Introduction

Tenofovir-based oral PrEP is the only biomedical approach currently approved to prevent HIV. Other ARV-based prevention modalities—including vaginal rings, microbicides, alternate oral agents, injectables—are currently in clinical development, with the long-acting integrase inhibitor Cabotegravir Long-Acting (Cab-LA) furthest along. With the prevention efficacy of Cab-LA under investigation in a global study of men who have sex with men (MSM) and transgender women who have sex with MSM [1], a series of sociobehavioral studies have begun assessing attitudes towards long-acting injectable PrEP (LAI-PrEP) among gay men in the US and other settings [1–9]. These studies show variable enthusiasm for an injectable PrEP product, with positive attitudes towards LAI PrEP (as measured by acceptability, preference or willingness to use) ranging from 25.2% [3] to 80.0% [6]. This variability is likely due to heterogeneity in study design, wording of survey questions, different explanation of LAI-PrEP, and/or real differences between populations. However, taken together, these studies do suggest that the addition of a second PrEP modality would likely attract PrEP-naïve users for whom the daily oral pill was not acceptable and consequently increase overall PrEP use [9]. While useful in establishing acceptability of attributes of a particular product-type at the most basic level, clinical trials are limited in the depth with which they can collect data on the acceptability of a specific biomedical prevention technology among potential users.

For this new biomedical prevention strategy to be effective, it is important to understand not only interest in Cab-LA during clinical development, but also its likelihood of being adopted by real users and the factors that might underlie decisions about uptake. People who are already on oral PrEP present a particularly informative population in which to study this question, as these individuals have experienced the physical and psychosocial benefits and challenges of daily oral PrEP use first-hand. For example, they may have experienced a decrease in anxiety around sex [10, 11], an increase in intimacy with partners [10, 12], an increase in sexual satisfaction [10, 13], worry about missing doses [14, 15], changes in frequency of condomless sex [10, 16, 17], infection with an STI [18], or increases in stigma related to sexual practices and choices in HIV prevention methods [13, 19]. In addition, current oral PrEP users understand the logistics of taking PrEP, including regular clinic visits [20], timely pharmacy refills [21], storing of pills [22, 23], and integrating the medication into their routines and lifestyles [22, 23]. The experience of integrating oral PrEP into their daily lives allows them to provide answers to questions about a hypothetical new injectable formulation of PrEP, informed by real world practice. Some of these experiences may continue to exist with the use of an injectable product, some may be overcome by its development, and some distinct new experiences are likely to arise as well.

The goal of this study was to investigate specific factors that might help direct a patient-provider shared decision-making process to optimize the choice of biomedical HIV prevention method. Such tools have been utilized in the contraceptive field [24–27] and our objective was to generate preliminary data that could usefully differentiate between current oral PrEP users who would or would not switch to LAI-PrEP. Specifically, the study focused on product-related and psychosocial factors that may impact implementation of LAI-PrEP. We tested measures that we hypothesized would help us understand who would be a good candidate for an intra-muscularly-injected versus an orally-administered PrEP product. Hypothesized domains included: 1) product-related factors, such as the physical experience of the PrEP modality and logistical features of administration; and 2) psychosocial factors, like agency and control, psychosocial aspects of adherence, and self-identification as an early-adopter or risk taker.

Materials and methods

Survey design and measurement

Participants were given the following written description of the “PrEP shot”:

“an anti-HIV medication that is made into an injection (shot) that can remain in the blood for a long period of time. You take one month of daily pills to make sure you don’t have a reaction to the medication. Then come to a clinic every three months to get two shots in the buttocks by a medical provider.”

Following this description, participants completed self-report survey items. The survey instrument included a series of close-ended and open-ended questions developed specifically for this study, based on the research team’s experience investigating the use of daily oral PrEP and injectable PrEP among users [28, 29] and clinical trial participants [30]. Because these measures were developed for this project, and because few (if any) acceptability measures exist specific to LAI PrEP, we have provided our complete list of measures in Table 1. We used four types of items and response choice sets to measure participant attitudes, including: a) pro/con judgements, in which participants read characteristics and were asked to rate the extent to which each was a “pro” or a “con” of the shot compared to the PrEP pill; b) Polar or “A versus B” items, in which participants are presented with two opposite statement and were asked to rate themselves between them; c) Likert-type scale items, in which participants responded to statements about injections and pills on a seven-point scale ranging from “very challenging” to “very easy”; and d) classic Likert-scale items, in which participants responded to statements on a seven-point scale ranging from “strongly agree” to “strongly disagree”. As demonstrated in Table 1, measures were divided into “product-level” factors, including physical experience and logistics/convenience and “psychosocial factors’, which included feelings of control, psychosocial attitudes/preferences, and feelings about adherence.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Hypothesized predictors of LAI PrEP switch intentions.

https://doi.org/10.1371/journal.pone.0200296.t001

Willingness to switch.

After completing these survey measures, participants were asked: “If the PrEP shot were FDA approved and available for the same cost as the daily PrEP pill, how likely would you be to switch to the PrEP shot?” on a 5-point scale from 1, definitely switch to 5, definitely not switch with mid-point “unsure”. Those who would definitely or probably switch were categorized as “switchers”, those who were unsure, or would probably or definitely not switch were categorized as “non-switchers”.

Demographic variables.

As part of the survey, participants were asked to report their age, race/ethnicity, sexual orientation, education, employment, and income. Additional items collected information on living situation and partnership characteristics.

We also asked three open-ended questions to elicit more details on participants’ perspectives on the relative pros and cons of the PrEP shot as compared to daily oral PrEP: (1) What do you think is the biggest benefit or “pro” of the PrEP shot compared to the PrEP pill?; (2) What do you think is the biggest drawback or “con” of the PrEP shot compared to the PrEP pill?; (3) Please tell us a little bit about why you would or wouldn’t switch.

We included open-ended questions for several reasons. First, given the preliminary nature of the constructs and measures we developed to capture them, we wanted participants to be able to raise issues that we may not have previously considered and provide more depth to their survey responses [31, 32]. Second, we anticipated that some items might lack variability in responses due to the potential homogeneity in a cohort of early oral PrEP adopters. Such a lack of variability may impact on the quantitative analysis and significance of findings; therefore, we included open-ended fields that would allow us to pick up common themes.

Across the three open questions, 97.5% of respondents provided answers. Text entered by participants in response to the three open-ended questions was organized thematically around the motivators and barriers to switching identified in the statistical analysis. One researcher coded and categorized the data. Two other researchers reviewed the coding and final coding was based on group discussion. Excerpts were chosen to provide a more detailed explanation of the respondents’ answers to the survey items.

Results

Intention to switch from oral to LAI-PrEP

The sample was largely gay (89.5%), ethnically diverse (46.7% non-white), and fairly well distributed in terms of education, income, and insurance status (Table 2). Almost 92% of the sample (n = 96) were still taking PrEP. Two-thirds of the sample reported that they would definitely (36.2%) or probably (30.5%) switch from daily oral PrEP to LAI-PrEP. About one fifth (21.0%) were unsure whether or not they would switch, a tenth (9.5%) would probably not switch, and 3 out of 106 respondents (2.9%) would definitely not switch.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Sample characteristics (n = 105).

https://doi.org/10.1371/journal.pone.0200296.t002

Factors associated with intention to switch

Intention to switch to LAI from oral PrEP was not associated with age, race/ethnicity, sexual orientation, education, insurance status, factors related to partnership type, or whether or not the participant was a current PrEP user. However, income was associated with intention to switch; specifically those with moderate income ($30,000-$50,000) were more likely to consider switching, compared to those who reported highest income (Table 3). Income was included in the final model.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Factors associated with intention to switch from oral to injectable PrEP (n = 105).

https://doi.org/10.1371/journal.pone.0200296.t003

In bivariate analysis, intention to switch to LAI-PrEP was positively associated (p < .10) with four product-level factors and five psychosocial factors (Table 3). Analysis of the qualitative data allowed us to include illustrative quotes for the bivariate factors.

Discussion

This is the first study to directly ask current oral PrEP users whether they would switch to an LAI-PrEP product, were it to become available, and probe the product-level and psychosocial aspects of their switch intentions. As such, these experienced PrEP users offer a unique perspective from which to assess the pros and cons of LAI-PrEP compared to daily oral PrEP. These men were among the earliest adopters of daily oral PrEP—enrolling into a PrEP demonstration within the first 24 months after FDA approval of daily oral Truvada. This makes the beliefs and attitudes of this group of men particularly relevant to the evaluation of the pros and cons of a new PrEP-modality, offering perspectives that are informed by the physical and psychological experience of oral PrEP use.

We developed twenty-four measures in five domains that we hypothesized would allow us to differentiate current oral PrEP users who would be motivated to switch to injectable PrEP from those who were likely to continue using daily oral PrEP. Measures in all five domains showed bivariate associations and four domains retained significance in the multivariable analysis.

Of the two product-level domains, physical experience was not, surprisingly, associated with switching intentions. Attitudes towards pain, injections, or pill swallowing did not differentiate those who had higher odds of intending to switch from those who were satisfied with oral PrEP in the multivariable model. However, this may be because none of our quantitative measures queried attitudes towards needles specifically, using instead, language of “shots” and “injections.” The literature on aversion to injections and needle phobia distinguishes between psychological concerns (fear of fainting, disgust), vasovagal concerns (fainting, loss of consciousness), and needle-phobia (avoidance, distress, fear) [33, 34]. This study appears to bear out this distinction as items eliciting reactions to injections were not associated with the outcome, whereas many non-switchers mentioned needles as a barrier to LAI-PrEP in their open-ended responses. Future research could use standardized scales to measure aversion to injections as compared to needles to assess whether scores on such scales could be useful in identifying someone as a candidate for injectable PrEP. These data suggest that while such concerns may not determine patient preferences, they should be discussed by providers as a critical component of a patient’s decision-making process.

Logistical concerns mattered for participants; concerns about the frequency of injections and clinic visits were both negatively associated with intentions to switch, whereas individuals who were not bothered by these return visits were more likely to report wanting to switch. Prior experiences during clinic visits for daily oral PrEP or other interactions with the health system may predispose individuals positively or negatively towards more frequent visits for an injectable agent. When working with patients around decision-making, providers should keep in mind that such logistical concerns may have farther-reaching influences on patients’ daily lives, and therefore may be more important than product-specific product factors such as concerns about injections. Deepening our understanding of what is difficult about clinic visits for injections—the frequency of visits, the lack of weekend and night-time hours at clinics, or the intimacy of such injections—would help planning for future implementation of LAI-PrEP as well as clinical development of future biomedical HIV prevention products.

Psychosocial-level factors also appeared to be quite important in people’s attitudes towards LAI-PrEP. The “pro” of not needing to take a daily oral pill (i.e., adherence) has been noted as an advantage of an LAI-PrEP modality [3, 6, 7]. However, in our data, endorsement of the statement “not having to take a pill every day” was not significantly associated with intention to switch, while the item “taking PrEP pills every day feels like an emotional burden” was. The latter item captures the psychological aspect underlying the “adherence” theme—that taking a pill every day, while doable, does take an emotional toll that seems to be operating in two ways: for some, the act of pill-taking appears to be a reminder of the risk of HIV; for others, it highlights the absence of sex. While the psychological burden of pill-taking has not been previously reported in the PrEP literature, it has been measured in the context of ARV and ARV adherence [35–37]. LAI-PrEP offers an alternative that avoids the daily psychological burden of PrEP use. Providers working with patients on decision-making may want to acknowledge the emotional burden of pill-taking as a factor in evaluation of prevention options.

Under the theme of agency, or control, the idea that taking PrEP made users “feel responsible” was most strongly associated with switching in the multivariable model. Factors related to a loss of control when switching from pill-taking to injectable PrEP were associated at the bivariate level only. Mental health benefits arising from an enhanced sense of self-efficacy and agency among oral PrEP users have been reported elsewhere [38–44], and is conceptually related to this finding. It is possible that patients who felt most affirmed by their oral PrEP use were also those who were most open to trying a new approach. Focusing on taking responsibility and being in control of HIV prevention may be an important motivating factor for providers to discuss with their patients regardless of the specific modality chosen for PrEP. It is also interesting to contrast this focus on agency with the above findings around the emotional burden of pill-taking. It may be that the ways in which individuals understand and make meaning from their PrEP behavior will be the most important factor in PrEP uptake, adherence, and persistence across modalities.

Conclusion

Psychosocial factors appear to motivate intentions to switch to injectable PrEP among experienced PrEP users. In anticipation of the availability of a LAI-PrEP product, there is a need for rigorous behavioral and social science research to inform the development of tools to support decision-making around PrEP modalities that take into account the psychosocial factors identified in this study. Such factors—especially issues around the meaning of pill-taking for patients as an emotional burden versus a symbol of agency and control—will be critical to the development of patient education materials and shared decision-making processes with providers.

Acknowledgments

READY is funded by NIH (R01MH106380; Golub, PI) and funded the contributions of Drs. Golub and Meyers and Ms. Wu and Ms. Brill. Dr. Sandfort’s contribution to this manuscript was supported by NIMH center grant P30-MH43520 (P.I.: Robert Remien, PhD) to the HIV Center for Clinical and Behavioral Studies. We gratefully acknowledge the efforts of Dr. Asa Radix, Stephanie Peña, and Anthony Surace, who all contributed to the success of this project. We also appreciate the contributions of two anonymous reviewers, whose critique and comments enhanced the quality of the final manuscript. Finally, we are grateful to the participants who gave their time and energy to READY.