62 (16.6) | REF | |
| Male partner type | ||||||||
| Casual only | 900 (36.2) | 408 (45.3) | 253 (28.1) | 90 (10.0) | 67 (7.4) | 82 (9.1) | 1.09 (0.89–1.34) | |
| Main and casual | 1009 (40.5) | 486 (48.2) | 304 (30.1) | 79 (7.8) | 59 (5.8) | 81 (8.0) | 1.14 (0.93–1.39) | |
| Main only | 479 (19.2) | 165 (34.4) | 122 (25.5) | 78 (16.3) | 39 (8.1) | 75 (15.7) | REF | |
| STI diagnosis | ||||||||
| Yes | 336 (13.5) | 195 (58.0) | 80 (23.8) | 16 (4.8) | 22 (6.5) | 23 (6.8) | 1.39 (1.13–1.73) | 1.14 (0.91–1.43) |
| No | 2253 (86.5) | 923 (42.9) | 616 (28.6) | 238 (11.1) | 151 (7.0) | 225 (10.5) | REF | REF |
| Illicit drug use | ||||||||
| Yes | 977 (39.3) | 463 (47.4) | 257 (26.3) | 90 (9.2) | 69 (7.1) | 98 (10.0) | 1.16 (1.00–1.35) | 1.12 (0.96–1.30) |
| No | 1512 (60.7) | 655 (43.3) | 439 (29.0) | 164 (10.8) | 104 (6.9) | 150 (9.9) | REF | REF |
| PrEP-related factors | ||||||||
| DO PrEP use | ||||||||
| None | 2031 (81.6) | 774 (38.1) | 570 (28.1) | 229 (11.3) | 149 (7.3) | 220 (10.8) | REF | REF |
| Current | 383 (15.4) | 244 (63.7) | 80 (20.9) | 16 (4.2) | 21 (5.5) | 17 (4.4) | 1.56 (1.27–1.92) | 1.39 (1.11–1.72) |
| Past | 161 (6.5) | 95 (59.0) | 43 (26.7) | 8 (5.0) | 2 (1.2) | 10 (6.2) | 1.44 (1.06–1.94) | 1.32 (0.97–1.79) |
| Willing to take DO PrEPb | ||||||||
| Yes | 1457 (69.4) | 723 (49.6) | 428 (29.4) | 115 (7.9) | 81 (5.6) | 87 (6.0) | REF | |
| No | 643 (30.6) | 92 (14.3) | 158 (24.6) | 118 (18.4) | 69 (10.7) | 138 (21.5) | 0.76 (0.64–0.90) | |
| DO PrEP adherence, past monthc | ||||||||
| <15 Doses | 356 (93.9) | 356 (93.9) | 71 (19.9) | 16 (4.5) | 20 (5.6) | 17 (4.8) | 1.29 (0.56–2.99) | |
| 16+ Doses | 23 (6.1) | 15 (65.2) | 7 (30.4) | 0 (0.0) | 1 (4.3) | 0 (0.0) | REF | |
| DO PrEP discontinuation, past 12 monthsd | ||||||||
| Yes | 101 (20.9) | 57 (56.4) | 28 (27.7) | 5 (5.0) | 2 (2.0) | 6 (5.9) | 0.83 (0.53–1.28) | |
| No | 383 (79.1) | 244 (63.7) | 80 (20.9) | 16 (4.2) | 21 (5.5) | 17 (4.4) | REF | |
Bold indicates significance at p < 0.05.
Results of ordinal logistic regression analyses.
Among those who have never taken DO PrEP.
Among those who were currently taking Do PrEP.
Among those who had taken DO PrEP in the past 12 months.
aPR, adjusted prevalence ratio; CI, confidence interval; DO PrEP, daily oral pre-exposure prophylaxis; PR, prevalence ratio; PrEP, pre-exposure prophylaxis; STI, sexually transmitted infection.
Model fit statistics for the LCA are presented in Table 2. A three-class model was chosen based on fit statistics and qualitative interpretation of classes. Results for the preferences LCA are presented in Table 3 as preference weights for each attribute level (e.g., mild, moderate, severe) and relative importance overall and by class (also see Fig. 1). Preference weights greater than zero indicate relative positive impact of a given level on participant preference, while weights lower than zero indicate a relative negative impact. Relative importance relates the attribute's preference weights to each other, indicating which attributes (i.e., side effects, location) have greater or lesser impact on overall preferences.
Table 2.
Latent Class Model Fit Statistics
| No. of classes | Log likelihood | AIC | CAIC | BIC |
|---|---|---|---|---|
| 1 | −8053.938 | 16147.88 | 16319.71 | — |
| 2 | −8015.769 | 16077.54 | 16231.87 | 16208.87 |
| 3 | −7837.821 | 15745.64 | 15980.5 | 15945.5 |
| 4 | −7755.044 | 15604.09 | 15919.47 | 15872.47 |
| 5 | −7667.403 | 15452.81 | 15848.71 | 15789.71 |
| 6 | −7619.065 | 15380.13 | 15856.55 | 15785.55 |
AIC, Akaike's information criterion; BIC, Bayesian information criterion; CAIC, corrected Akaike's information criterion.
Table 3.
Preference Weights of LA PrEP Characteristics by Class Among 2241 Men Who Have Sex with Men, 2019
| Barriers to PrEP implementation | Total DCE sample |
Side effects averse class (64%) |
Ambivalent class (20%) |
Cost-conscious class (16%) |
|---|---|---|---|---|
| Preference weights | Preference weights | Preference weights | Preference weights | |
| Perceived side effects | ||||
| Mild | Ref. | Ref. | Ref. | Ref. |
| Moderate | −0.611 | −0.19 | −0.963 | −0.238 |
| Severe | −2.412 | −1.068 | −4.397 | −1.763 |
| Injection frequency | ||||
| Every 2 months | Ref. | Ref. | Ref. | Ref. |
| Every 3 months | 0.551 | 0.775 | 0.851 | 0.163 |
| Every 6 months | 0.580 | 1.312 | 0.416 | 0.741 |
| Out-of-pocket cost | ||||
| $0 | Ref. | Ref. | Ref. | Ref. |
| $25 | −0.552 | −0.202 | −0.506 | −2.042 |
| $100 | −1.63 | −1.035 | −1.527 | −5.066 |
| Service location | ||||
| Private doctor's office | Ref. | Ref. | Ref. | Ref. |
| Pharmacy | −0.124 | −0.255 | −0.319 | −0.39 |
| Sexual health clinic | −0.083 | −0.398 | −0.14 | 0.145 |
| Risk of negative judgment | ||||
| High | Ref. | Ref. | Ref. | Ref. |
| Medium | 0.234 | 0.464 | 0.102 | −0.074 |
| Low | 0.305 | 0.799 | 0.015 | 0.164 |
DCE, discrete-choice experiments; LA PrEP, long-acting injectable HIV pre-exposure prophylaxis.
FIG. 1.
Relative attribute importance, overall and by class.
Class 1, the “side effects-averse” class (64% of the sample), exhibited a strong dislike for perceived side effects (61.1% of relative importance), some dislike for higher cost (21.2%), and cared somewhat about injection frequency (12.8%). The relative importance of perceived side effects was 2.9 times greater than the relative importance of cost (e.g., relative importance of side effects divided by relative importance of cost).
Class 2, the “ambivalent” class (20% of the sample), is differentiated from the others by the relative importance placed across all attributes, with no attributes dominating. This group placed 17.3% of importance on avoiding stigma (negative judgment), compared with only 2.0% and 1.4% in the other classes. This class also placed moderate importance on perceived side effects (23.2%) and cost (22.4%). While service location was relatively unimportant compared with the other attributes for this class, they placed two times the importance on it versus the other classes (8.6% vs. 4.8% and 4.4%).
Class 3, the “cost-conscious” class, making up 16% of the sample, exhibited strong dislike for higher out-of-pocket cost (62.4% of relative importance), and they valued perceived side effects (21.7%) and injection frequency with less relative importance (about 9.1%). In relative terms, for this group, cost carried 2.8 times and 6.8 times greater importance than perceived side effects and injection frequency, respectively (e.g., relative importance of cost divided by relative importance of injection frequency).
WTP analysis (Table 4) indicates willingness to pay a certain level of money (USD) per injection for a one-unit change in each attribute level. This analysis demonstrated that the side effects-averse class would pay $314 ($156.91 per level), the ambivalent class would pay $99 ($48.83 per level), while the cost-conscious class would pay only $32 ($15.85 per level) to experience only mild perceived side effects. The ambivalent group would pay much more than the other two groups to reduce injection frequency, change service location to more private, and would pay the most for reducing negative judgment ($104 to reduce from high-to-low risk of negative judgment). This compares to only $3 the cost-conscious group would pay for the same reduction in negative judgment. Interestingly, the cost-conscious and ambivalent groups would pay to change the service location to a private doctor's office, while the side effects-averse group had a negative $6.47 WTP for this attribute, meaning that they would need to see a reduction in price to be willing to access LA PrEP at a private doctor's office.
Table 4.
Willingness to Pay by Class
| Attribute | Unit change | Side effects-averse class USD (95% CI) | Ambivalent class USD (95% CI) | Cost-conscious class USD (95% CI) |
|---|---|---|---|---|
| Side effects | Decrease from severe to moderate or moderate to mild | 156.91 (145.66 to 168.16) | 48.83 (40.86 to 56.80) | 15.85 (13.96 to 17.77) |
| Injection frequency | Decrease from 6 to 3 months, or 3 to 2 months | 4.77 (2.92 to 6.63) | 34.88 (29.76 to 40.01) | 3.36 (2.56 to 4.17) |
| Service location | Change from pharmacy to private office or private office to sexual health clinic | −6.47 (−9.58 to −3.37) | 16.79 (10.02 to 23.55) | 4.91 (3.09 to 6.72) |
| Negative judgment | Decrease from high to medium or medium to low | 9.42 (6.19 to 12.65) | 52.39 (41.31 to 63.48) | 1.63 (−0.26 to 3.52) |
| Willing to pay × dollars to avoid one unit change in attribute level | ||||
CI, confidence interval; USD, US dollars.
Discussion
This analysis indicates preference heterogeneity for implementation characteristics for LA-PrEP among over 2000 MSM in the 2019 AMIS. Out-of-pocket costs and perceived side effects emerged as the most important attributes overall, but different MSM subgroups put higher relative importance on one attribute versus another.
About two thirds of respondents, the “side effects-averse” group, were primarily concerned about perceived side effects and were much less concerned about cost. This is exhibited by their willingness to pay a higher price for a product with only mild side effects. To maximize the uptake of LA PrEP in this group, the perception or fear of side effects will need to be addressed. This can perhaps be achieved through tailored messaging and information strategies. Perceived side effects were important to potential users for DO and LA PrEP in both qualitative and quantitative studies among MSM in the United States.13,35,36 Among actual LA PrEP trial participants, however, although the majority did experience some side effects (mostly, mild to moderate pain at the injection site) and there were few discontinuations due to adverse effects,37 they felt that LA PrEP would be “worth the pain.”12
The “ambivalent” class comprised about one fifth of the sample. While they did not place the highest relative importance on risk of negative judgment, they placed significantly more on it than the other two groups. This is starkly seen in their willingness to pay $52 for a reduction in stigma (negative judgment), compared with only $9.42 and $1.63 for the other two classes. This group placed relatively the same utility on injection frequency, cost, and side effects. This indicates that this group may be the least likely to find LA PrEP relevant for their lives. However, messaging that places emphasis on the privacy of using a bimonthly injectable product versus DO pills may entice members of this group to consider PrEP for themselves.
A small minority (16%) of respondents were highly “cost-conscious,” placing most importance on out-of-pocket costs. This group is likely to be deterred by high cost alone, even for a product with low treatment burden (minimal perceived side effects and injection frequency) and may be an important barrier particularly among young MSM. In another DCE/LCA analysis on PrEP preferences among US MSM, five classes emerged, with one class dominated by concerns about cost.20 At the time of study design and data collection, LA PrEP costs were still unknown, but in 2022 were priced at $3700 per dose, comparable to 2 months of brand-name DO PrEP, but more expensive than generic DO PrEP.38 While federal insurers in the United States are required to cover PrEP at no cost to patients, it is unclear if that applies to LA PrEP.38 Additional implementation research is needed to determine to what degree cost remains a barrier to patients, what insurance companies cover, and how this affects equitable access and cost-effectiveness.39
The dominance of cost (cost-conscious group) and side effects (side effects-averse group) is found in other preference research as well. A systematic review of DCEs focused on cancer treatment found that most attributes were focused on process (i.e., mode of administration), outcomes (i.e., side effects, effectiveness), and cost (i.e., out-of-pocket amount). The review found that while the minority of all attributes in the included studies were cost-based (5%) or outcomes-based (25%), cost was most significant to patients 67% of the time, and outcomes 81% of the time.40 In a qualitative study, gay and bisexual men were concerned about LA PrEP efficacy, side effects, needles, cost, and frequency of clinic visits.41 For successful uptake and persistence in both groups, comprehensive insurance coverage and other subsidies for LA PrEP are likely key, along with perceptions of outcomes (side effects, effectiveness), and not simply formulation (e.g., DO vs. LA).
This study should be interpreted given its limitations. We recruited the sample online; this may restrict generalizability to people with internet access. Our sample was underrepresentative of Black and Hispanic MSM relative to the disproportionate impact HIV has on those groups. PrEP implementation strategies tailored to these groups' needs and preferences are essential, with recognition of preference heterogeneity and the need for audience segmentation to effectively increase uptake and persistence among various subgroups. Finally, we developed the DCE through literature review and internal expert opinion, but more formative work with the target audience may have found other more salient attributes and levels. This pilot research should inform future data collection and analysis.
These findings can inform future implementation of LA PrEP. Given the disparate relative importance the three groups of MSM placed on out-of-pocket costs, stigma, and perceived side effects, communication strategies should consider audience segmentation to overcome different uptake barriers. For example, easy-to-access subsidy programs that ensure low or no out-of-pocket cost LA PrEP to those with public or no insurance may benefit the cost-conscious potential user. For the side effects-averse group, messaging that addresses fears and provides accurate information about actual side effects may influence their uptake and persistence, more so than subsidies or lower out-of-pocket costs. An LCA conducted among young MSM found that three classes of decision-making approaches related to PrEP modalities, including provider-led, patient-driven, and shared decision-making.22 This indicates that some clients may not ask for PrEP but accept it when offered by providers, or may accept PrEP after a conversation with providers in which they feel that they shared in the decision.
Perceived side effects emerged as an important barrier for uptake of the hypothetical injectable form of LA PrEP for a large proportion of potential MSM users. Minimizing out-of-pocket costs is important to equitable access. The bimonthly doses and privacy and secrecy afforded to LA PrEP versus DO PrEP users may encourage some users. Tailored communication strategies are recommended for the three different groups of potential LA PrEP users.
Acknowledgments
The authors thank Connor Volpi for assistance on the article.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
This study was funded by ViiV Healthcare.
References
- 1. AVAC. PrEP Watch: United States; 2022 [cited November 28, 2022]. Available from: https://www.prepwatch.org/country/united-states/ [Last accessed: June 6, 2023].
- 2. Food and Drug Administration. Supplemental New Drug Application for Truvada for HIV Pre-exposure Prophylaxis. Department of Health and Human Services: Rockville, MD; 2018. [cited November 20, 2019]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2018/021752Orig1s055ltr.pdf [Last accessed: September 19, 2023]. [Google Scholar]
- 3. CDC. Monitoring Selected National HIV Prevention and Care Objectives by Using HIV Surveillance Data-United States and 6 Dependent Areas, 2020. HIV Surveill Suppl Rep 2022(3). Available from: https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-report-vol-26-no-2.pdf [Last accessed: June 6, 2023].
- 4. Sanchez TH, Zlotorzynska M, Sineath RC, et al. National trends in sexual behavior, substance use and HIV testing among United States men who have sex with men recruited online, 2013 through 2017. AIDS Behav 2018;22(8):2413–2425; doi: 10.1007/s10461-018-2168-4. [DOI] [PubMed] [Google Scholar]
- 5. Mayer KH, Agwu A, Malebranche D. Barriers to the wider use of pre-exposure prophylaxis in the United States: A narrative review. Adv Ther 2020;37(5):1778–1811; doi: 10.1007/s12325-020-01295-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Shorrock F, Alvarenga A, Hailey-Fair K, et al. Dismantling barriers and transforming the future of pre-exposure prophylaxis uptake in Young Black and Latinx sexual minority men and transgender women. AIDS Patient Care STDs 2022;36(5):194–203; doi: 10.1089/apc.2021.0222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Sidebottom D, Ekström AM, Strömdahl S. A systematic review of adherence to oral pre-exposure prophylaxis for HIV—How can we improve uptake and adherence? BMC Infect Dis 2018;18(1):581; doi: 10.1186/s12879-018-3463-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. HIV Prevention Trials Network (HPTN). HPTN 083 Study Demonstrates Superiority of Cabotegravir for the Prevention of HIV; July 7, 2020. [cited August 1, 2020]. Available from: https://www.hptn.org/news-and-events/press-releases/hptn-083-study-demonstrates-superiority-cabotegravir-prevention-hiv?utm_source=IAS&utm_campaign=3b6214676e-daily-delegate-11-July&utm_medium=email&utm_term=0_58c4aa5b50-3b6214676e-103223869 [Last accessed: April 12, 2023].
- 9. Landovitz RJ, Donnell D, Clement ME, et al. Cabotegravir for HIV prevention in cisgender men and transgender women. N Engl J Med 2021;385(7):595–608; doi: 10.1056/NEJMoa2101016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. HIV Prevention Trials Network. HPTN 084 Study Demonstrates Superiority of CAB LA to Oral TDF/FTC for the Prevention of HIV; November 9, 2020. [cited November 28, 2022]. Available from: https://www.hptn.org/news-and-events/press-releases/hptn-084-study-demonstrates-superiority-of-cab-la-to-oral-tdfftc-for [Last accessed: April 12, 2023].
- 11. FDA. FDA Approves First Injectable Treatment for HIV Pre-Exposure Prevention; December 20, 2021. [cited August 29, 2022]. Available from: https://www.fda.gov/news-events/press-announcements/fda-approves-first-injectable-treatment-hiv-pre-exposure-prevention [Last accessed: April 12, 2023].
- 12. Kerrigan D, Mantsios A, Grant R, et al. Expanding the menu of HIV prevention options: A qualitative study of experiences with long-acting injectable cabotegravir as PrEP in the context of a Phase II Trial in the United States. AIDS Behav 2018;22(11):3540–3549; doi: 10.1007/s10461-017-2017-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Biello KB, Hosek S, Drucker MT, et al. Preferences for injectable PrEP among young U.S. cisgender men and transgender women and men who have sex with men. Arch Sex Behav 2018;47(7):2101–2107; doi: 10.1007/s10508-017-1049-7. [DOI] [PubMed] [Google Scholar]
- 14. Biello KB, Mimiaga MJ, Santostefano CM, et al. MSM at highest risk for HIV acquisition express greatest interest and preference for injectable antiretroviral PrEP compared to daily, oral medication. AIDS Behav 2018;22(4):1158–1164; doi: 10.1007/s10461-017-1972-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. John SA, Whitfield THF, Rendina HJ, et al. Will gay and bisexual men taking oral pre-exposure prophylaxis (PrEP) switch to long-acting injectable PrEP should it become available? AIDS Behav 2018;22(4):1184–1189; doi: 10.1007/s10461-017-1907-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Levy ME, Patrick R, Gamble J, et al. Willingness of community-recruited men who have sex with men in Washington, DC to use long-acting injectable HIV pre-exposure prophylaxis. PLoS One 2017;12(8):e0183521; doi: 10.1371/journal.pone.0183521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Meyers K, Wu Y, Brill A, et al. To switch or not to switch: Intentions to switch to injectable PrEP among gay and bisexual men with at least twelve months oral PrEP experience. PLoS One 2018;13(7):e0200296; doi: 10.1371/journal.pone.0200296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. John SA, Zapata JP, Dang M, et al. Exploring preferences and decision-making about long-acting injectable HIV pre-exposure prophylaxis (PrEP) among young sexual minority men 17–24 years old. Sci Rep 2023;13(1):5116; doi: 10.1038/s41598-023-32014-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Howe ZW, Norman S, Lueken AF, et al. Therapeutic review of cabotegravir/rilpivirine long-acting antiretroviral injectable and implementation considerations at an HIV specialty clinic. Pharmacotherapy 2021;41(8):686–699; doi: 10.1002/phar.2605. [DOI] [PubMed] [Google Scholar]
- 20. Dubov A, Ogunbajo A, Altice FL, et al. Optimizing access to PrEP based on MSM preferences: Results of a discrete choice experiment. AIDS Care 2019;31(5):545–553; doi: 10.1080/09540121.2018.1557590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Levy ME, Agopian A, Magnus M, et al. Is long-acting injectable cabotegravir likely to expand PrEP coverage among MSM in the District of Columbia? J Acquir Immune Defic Syndr 2021;86(3):e80–e82; doi: 10.1097/qai.0000000000002557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Valente PK, Bauermeister JA, Lin WY, et al. Preferences across pre-exposure prophylaxis modalities among young men who have sex with men in the United States: A Latent Class Analysis Study. AIDS Patient Care STDs 2022;36(11):431–442; doi: 10.1089/apc.2022.0111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Sullivan PS, Satcher Johnson A, Pembleton ES, et al. Epidemiology of HIV in the USA: Epidemic burden, inequities, contexts, and responses. Lancet 2021;397(10279):1095–1106; doi: 10.1016/s0140-6736(21)00395-0. [DOI] [PubMed] [Google Scholar]
- 24. Bridges J, Onukwugha E, Johnson F, et al. Patient preference methods—A patient centered evaluation paradigm. ISPOR Connect 2007;13(6):4–7. [Google Scholar]
- 25. Beckham SW, Crossnohere NL, Gross M, et al. Eliciting preferences for HIV prevention technologies: A systematic review. Patient 2021;14(2):151–174; doi: 10.1007/s40271-020-00486-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hauber AB, Gonzalez JM, Groothuis-Oudshoorn CG, et al. Statistical methods for the analysis of discrete choice experiments: A report of the ISPOR Conjoint Analysis Good Research Practices Task Force. Value Health 2016;19(4):300–315; doi: 10.1016/j.jval.2016.04.004. [DOI] [PubMed] [Google Scholar]
- 27. Zhou M, Thayer WM, Bridges JFP. Using latent class analysis to model preference heterogeneity in health: A systematic review. Pharmacoeconomics 2018;36(2):175–187; doi: 10.1007/s40273-017-0575-4. [DOI] [PubMed] [Google Scholar]
- 28. Sanchez TH, Sineath RC, Kahle EM, et al. The Annual American Men's Internet Survey of Behaviors of men who have sex with men in the United States: Protocol and Key Indicators Report 2013. JMIR Public Health Surveill 2015;1(1):e3; doi: 10.2196/publichealth.4314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Ingram DD, Franco SJ. 2013 NCHS Urban-Rural Classification Scheme for Counties. National Center for Health Statistics. Vital Health Stat 2014;2(166):1–73. [PubMed] [Google Scholar]
- 30. Reed Johnson F, Lancsar E, Marshall D, et al. Constructing experimental designs for discrete-choice experiments: Report of the ISPOR Conjoint Analysis Experimental Design Good Research Practices Task Force. Value Health 2013;16(1):3–13; doi: 10.1016/j.jval.2012.08.2223. [DOI] [PubMed] [Google Scholar]
- 31. ChoiceMetrics. Ngene 1.2 User Manual & Reference Guide, Australia, 2014. [Google Scholar]
- 32. StataCorp. Stata Statistical Software: Release 15. StataCorp LLC: College Station, TX; 2017. [Google Scholar]
- 33. Orme BK. Getting Started with Conjoint Analysis: Strategies for Product Design and Pricing Research, 2nd ed. Research Publishers: Madison, WI; 2009. [Google Scholar]
- 34. Huber J, Orme B, Miller R.. The Value of Choice Simulators. In: Conjoint Measurement: Methods and Applications. (Gustafsson A, Herrmann A, Huber F. eds.) Springer: New York, NY; 2007; pp. 347–362. [Google Scholar]
- 35. Parsons JT, Rendina HJ, Whitfield TH, et al. Familiarity with and preferences for oral and long-acting injectable HIV pre-exposure prophylaxis (PrEP) in a National Sample of Gay and Bisexual Men in the U.S. AIDS Behav 2016;20(7):1390–1399; doi: 10.1007/s10461-016-1370-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Goedel WC, Halkitis PN, Greene RE, et al. Correlates of awareness of and willingness to use pre-exposure prophylaxis (PrEP) in gay, bisexual, and other men who have sex with men who use geosocial-networking smartphone applications in New York City. AIDS Behav 2016;20(7):1435–1442; doi: 10.1007/s10461-016-1353-6. [DOI] [PubMed] [Google Scholar]
- 37. Landovitz RJ, Li S, Grinsztejn B, et al. Safety, tolerability, and pharmacokinetics of long-acting injectable cabotegravir in low-risk HIV-uninfected individuals: HPTN 077, a phase 2a randomized controlled trial. PLoS Med 2018;15(11):e1002690; doi: 10.1371/journal.pmed.1002690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Highleyman L. Injectable PrEP Is Here. What Does This Mean for You? 2022. [cited November 28, 2022]. Available from: https://www.sfaf.org/collections/beta/injectable-prep-is-here/ [Last accessed: April 12, 2023].
- 39. Neilan AM, Landovitz RJ, Le MH, et al. Cost-effectiveness of long-acting injectable hiv preexposure prophylaxis in the United States: A cost-effectiveness analysis. Ann Intern Med 2022;175(4):479–489; doi: 10.7326/m21-1548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Bien DR, Danner M, Vennedey V, et al. Patients' preferences for outcome, process and cost attributes in cancer treatment: A systematic review of discrete choice experiments. Patient Centered Outcomes Res 2017;10(5):553–565; doi: 10.1007/s40271-017-0235-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Tran NK, Martinez O, Scheim AI, et al. Perceived barriers to and facilitators of long-acting Injectable HIV PrEP use among Black, Hispanic/Latino, and White Gay, Bisexual, and other men who have sex with men. AIDS Educ Prev 2022;34(5):365–378; doi: 10.1521/aeap.2022.34.5.365. [DOI] [PubMed] [Google Scholar]