PrEP willingness and adherence self-efficacy among men who have sex with men with recent condomless anal sex in urban China

Abstract

This study investigates PrEP willingness, adherence self-efficacy and potential impact of PrEP among HIV-negative, Chinese men who have sex with men (MSM; n = 622) with recent condomless anal sex. Facilitative factors of PrEP willingness included migrant status, sexual risk, and prior PrEP use, whereas barriers included concerns over being treated as an HIV/AIDS patient, recent HIV testing, identity concealment, and HIV prevention service usage. Adherence self-efficacy was associated with PrEP knowledge and confidence in PrEP efficacy of HIV prevention. A total of 39.3% anticipated increase in sex partners, 25.6% anticipated decrease in condom use, and 38.0% anticipated increased HIV testing following PrEP uptake. Results suggest a two-step approach to (1) promote PrEP acceptance among Chinese MSM and (2) enhance adherence and risk monitoring among PrEP-willing MSM. Efforts to reduce stigma, incorporate PrEP in the HIV prevention continuum, and increase PrEP knowledge will be crucial to optimize PrEP implementation.

Introduction

Numerous clinical trials have demonstrated the safety and efficacy of HIV antiretroviral pre-exposure prophylaxis (PrEP) for HIV prevention in several countries [1]. Regions that have implemented PrEP have witnessed significant impact on controlling the HIV epidemic among key populations, including men who have sex with men (MSM) [2, 3]. The World Health Organization (WHO) has recommended PrEP for populations with substantial risk of HIV infection (> 3% incidence) [4]. In China, the most populous country in the world, MSM have had the highest HIV infection rate among all key populations since 2010 [5]. The national prevalence of HIV infection was 6.9% among MSM in 2018 [6], with rates even higher in urban areas. Among MSM in China, sexual risk has been identified as the primary route of HIV transmission [7]. Condomless anal sex (CAS) is prevalent; a recent meta-analysis estimated 34–60% Chinese MSM engaged in CAS in the past year [8].

The scaling up of PrEP could be a promising and effective public health strategy to significantly reduce the burden of HIV among Chinese MSM. One epidemiological modelling study forecasted 1.1–3.0 million new HIV infections in China in the next two decades, and moderate coverage of PrEP (50%) could prevent 0.17–0.32 million of these infections [9]. There is a government-funded, multi-center clinical trial of oral PrEP among Chinese MSM underway [10, 11], and recently, tenofovir/emtricitabine has been approved for PrEP use in China [12]. As such, there is a crucial need to understand factors that can facilitate or impede PrEP implementation with MSM as a priority population.

Recent studies have documented varied levels of PrEP willingness among the general MSM population (i.e., no risk criteria) in China [13–18]. However, there are several gaps in the current literature on PrEP willingness and adherence self-efficacy among Chinese MSM. First, limited research has focused on MSM eligible for and may benefit from PrEP under current guidelines (e.g., engaged in CAS in recent months) [19, 20], despite the more urgent need to target this population [21]. Specifically, a recent cross-sectional research with 708 MSM in four cities in China found that behavioral indication for PrEP (e.g., sexual risk behaviors) not associated with self-perception of PrEP candidacy or progression through the motivational PrEP cascade [22, 23], suggesting the need to further understand factors that facilitate and impede PrEP willingness and uptake within Chinese MSM behaviorally indicated for PrEP. Second, limited research has documented low self-efficacy for adherence to daily PrEP uptake among Chinese MSM [24]. Understanding adherence self-efficacy among MSM who are willing to use PrEP (i.e., PrEP-willing MSM) will be important to anticipate potential barriers of PrEP uptake in future implementation. Third, psychosocial and behavioral correlates of PrEP willingness and adherence self-efficacy among MSM in China need further exploration. Specifically, low PrEP knowledge has been documented [13–17, 25], yet its role in determining willingness and adherence self-efficacy among potential PrEP users has not been investigated. Recent HIV risk and prevention behaviors may also affect PrEP willingness. CAS has been associated with PrEP willingness among the general MSM population [13, 16, 17], yet the role of elevated risk (e.g., sex under the influence of substance, sex with an HIV+ partner) has not been examined among Chinese MSM who may benefit from PrEP to reduce HIV infection risk given recent CAS. With regard to HIV prevention behaviors, history of HIV testing has been positively associated with PrEP willingness among general Chinese MSM [15]. Other preventive behaviors, including prior usage of PrEP and usage of internet-based HIV prevention services (e.g., via WeChat – the most widely used mobile phone application in China), may be other potential correlates of PrEP willingness. Furthermore, evaluating the potential impact of PrEP uptake on sexual risk (i.e., risk compensation) and HIV testing behaviors will also be crucial to guide comprehensive HIV prevention efforts.

To fill these gaps, the current study examined factors associated with PrEP willingness and adherence self-efficacy among a sample of HIV-negative MSM with recent CAS in urban China. Specifically, we investigated the willingness to use oral PrEP among MSM if it were to be provided for free. We explored potential correlates of PrEP willingness, including demographic factors (e.g., education, migrant status, sexual orientation), PrEP-related social cognitive factors including PrEP awareness, PrEP knowledge, and specific concerns about PrEP usage, and recent sexual risk and HIV prevention behaviors. Among PrEP-willing MSM, we examined their PrEP adherence self-efficacy and explored its demographic, psychosocial, and behavioral correlates. Finally, among PrEP-willing MSM, we described potential impacts of PrEP use on sexual risk behaviors and HIV testing following PrEP uptake.

Methods

Procedure

We analyzed data from a multisite study of oral HIV self-testing and prevention for Chinese MSM with recent CAS, conducted in Hefei (the capital of Anhui, northcentral China), Chengdu, (the capital of Sichuan, southwest China), and Guangzhou (the capital of Guangdong, southeast China). The study was approved by the Institutional Review Board (IRB) at Anhui Medical University. Participants were recruited at three non-government organizations (NGOs) that provide HIV prevention and education services in these cities as well as via MSM-focused social media in China (i.e., Blued). Eligibility criteria included: (a) male at birth; (b) aged 18 or older; (c) engaged in at least one condomless anal sex with a man in the past 6 months; (d) self-reported HIV-negative or status unknown; (e) willing to self-administer an oral HIV self-test.

Potential participants were approached by a trained research staff in targeted community venues (e.g., bars, clubs, other commercial venues) and online settings (e.g., MSM-themed chat-rooms and dating sites) and provided with an overview of the study. Initial screening was conducted by phone, in person, or online. If eligible and interested in the study, individuals were invited to meet at the collaborating NGO where they underwent written informed consent, and then were escorted to a private room to watch a video demonstration of oral HIV-self testing and asked to self-administer an oral HIV self-test. Any individuals who tested preliminary HIV-positive were excluded from further participation in the study and immediately linked to HIV confirmatory testing at the local health department. HIV negative participants then completed a computer-administered survey following receipt of their HIV-negative self-test result. Data was collected between June and October, 2018.

Measures

PrEP related outcomes

Regarding PrEP awareness and willingness, participants were first asked if they have heard of taking daily antiretroviral medication to lower risks of contracting HIV (yes/no). All participants were provided with a brief description of PrEP: “PrEP (pre-exposure prophylaxis) is a medication for people who do not have HIV but who may benefit from reduced risk of HIV. Taking the pill every day substantially reduces your risk of being infected with HIV.” Then, participants were presented with a question on oral PrEP willingness: “If daily oral intake of PrEP can achieve 90% effectiveness to prevent HIV and you can get PrEP for free, how likely would you take PrEP?” Participants who were “absolutely willing” or “probably willing” to take PrEP were categorized as “willing” and compared against the remainder (those who were “uncertain,” “probably not willing,” and “absolutely not willing”).

PrEP adherence self-efficacy was measured by an adapted and abbreviated version of the Condom Use Self-Efficacy Scale [26], which itself has been validated among Chinese MSM [27, 28]. The adapted measure consisted of six items regarding daily and sustained PrEP use (e.g., “I can continue to take PrEP even if my sex partner is against it.” “I can continue to with my PrEP regimen even if I experience some side effects.”). Responses for each item were dichotomized as confident (1 for those responded as “I am confident”) or not (0 for those responded as “I’m not confident” or “not sure”) and summed into a composite score ranging from 0 to 6. Higher total scores represented higher adherence self-efficacy. This newly adapted PrEP adherence self-efficacy has not been validated. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) indicated a one-factor structure, with all items having sufficient loadings (≥ .40, all items had a standardized loading > .75). Cronbach’s alpha was 0.93.

Three items measured potential impact of PrEP on sexual risk and HIV testing, including (1) “If using PrEP to prevent HIV, how do you think your likelihood of using condoms during sex may change compared to before?” (2) “If using PrEP to prevent HIV, how do you think the number of your sex partners may change compared to before?” and (3) “If using PrEP to prevent HIV, how do you think the frequency of your HIV testing may change compared to before?” Participants responded to each item on a 5-point scale, including “definitely increase”, “likely increase”, “same as before,” “likely decrease,” and “definitely decrease”.

Predictor Variables

Demographic information was collected, including participants’ education background, monthly income, residence/“hukou”, marital and relationship status, and history of sextually transmitted disease. Participants reported their sexual orientation as well as sexual orientation concealment, measured via the Concealment Behavior Scale on American Men’s Internet Survey [29], which asked participants whether or not they have disclosed their sexual relations with men to seven types of people, including LGB friends, parents, friends, relatives, co-workers, and medical workers (1 = concealed; 0 = disclosed). Higher total scores indicate higher levels of concealment.

PrEP knowledge was measured by an exploratory 8-item scale that queried participants’ knowledge on key facts about PrEP, informed by China CDC guidelines on HIV prevention for MSM [30]. Sample items included “PrEP can replace the function of condoms in sex” and “Besides HIV, PrEP can prevent other sexually transmitted illnesses.” Participants were asked to respond each item with “True”, “False,” and “I don’t know.” The total score was the sum score of participants’ correct answers. Cronbach’s alpha was 0.82.

Concerns of PrEP use.

Five items assessed potential concerns regarding PrEP use (1 = has concern; 0 = not concerned), including side effects, the efficacy of PrEP, worries about discussing one’s sex life with a physician, worries about being treated by others as an HIV/AIDS patient, and being rejected by men regarding sexual relations (see Table 2).

Table 2.

Multivariate analysis of MSM’s PrEP willingness and adherence self-efficacy

	Full sample (N = 622) DV: PrEP willingness				PrEP-willing MSM (N = 402) DV: PrEP adherence self-efficacy
	Multivariate logistic regression				Multivariate regression
	B	aOR	95%CI	p	B	95%CI	beta	p
Categorical Variables
Demographic characteristics
Education level
High school or less	0.25	1.29	[0.81, 2.07]	.293
More than high school	ref
Hukou (residence)
Migrant	0.70	2.01	[1.38, 2.92]	< .001***
Local	ref
PrEP awareness
Heard of PrEP					0.32	[−0.16, 0.80]	0.06	.192
No
Yes
Concerns for PrEP usage
The efficacy of PrEP in preventing the transmission of HIV virus					−0.80	[−1.50, −0.09]	−0.10	.027*
Not concerned
Has concern
People may treat me as an HIV/AIDS patient
Not concerned	ref
Has concern	−0.60	0.55	[0.38, 0.80]	.002**
Sexual risk behaviors
Had an HIV+ sex partner in the past six months
No	ref
Yes	1.43	4.19	[1.82, 11.43]	.002**
Sex under influence of substance in past six months
No	ref
Yes	0.94	2.57	[1.67, 4.03]	< .001***
HIV prevention behaviors
Prior use of PrEP	ref
No	1.82	6.17	[1.98, 27.40]	.005**
Yes
Tested for HIV in past 6 months
No	ref
Yes	−0.68	0.50	[0.34, 0.74]	< .001***
Continuous Variables
Sexual orientation concealment	−0.19	0.83	[0.70, 0.96]	.015*
PrEP knowledge					0.34	[0.24, 0.44]	0.32	< .001***
Use of WeChat for HIV prevention	−0.17	0.84	[0.72, 0.98]	.032*

Note. Only variables emerged as significant on the bivariate level (see Table 1) were entered into multivariate regression analysis. Binary variables were dummy-coded (0 = No; 1 = Yes). aOR= adjusted odds ratio; CI = confidence interval;

^*p < .05,

^**p < .01,

^***p < .001

Sexual risk behaviors were indicated by three single-item questions (1= yes; 0 = no), including (1) whether participants had receptive condomless anal sex with a male partner in the past six months, (2) whether participants had sex under influence of substance (poppers, marijuana, psychedelics such as 5-MeO, etc.) in the past six months, and (3) whether participants had a sex partner living with HIV in the past six months.

HIV preventive behaviors included (1) a single-item question assessing HIV testing history in the past six months and (2) a single-item question assessing participants’ previous PrEP use for preventive purposes. In addition, we asked participants about their use of WeChat for HIV prevention services to: (1) seek knowledge about HIV; (2) consult with someone on HIV risk; (3) schedule HIV testing; and (4) access knowledge regarding HIV treatment and care. Higher total scores indicate higher usage of WeChat-based HIV prevention services.

Statistical Analysis

Descriptive statistics were calculated to understand PrEP awareness and willingness in the full sample, and to examine adherence self-efficacy and potential impact on sexual risk and HIV testing behaviors in the subsample of PrEP-willing MSM. Bivariate analyses were calculated (Chi-square statistics, independent sample t-test, analysis of variance, and Pearson’s correlation) to test for associations between variables of interest and PrEP willingness and adherence self-efficacy. Regression analyses were performed to examine predictors of PrEP willingness and adherence self-efficacy, using logistic regression and multiple regression, respectively. Predictor variables were only entered into the model if they were significantly associated with the outcome in bivariate analyses. Multivariable analyses complied with assumptions regarding variable distribution and there was no evidence for collinearity. All analyses were conducted in R, where statistical significance was determined with an alpha level of .05.

Results

Sample demographic characteristics

A total of 622 MSM across three cities completed the survey. Participants’ age ranged from 18 to 62 (Mean = 29.75, Median = 28, Standard Deviation [SD] = 8.32). Participants’ education level varied: 53.54% (n = 333) had college degree or higher, 25.24% had associate degree (n = 157), 16.7% completed high school (n = 104), and 4.50% had junior high school or below (n = 28). About half of participants (46.3%) were migrants (i.e., without local “hukou”). The majority of MSM received HIV testing before (n = 530, 85.2%) and at least one time testing in the past 6 months (n = 397, 63.8%), in addition to testing provided at the study sites. Participants reported a median of 8 lifetime male anal sex partners (Mean = 17.12, SD = 37.18, Range = [1, 500]). Less than half participants reported lifetime sex with a female (n = 260, 41.8%), and 11.1% (n = 69) reported sex with a female in the past six months. Detailed demographic information is presented in Table 1.

Table 1.

Demographic characteristics, and bivariate analyses on PrEP willingness and PrEP adherence self-efficacy

		Full sample (N = 622)				PrEP-willing MSM (N = 402)
		Comparison on PrEP willingness				Comparison on PrEP adherence self-efficacy
Categorical variables	N (%)	Not willing n (%)	Willing n (%)	χ2	p	Adherence self-efficacy M (SD)	t/F	p
Demographic characteristics
Education level				4.37	.037*		0.14	.886
High school or less	132 (21.2)	36 (27.3)	96 (72.7)			2.97 (2.62)
More than high school	490 (78.8)	184 (37.6)	306 (62.4)			2.93 (2.52)
Monthly income in most recent six months				1.98	.576		1.75	.156
< 1,500 Yuan ($214)	93 (15.0)	34 (36.6)	59 (63.4)			3.20 (2.52)
1,501–5,300 Yuan ($214- $757)	248 (39.9)	80 (32.3)	168 (67.7)			2.74 (2.50)
5,301–11,000 Yuan ($757- $1,571)	203 (32.6)	75 (36.9)	128 (63.1)			2.88 (2.56)
> 11,000 Yuan ($1,571)	78 (12.5)	31 (39.7)	47 (60.3)			3.64 (2.57)
Hukou (residence)				26.08	< .001***		−0.73	.469
Have hukou	334 (53.7)	71 (24.7)	217 (75.3)			2.88 (2.58)
Without hukou	288 (46.3)	149 (44.6)	185 (55.4)			3.06 (2.49)
Sexual orientation				2.92	.404		0.04	.846
Gay	506 (81.4)	185 (36.6)	321 (63.4)			2.96 (2.51)
Bisexual	93 (15.0)	29 (31.2)	64 (68.8)			2.95 (2.70)
Heterosexual	3 (0.5)	0 (0)	3 (100)			2.33 (2.52)
Unsure	20 (3.2)	6 (42.9)	14 (57.1)			3.21 (2.64)
Marital status				1.31	.726		0.003	.956
Unmarried and no female partner	484 (77.8)	171 (35.3)	313 (64.7)			2.97 (2.53)
Unmarried and have stable female partner	27 (4.3)	11 (40.7)	16 (59.3)			2.69 (2.73)
Married	71 (11.4)	22 (31.0)	49 (69.0)			2.98 (2.62)
Divorced	40 (6.4)	16 (40.0)	24 (60.0)			2.96 (2.58)
Current male partner status				2.42	.120		1.59	.113
Have a stable male partner	282 (45.4)	130 (38.2)	210 (61.8)			3.15 (2.54)
Without a stable male partner	340 (54.7)	90 (31.9)	192 (68.1)			2.75 (2.53)
Sexually transmitted disease				2.90	.089		−0.69	.493
Does not have an STI diagnosis	542 (87.1)	199 (36.7)	343 (63.3)			2.92 (2.54)
Have an STI diagnosis (e.g., syphilis)	80 (12.9)	21 (26.2)	59 (73.8)			3.17 (2.53)
PrEP awareness
Heard of PrEP				0.16	.691		−2.68	.008**
No	271 (43.6)	93 (34.3)	178 (65.7)			2.58 (2.45)
Yes	351 (56.4)	127 (36.2)	224 (63.8)			3.26 (2.58)
Concerns for PrEP usage
Side effects				0.90	.342		1.31	.194
Not concerned	545 (87.6)	197 (36.1)	348 (63.9)			3.03 (2.51)
Has concern	77 (12.4)	23 (29.9)	54 (70.1)			2.52 (2.68)
The efficacy of PrEP in preventing the transmission of HIV virus				0.13	.724		2.70	.009**
Not concerned	546 (87.8)	195 (35.7)	351 (64.3)			3.09 (2.51)
Has concern	76 (12.2)	25 (32.9)	51 (67.1)			2.06 (2.56)
May need to talk about my sex life with a physician				1.13	.287		0.43	.668
Not concerned	202 (32.5)	65 (32.2)	137 (67.8)			3.04 (2.59)
Has concern	420 (67.5)	155 (36.9)	265 (63.1)			2.92 (2.52)
People may treat me as an HIV/AIDS patient				12.86	< .001***		0.89	.377
Not concerned	412 (66.2)	125 (30.3)	287 (69.7)			3.03 (2.54)
Has concern	210 (33.8)	95 (45.2)	115 (54.8)			2.78 (2.55)
Men may refuse to have sex with me				2.06	.151		−0.51	.610
Not concerned	297 (47.7)	96 (32.3)	201 (67.7)			2.90 (2.58)
Has concern	325 (52.3)	124 (38.2)	201 (61.8)			3.02 (2.50)
Sexual risk
Sex with a woman in past six months				0.60	.438		−0.23	.823
No	533 (88.9)	199 (36.0)	354 (64.0)			2.95 (2.52)
Yes	69 (11.1)	21 (30.4)	48 (69.6)			3.04 (2.69)
Ever had receptive condomless anal sex in past six months				1.81	.178		0.359	.720
No	304 (48.9)	99 (32.6)	205 (67.4)			3.00 (2.51)
Yes	318 (51.1)	121 (38.1)	197 (61.9)			2.91 (2.57)
Sex under influence of substance in past six months (poppers, marijuana, psychedelics such as 5-MeO, etc.)				20.82	< .001***		−0.94	.350
No	450 (72.4)	184 (40.9)	266 (59.1)			2.88 (2.56)
Yes	172 (27.7)	36 (20.9)	136 (79.1)			3.13 (2.51)
Had an HIV+ sex partner in past six months				14.08	< .001***		0.64	.524
No	568 (91.3)	214 (37.7)	354 (62.3)			2.99 (2.52)
Yes	54 (8.7)	6 (11.1)	48 (88.9)			2.73 (2.68)
HIV prevention behaviors
HIV testing in past six months				14.85	< .001***		−0.57	.567
Not tested for HIV in past six months	225 (36.2)	57 (25.3)	168 (74.7)			2.88 (2.50)
Tested for HIV	397 (63.8)	163 (41.1)	234 (58.9)			3.02 (2.57)
Used PrEP to prevent HIV after high risk sex				6.20	.013*		−1.42	.166
No	597 (96.0)	217 (36.5)	378 (63.5)			2.91 (2.63)
Yes	25 (4.0)	3 (11.1)	24 (88.9)			3.71 (2.66)
Continuous variables	M (SD)	M (SD)	M (SD)	t	p	r	t	p
Age	29.75 (8.32)	30.28 (8.16)	29.46 (8.41)	1.20	.233	.01	0.22	.829
Sexual orientation concealment	5.97 (1.34)	6.23 (1.08)	5.83 (1.45)	3.90	< .001***	0.06	1.30	.195
PrEP knowledge	5.12 (2.50)	5.11 (2.70)	5.12 (2.39)	−0.05	.962	.34	7.22	< .001***
Use of WeChat for HIV prevention	1.77 (1.21)	2.03 (1.17)	1.63 (1.20)	4.02	< .001***	.06	1.15	.252

Note. M = Mean; SD = Standard Deviation;

^*p < .05,

^**p < .01,

^***p < .001

PrEP awareness and willingness.

A total of 56.4% of participants were aware of PrEP and 64.6% indicated a willingness to take oral PrEP if provided for free (34.2% reported “may use it” and 30.4% reported “will absolutely use it”). In multivariate analysis (Table 2), willingness to take oral PrEP was positively associated with being a migrant, aOR (adjusted odds ratio) = 2.01, 95%CI (confidence interval) = [1.38, 2.92], prior PrEP use (aOR = 6.17 [1.98, 27.40]), sex under the influence of substance in the past six months, (aOR = 2.57 [1.67, 4.03]), and having an HIV+ partner in the past six months (aOR = 4.19 [1.82, 11.43]). Willingness to take oral PrEP was negatively associated with sexual orientation concealment (aOR = 0.83 [0.70, 0.96]), having tested for HIV in the past six months, (aOR = 0.50 [0.34, 0.74]), and WeChat use for HIV prevention, (aOR = 0.84 [0.72, 0.98]).

PrEP adherence self-efficacy.

Analysis on participants’ PrEP adherence self-efficacy was only conducted among PrEP-willing MSM (N = 402). Participants’ average total score on PrEP adherence self-efficacy was 2.96 (SD = 2.54, range = [0, 6], skewness = 0.02, kurtosis = −1.72). Figure 1 presents the distribution of participants’ answers to PrEP adherence self-efficacy by item. As shown in Table 2, at a multivariate level, PrEP adherence self-efficacy was positively associated with PrEP knowledge (β = 0.32) and negatively associated with concerns regarding the efficacy of PrEP in preventing the transmission of HIV virus (β = −0.10). The model explained 12.3% variance in PrEP adherence self-efficacy.

Figure 1.

Adherence self-efficacy and potential impact of PrEP on sexual risk behaviors and HIV testing for oral PrEP uptake among PrEP-willing MSM (N = 402)

Potential impact of PrEP on sexual risk and HIV testing among MSM who were willing to take PrEP (N = 402) is illustrated in Figure 1. If using PrEP to prevent HIV, 41.5% of participants anticipated an increase in their condom use, with 32.8% and 25.6% saying they would use condoms about the same or decrease their use, respectively. Interestingly, nearly 39.3% of participants reported they would increase their number of sexual partners while taking PrEP. Finally, 38.1% of participants reported they would increase their HIV testing on PrEP while 46.5% anticipated undergoing HIV testing at about the same frequency as before.

Discussion

This study is an initial effort to advance knowledge on PrEP willingness and adherence self-efficacy among Chinese MSM with recent CAS and may benefit from PrEP. Overall, findings suggest that over half (56.4%) of this subpopulation of MSM were aware of PrEP and two-thirds (64.6%) were willing to use PrEP, yet adherence self-efficacy was only moderate among those who were willing to use PrEP, which may be a potential barrier for future PrEP implementation in China. To meaningfully engage Chinese MSM in PrEP uptake, it may be helpful to conceptualize a two-step approach to (i) promote PrEP willingness among potential users and (ii) enhance PrEP adherence self-efficacy among PrEP-willing MSM. Potential impact on sexual risk and HIV testing behaviors following PrEP implementation are also highlighted.

Findings from this analysis are consistent with “PrEP cascade” frameworks that posit a series of interconnected cognitive-behavioral stages needed to optimize PrEP outcomes at the population level [31–33]. For example, Parsons et al. (2017) applied principles from the Transtheoretical Model [34] to describe a progression of PrEP behaviors ranging from pre-contemplation to preparation to maintenance. A recent cross-sectional research applied the PrEP cascade framework among Chinese MSM found the majority of sexually active MSM in early stages of the cascade (e.g., precontemplation, contemplation, preparation) as well as the need to address barriers in these stages for PrEP uptake promotion [22]. Similarly, our findings suggest that varied public health intervention strategies are needed to support PrEP use among MSM, depending on the discrete objectives to either promote willingness to initiate PrEP use or to motivate efficacy and engagement in PrEP use.

Various demographic, psychosocial and behavioral correlates of PrEP willingness were identified. Consistent with prior research [13], migrant Chinese MSM reported higher PrEP willingness compared with non-migrant peers. Migrant Chinese MSM are also at a higher risk for HIV as well as other STIs (e.g., syphilis) [35]. Migrant Chinese MSM often lack the support of a stable social network and are disadvantaged by generally lower education levels. Moreover, due to China’s household registration system (“hukou”), migrants are marginalized in the urban health care system and often experience limited access to comprehensive, accessible, and long-term health services [36]. Thus, targeted PrEP interventions with this population will need to consider effective strategies to reach migrant MSM and address potential challenges they face, such as PrEP access, cost, lack of support, and poor mental health [37, 38].

Findings point to stigma as a barrier to PrEP willingness among MSM in China, and in particular highlight the roles of sexual orientation concealment and concerns about being treated as an HIV/AIDS patient [39]. Chinese MSM largely conceal their sexual identity to conform with the heteronormative culture and protect themselves from discrimination [38, 40]. Consequently, PrEP implementation programs will need to consider the fear of identity exposure that many Chinese MSM may experience with PrEP uptake. In addition, societal stigma against people living with HIV is prevalent in China [41, 42]. Thus, MSM may be unwilling to adopt PrEP as a prevention method to avoid being viewed as sick or living with HIV. Some previous research found recent sexual risk behaviors associated with higher PrEP willingness [13, 16, 17] while a study found behavioral indication for PrEP not associated with self-perceived appropriateness for PrEP candidacy [23]. Findings of this study provide a more nuanced perspective. Specifically, recent behaviors including sex under the influence of substance and having an HIV+ sex partner were associated with higher PrEP willingness, yet receptive CAS was not associated with PrEP willingness, indicating a need to further understand circumstances and risk perception related to receptive CAS for individuals as well as to educate MSM on specific behavioral indications for PrEP [43].

Contrary to previous research [15], we found HIV testing in the past six months was associated with lower PrEP willingness. There are two possible explanations for this association, which warrant further consideration. Potentially, MSM who had recent CAS and engaged in recent testing were also more likely to be regular testers and, consequently, they may feel less anxious about their HIV status due to their regular testing regimen [44]. Alternatively, MSM who recently received an HIV-negative test result might perceive themselves as having low vulnerability to HIV infection and, consequently, consider PrEP less relevant to their needs. We also found that MSM who had higher use of WeChat for seeking HIV prevention information or services were less willing to use PrEP, potentially due to lower perceived need for additional preventive methods. Future PrEP promotion interventions should consider targeting MSM at HIV testing sites and via WeChat-based services in order to effectively inform MSM the benefits of PrEP (even for those who recently tested HIV-negative) and the role of PrEP on HIV prevention continuum. Internet-based platforms are particularly useful for reaching Chinese MSM due to their use of mobile apps for seeking partners, social support, and MSM-related information [45–47].

Among those who expressed PrEP willingness, adherence self-efficacy was associated with cognitive variables such as PrEP knowledge and having concerns about HIV prevention efficacy (i.e., perceiving PrEP as ineffective). As a note of caution, adherence self-efficacy was assessed among PrEP-naïve MSM, and anticipated adherence self-efficacy in various circumstances may differ from actual adherence following prescription of PrEP. Nonetheless, these findings suggest the need for adherence interventions and ongoing support for MSM in China who initiate PrEP. Intervention strategies may include counseling and motivational messaging (in person or via mobile app) to support PrEP adherence, personalized feedback on PrEP adherence levels (e.g., assessed via biospecimens), and messaging to facilitate “good enough adherence” (e.g., four doses per week [48]) in the context of risk and other preventive behaviors (e.g., condom use) [49]. Although the Chinese government has only recently approved daily PrEP, recent research on the protective levels of intermittent PrEP dosing may prompt consideration about national endorsement of non-daily PrEP use (e.g., 2–1-1) [50].

Findings on the potential impact of PrEP uptake on risk and testing behaviors indicate a need for monitoring behavioral risks among PrEP users. One quarter (25.6%) of participants who reported PrEP willingness reported a potential decrease in condom use, and 39.3% reported that they would likely have more sex partners following PrEP initiation. PrEP trials and future implementation in China may also need to monitor risk compensation and engage MSM in ongoing efforts to reduce activities that can further the spread of other STIs among PrEP users and their partners [51–53]. Notably, 38.0% reported that they would likely increase their frequency of HIV testing following initiation. As regular HIV testing is recommended as part of PrEP implementation guidelines in U.S. and elsewhere [19, 20], this suggests potentially an accepting attitude towards HIV testing and positive effect of PrEP in engaging MSM to HIV prevention services. Given that the prevalence of regular HIV testing is overall unsatisfactory (e.g., 48.4% in a study of sexually active, largely educated and urban Chinese MSM [44]), PrEP could offer an opportunity to enhance HIV testing efforts among Chinese MSM. In this regard, coupling with convenient, innovative, and technology-assisted testing strategies (e.g., at-home HIV self-testing with app-based follow-up services [46]) could greatly reduce the burden of testing and encourage uptake.

PrEP implementation trials in the U.S. and elsewhere have highlighted the need for continued risk monitoring [52, 54] as well as PrEP use in the context of promoting sexual well-being and health services [55, 56]. Overall, findings of the current study suggest the potential promise of a PrEP promotion strategy that is community-based, culturally-competent, and centered around providing health benefits that go beyond pure risk-reduction among Chinese MSM. Specifically, benefits of PrEP might include individual-level (e.g., improved sexual well-being, reduced worries about HIV), interpersonal-level (e.g., greater sexual intimacy and communication), and community-level (e.g., reduced HIV stigma and MSM stigma) outcomes [56–58]. Given that psychosocial factors including HIV stigma and identity concealment emerged as important determinants of PrEP willingness, PrEP awareness campaigns focused on Chinese MSM should employ stigma reduction focused strategies in addition to providing PrEP knowledge. MSM-focused stigma reduction interventions are lacking in general in China [38], albeit recently growing efforts [59]. It is equally important to train PrEP providers competent in LGBTQ care and provide equal access to PrEP and other health benefits for subpopulations of MSM who are historically marginalized (e.g., migrant MSM). Recently, a study with lay HIV workers in China found significant PrEP stigma decrease via a training intervention on PrEP implementation, suggesting potential promise of engaging providers in addressing PrEP stigma and enhancing uptake [60]. Working with community stakeholders could also be effective to influence peer norms regarding PrEP usage, particularly since opinions of the LGBTQ community was the most influential factor in PrEP adherence self-efficacy (see Figure 1). Given the self-reported anticipation in reduced condom use and increased sex partners among some MSM, PrEP implementation programs in China should consider branding PrEP to MSM community as an opportunity for sex education, overall sexual health promotion, and linkage to community-based services. Sex-positive messages may be particularly important to combat the strong sociocultural stigma against MSM in China [38, 40, 61]. This may facilitate continued risk monitoring (e.g., HIV and other STI testing) in the context of sexual health empowerment rather than being managed as a “high risk” patient, which could elicit stigma and deter MSM from continued PrEP engagement.

Several study limitations need to be considered. First, given the focus on MSM with recent CAS in urban China, results may not be generalizable to the general MSM population in China, especially those in more rural areas. In a related vein, since participants were recruited through a mix of online and in-person efforts, we do not know how the proportion of MSM who may reviewed the study information and uninterested in participating, resulting in a lack of knowledge regarding generalizability to the larger MSM community in urban China. Second, the cross-sectional nature of our data prohibits casual inferences. Third, we provided a conservative description of PrEP effectiveness as 90% [62], yet evidence suggests PrEP is up to 99% effective in preventing HIV acquisition through sex with a daily intake [48]. Thus, this language was potentially confusing for participants [62] and may lead to an underestimation of PrEP willingness in our sample. Fourth, questions regarding adherence self-efficacy and potential impact of PrEP uptake on sexual risk and HIV testing behaviors are hypothetical in nature and so our results need to be confirmed by future clinical trial outcomes. Fifth, data for this analysis were collected in the context of a parent study on HIV self-testing, and therefore participants might have been particularly sensitized to HIV risk reduction at the time of data collection.

Conclusions

In conclusion, this study highlights the relevance of stigma, risk and HIV prevention behaviors in PrEP willingness, the important role of cognitive factors in adherence self-efficacy, and the potential impact of PrEP on risk and testing behaviors among MSM with recent CAS in urban China. To meaningfully promote PrEP acceptance among in this population, intervention efforts should incorporate strategies to reduce HIV and PrEP stigma, attend to the needs of vulnerable subpopulations (e.g., migrants, MSM who use substance), and incorporate PrEP with existing preventive behaviors as part of a comprehensive HIV prevention strategy among Chinese MSM. Enhancing PrEP knowledge, including information about its efficacy and the importance of adherence, and continued monitoring of sexual risk and engagement in HIV prevention programs and services, may help achieve optimal outcomes for PrEP users and their partners. Given its proven effectiveness and safety, successful implementation of PrEP has the potential of impacting the HIV epidemic curve in the region.