“Talking about it publicly made me feel both curious and embarrassed”: Acceptability, feasibility and appropriateness of a stigma-mitigation training to increase health worker comfort discussing anal sexuality in HIV services

Abstract

Health workers report challenges to broaching sexual behavior with gay and bisexual men (MSM). We conducted a stigma-mitigation training to increase provider-initiated conversation about anal sexuality among Chinese health workers. The two-day workshop, titled Smarter Sex is the New Safer Sex: Anal Pleasure and Health, coupled information about anal physiology and sexual response with gradual exposure to trainees’ emotional and cognitive responses. We analyzed surveys and one-time interviews. Acceptability (satisfaction with training activities and recommended practices), feasibility (compatibility with participants’ current practices), and appropriateness (perceived fit with participants’ work mission and goals) were high, reaching 84-95% of the scale range, though with variable ratings for trainee comfort during participation. Qualitative data confirm health workers’ motivation to discuss anal sexuality and the value of learning about anal physiology and sexual response, but revealed continued uncertainty about how to broach and navigate discussion with clients. Refinement of the training to align with health worker recommendations and to augment communication options and skills is warranted, as is involvement of MSM clientele themselves.

Introduction

Stigma within health services against gay, bisexual and other men who have sex with men (MSM) limits their access and engagement in HIV programming [1-7]. For MSM, stigma specifically inhibits willingness to disclose sexual behavior [8,9] and therefore the potential benefit of targeted delivery of HIV interventions [5,7,10-15]. For healthcare providers (HCPs), personal discomfort conducting sexual histories [16] and a lack of education about MSM-relevant health issues impedes communication and assessment [17], thereby limiting their ability to help MSM curb their HIV risk [5,10,11]. Training interventions have traditionally served an important role in the global fight against HIV/AIDS [7,12,13] to develop knowledge and skills and, more recently, to build competence working with MSM [1,3]. Although evidence in support of competency training is limited, systematic reviews encourage continued efforts to target stigma as a pervasive contributor to health disparities [14,15,18-20].

Few peer-reviewed evaluations of trainings focus explicitly on stigma toward sexual behavior, despite the fact that sexual stigma bears heightened relevance for MSM and their interactions in HIV services [17,21]. Anal sex is the most proximate risk factor for HIV among MSM [22]. However, the discussion of sexual behavior appears to be exceedingly rare in clinical encounters, by some estimates occurring in less than 50% of meetings even in HIV care settings. Such discussion is limited by both HCP motivation and skill as well as reticence among MSM to reveal sexual orientation and specific anal sex practices [21-26]. To this point, experts in sexual health strongly recommend that competencies in medical education include the ability to discuss and assess specific sexual behavior [27]. However, there is no empirically-supported guidance to improve HCP knowledge, comfort and intervention skills related to anal health and sexuality. One peer-reviewed evaluation of competence training has included knowledge related to anal sexuality, albeit focused on risk reduction [21]. However, MSM prefer holistic interventions that address sexual health and pleasure more broadly, rather than a primary focus on infectious disease [28,29].

Scientific knowledge about competence training to help HCPs work with MSM has also been limited by geography. Peer-reviewed publications have evaluated trainings conducted in North America [30], Europe [31], and sub-Saharan Africa [32-35], but rarely in Asia. In China, where same-sex behavior is characterized by the state as abnormal, pornographic, and sanctionable [36], HIV incidence among MSM is 5.5 per 100 person-years, almost ten times the rate among MSM in the United States [37,38]. Chinese MSM exhibit many of the same HIV risk behaviors associated with anal sex that researchers have identified in MSM populations in other countries [39,40] as well as similar patterns of stigma and discrimination that inhibit disclosure of sexual behavior – as low as 16% in one study [41] – and deter engagement in health services [2,41-44]. The role of Chinese healthcare providers in the MSM epidemic is central to the future of HIV prevention and control. Biomedical innovations like pre-exposure prophylaxis (PrEP) are on the cusp of regulatory approval in China and successful engagement and screening of MSM clients for new HIV prevention options will require that HCPs communicate skillfully with MSM specifically about sexual behavior [45].

In an effort to understand how to mitigate stigma toward anal sexuality among HCPs, we piloted a training intervention to develop knowledge, comfort, and skill discussing anal sexuality among HIV workers in China, where stigma toward MSM is high and cultural competence training is rare but potentially of interest to HCPs [46]. We evaluated acceptability, feasibility, and appropriateness of the training and report on the practices recommended over the course of the training.

Materials and Methods

To assess acceptability, feasibility, and appropriateness of the intervention, we conducted: A) an online survey immediately after the training; B) in-depth telephone interviews approximately one to two weeks post-training; and C) a follow-up online survey approximately two months post-training with additional open-ended qualitative questions. The Institutional Review Board of the New York State Psychiatric Institute approved all procedures, as did Fudan University in China through an amendment to an open protocol from an ongoing two-year collaboration. Informed consent was obtained from all individual participants included in the study.

Participants

We targeted recruitment through contacts from a Good Participatory Practices (GPP) collaboration with health workers and policymakers across six Chinese cities, including the Centers for Disease Control and regional hospitals, and community-based organizations (CBOs) dedicated to serving sexual minorities. Potential participants received an email invitation to attend a sexual health workshop and, if interested, a follow-up hyperlink to an information statement to consent for study participation. Potential participants could decline to attend the training or could participate in the training and decline to complete data collection without repercussion for their involvement in the larger GPP collaboration.

Procedures

Surveys, interview guides and training materials were originally developed in English and reviewed iteratively to ensure fit to the sociocultural context. After translation into Chinese, bilingual research staff back-translated into English, then reviewed the translation and conducted pilot testing to ensure quality. Surveys were hosted through a China-based survey platform. An anonymous unique identifier linked participants across timepoints.

Interviews lasted about one hour and were conducted in Chinese in September 2018 by research assistants trained in qualitative inquiry. Participants who completed in-depth interviews were reimbursed RMB200 (~US$30) for their time and those who completed all assessments could enter a lottery to receive one gift certificate worth RMB500 (~US$80). Interviews were audio-recorded on a password-protected device and downloaded to a secure server, after which the original file was deleted. Research assistants also completed a guided debriefing report after each interview [47]. A third-party vendor transcribed interviews, which the study team then reviewed for accuracy and to redact any identifying information. The team reviewed transcripts prior to completion of all interviews, to further guide qualitative inquiry.

Training Intervention

The two-day training, Smarter Sex is the New Safer Sex: Anal Pleasure and Health 1.0, was developed by the first author (BAK) and originally delivered through a training institute for CBO staff working on HIV prevention among sexual minorities in the United States [48]. It was not evaluated for research purposes in this context. In China, we piloted the intervention in August 2018 with the specific goal of studying its implementation. Exercises focused on building HIV workers’ knowledge and communication skills to promote disclosure of anal sexuality among clients who may otherwise remain reluctant to share their sexual concerns, including their HIV-related risk. The training relied on the thoughts-feelings-behavior model from cognitive behavior therapy (CBT) [49,50] to lessen health workers’ anxiety about discussing anal sexuality. Activities targeted health worker mindfulness and acceptance of their own emotional and cognitive responses to discussing anal sexuality, and simple information exchange skills based in Motivational Interviewing (i.e., elicit-provide-elicit) [51] to build confidence evoking and responding to disclosure among their clients. This process included exposure-based procedures and cognitive restructuring exercises (e.g., values clarification, loss- vs. gain-framed messaging) as well as basic information about anal physiology and sexual response. Training activities reinforced learning in several ways (e.g., individual reading, small group discussion, review), coupling gradual exposure to discussing anal sexuality with skills-based practice in client-centered communication (e.g., teach back, question carousel, roleplays). Trainers modeled skills in response to participant questions.

Measures

Quantitative Measures

Shortly after the training, we measured its implementation with two sets of scales. The 13-item Acceptability, Feasibility, Appropriateness Scale (AFAS; α = .92) comprised three subscales [52-54]. Acceptability referred to satisfaction with the training activities and recommended practices (α = .84; e.g., “How comfortable are you with the exercises contained within the training?”). Feasibility referred to compatibility with participants’ current practices (α = .78; e.g., “To what extent do you expect to be able to incorporate the concepts and techniques from the training into your daily work activities?”). Appropriateness referred to perceived fit with participants’ work mission and goals (α = .88; e.g., “How relevant are the information and practices to your client population?”). Response options ranged from 1 (Not at all) to 5 (Extremely).

We additionally relied on an implementation science framework, the Theoretical Domains Framework (TDF) [55], to anticipate domains that could influence implementation (e.g., knowledge, skills, social/professional role and identity, social influences), and to characterize adaptations to the training content or processes that trainees might recommend. The TDF has been used to explain why recommended interventions have not been translated into best practices among HCPs and to develop further theories to explain behavior change [56]. We adapted the Determinants of Implementation Behavior Questionnaire (DIBQ) [55,57] to measure elements of the TDF. The DIBQ is a validated scale which comprises a set of adaptable items to measure TDF-related factors that influence healthcare providers’ implementation of recommended practices in real-world settings [57]. Our 25-item scale (α = .82) comprised 10 domains: perceived behavioral control (α = .85), optimism (α = .67), attitude (α = .93), outcome expectancy (α = .58), reinforcement (α = .83), intentions (α = .68), goals/priority (α = .90), innovation characteristics (α = .47), innovation strategies (α = .83), and 1 item for organization. Response categories ranged from 1 to 7 with variably-defined anchors (e.g., Very difficult to Very easy; Not worthwhile at all to Very worthwhile).

Qualitative Inquiry

In-depth interview guides aimed to elicit reflections on training implementation based in the TDF [56,57], including aspects of acceptability, feasibility, and appropriateness (along with potential adaptations for each), and emotional and social influences. Our follow-up open-ended online survey included prompts for additional reflections (e.g., “How have you tailored, or how might you tailor, what you learned in the training to your MSM clients’ needs?”).

Analyses

For quantitative analyses, we examined means and measures of central tendency. For qualitative analyses, three co-authors began by reviewing 9 of 30 transcripts (29%) and all 49 two-month follow-up survey responses. During weekly meetings, this workgroup presented and discussed individually-developed diagrams of each transcripts’ salient content [58]. Meetings periodically included discussion of project-level diagrams, compiled from transcript-level diagrams, to guide exploration of themes for further analysis [59]. We then refined a codebook based on these themes and our a priori qualitative inquiry prompts, double-coded 10 percent of transcripts, and resolved discrepancies by consensus. Using this final codebook, we double-coded all transcripts, analyzing in memos how the content contributed to or contradicted already-developed themes or presented new insights. We then reviewed coding reports for acceptability, feasibility, and appropriateness to develop new diagrams. We interrogated our analysis in relation to existing literature and gradually came to consensus of the major themes and findings.

Results

Sample

Of the 51 people who registered for the training, all consented to participate in the research. Forty-one (80.4%) attended both days of training, with the remaining nine attending 50-75% and one person attending only 25% (a half-day). Forty (78.4%) completed survey assessments of implementation immediately after the training. We invited all training attendees to be interviewed and interviewed all 30 (59%) who expressed interest, most of whom (90%) completed both training days. Two months later, 49 (96.1%) completed our follow-up open-ended survey questions.

Participants (see Table I) skewed toward younger age (M = 34.5, SD = 8.6), with most (86.3%) identifying as cisgender male. Most resided in cities located in the North (23.5%), Central (21.6%), and Southwest (19.6%) of China. Education level was generally high, with half reporting a college degree and a little more than one-third (37.3%) reporting more than a college degree. About half identified as gay/homosexual and another sizable minority (13.7%) as bisexual. With regard to work responsibilities, just over half of participants conducted HIV counseling and testing, about 20% provided primarily HIV/STI treatment (as medical doctors), and another 20% provided general public health interventions. Modal career length was 5-10 years. MSM clients comprised about 70% of their service delivery population.

Table I.

Demographic characteristics (N = 51)

Variable	n	(%)	Mean	(SD)
Age (in years)			34.5	(8.6)
City of Residence
North	12	(23.5)
Central	11	(21.6)
South	5	(9.8)
Southwest	10	(19.6)
East	4	(7.8)
Other	9	(17.6)
Education
Middle school	1	(2.0)
High school or equivalent	6	(11.8)
College	25	(49.0)
Above college	19	(37.3)
Ethnicity
Han ethnicity	46	(90.2)
Additional ethnicity	5	(9.8)
Gender Identity
Cisgender male	44	(86.3)
Cisgender female	7	(13.7)
Sexual orientation
Gay/homosexual	24	(47.1)
Straight/heterosexual	20	(39.2)
Bisexual	7	(13.7)
Client population (% of all clientele)	69.9	(31.0)
Gay/bisexual men	15.2	(19.7)
Heterosexual men	8.8	(13.4)
Heterosexual women	2.2	(6.4)
Lesbian/bisexual women	2.0	(3.8)
Transgender men or women
Primary work duties
HIV counseling & testing	26	(51.0)
Public health interventions	10	(19.6)
HIV treatment*	7	(13.7)
STI treatment*	4	(7.8)
Other	4	(7.8)
Work setting
Community based organization	27	(52.9)
Hospital	16	(31.4)
Chinese Centers for Disease Control	14	(27.5)
STI Clinic	8	(15.7)
Additional	3	(5.9)
Employment
Full-time	43	(84.3)
Part-time	6	(11.8)
Other	2	(3.9)
Monthly income
No income	3	(5.9)
2000 RMB – 3999 RMB	6	(11.8)
3000 RMB – 4999 RMB	16	(31.4
5000 RMB – 9999 RMB	19	(37.3)
10000 RMB – 19999 RMB	5	(9.8)
20000 RMB – 29999 RMB	2	(3.9)
Career length
Less than 1 year	8	(15.7)
1-2 years	5	(9.8)
2-5 years	11	(21.6)
5-10 years	15	(29.4)
10-20 years	9	(17.6)
20 or more years	3	(5.9)

Quantitative Results

Acceptability, Feasibility and Appropriateness

As seen in Table II, ratings of training acceptability (M = 4.73, SD = .49), feasibility (M = 4.33, SD = .71), and appropriateness (M = 4.19, SD = .75) were all high on the 5-point scale range. Responses varied most for items related to compatibility of the training’s recommended practices with workflow and job responsibilities. The lowest endorsement was for workflow timing (M = 3.93, SD = .76), but across these items (e.g., compatibility with daily work activities, mission, and service provision mandate, and work approach), variability was similarly dispersed (SD range: .72 - .82). Respondents rated items more highly for the usefulness and relevance of the training’s information and practices for their clients (respectively M = 4.60, SD = .67; M = 4.53, SD = .68).

Table II.

Acceptability, feasibility, appropriateness of stigma-mitigation training (n = 40)

	Mean	(SD)
Acceptability, Feasibility, Appropriateness Scale (AFAS)	4.43	(.64)
Acceptability	4.73	(.49)
To what extent are you satisfied with the training you received?	4.75	(.49)
How credible did you find the trainers?	4.83	(.38)
How well organized and executed do you believe the training program to be?	4.73	(.45)
How satisfied are you with the content of the training covered?	4.65	(.53)
How comfortable are you with the exercises contained within the training?	4.70	(.56)
Feasibility	4.33	(.71)
How useful are the information and practices from the training to you in your everyday practice?	4.60	(.67)
To what extent do you expect to be able to incorporate the concepts and techniques from the training into your daily work activities?	4.23	(.80)
How compatible do you expect the practices from the training to be with the practical realities and resources of your work setting?	4.15	(.66)
Appropriateness of Information and practices	4.19	(.75)
How compatible are the information and practices with your work mission or service provision mandate?	4.00	(.82)
How relevant are the information and practices to your client population?	4.53	(.68)
How well do the information and practices fit with your current approach to work?	4.30	(.72)
How compatible are the information and practices with your workflow timing (e.g., when and for how long you see clients)?	3.93	(.76)
How well do the information and practices from the training fit with your overall approach to service delivery in the setting in which you provide care?	4.18	(.75)

Likert responses ranged from 1-5, with higher scores indicating greater positive valence (1 Not at all, 2 Slightly, 3 Moderately, 4 Very, 5 Extremely)

Theoretical Domains Framework Constructs

TDF domains (see Table III) reflected generally high endorsement of the training itself and its recommended practices, with notable variability. For the training itself, participants rated seven of the domains between 81% - 94% of the 7-point scale range. Highly-rated domains included attitudes toward the training (i.e., useful, worthwhile, and interesting; M = 6.59, SD = .68); optimism about the benefits to future participants (M = 6.46, SD = .73); innovation strategies (i.e., sufficiency of materials and skills to discuss anal health with MSM; M = 6.14, SD = 1.06); and anticipated reinforcement for their participation in the training from clients and colleagues (M = 6.13, SD = .90). Lower, but still highly rated, were perceived behavioral control (i.e., contributing, participating, and engaging in the training; M = 5.69, SD = 1.30) and outcome expectancy (e.g., anticipated advantages of participation for future participants; M = 5.68, SD = 1.10). Goals/priority (e.g., relative urgency to attend this training compared to another) was rated lowest (M = 5.00, SD = 1.37).

Table III.

Determinants of Implementation Behavior Questionnaire for training implementation (n = 40)

Variable	Mean	(SD)
Perceived Behavioral Control*	5.69	(1.30)
Contributing during the training	5.70	(1.29)
Participating in activities during the training	5.55	(1.36)
Engaging in the training	5.83	(1.26)
Optimism**	6.46	(.73)
If the training were offered to others, I expect more good things to happen than bad.	6.43	(.81)
I’m optimistic about what will result from the training.	6.50	(.64)
Attitude***	6.59	(.68)
Usefulness of the training	6.58	(.71)
Worthwhile	6.65	(.62)
Interesting	6.55	(.71)
Outcome Expectancy**	5.68	(1.10)
Other people will appreciate the training	6.10	(1.01)
Other people will feel satisfied by the training	5.90	(.90)
Other people will feel comfortable with the subject matter	5.23	(1.12)
The training will benefit public health	6.35	(.77)
The training will have disadvantages for other participants [reverse-scored]	4.83	(1.68)
Reinforcement**	6.13	(.90)
Future participants would get positive recognition from clients for participating in this training	6.15	(.89)
Future participants would get positive recognition from colleagues for participating in this training	6.10	(.90)
Intentions	6.37	(.70)
I will definitely recommend this training to colleagues in the next few weeks.**	6.28	(.88)
Strength of intention [n = 37; skipped if previous response was undecided)	6.27	(.80)
Goals/Priority*****	5.00	(1.37)
How often do you think attending other trainings would be a higher priority than attending this training?	5.10	(1.43)
How often do you think attending other trainings would be more urgent than attending this training?	4.90	(1.32)
Innovation Characteristics**	5.63	(1.24)
What we learned is compatible with my daily work practices	5.83	(1.20)
What we learned is simple to deliver	6.15	(.89)
What we learned costs little time to deliver	4.93	(1.62)
Organization**
Within my workplace, there is simply not enough time to implement the recommended practices from the training.	3.08	(1.56)
Innovation Strategies**	6.14	(1.06)
The training provides sufficient materials to discuss anal health with MSM clients.	6.13	(1.14)
The training provides sufficient skills training to discuss anal health with MSM clients.	6.15	(.98)

All item Likert responses ranged from 1-7, with higher scores indicating greater positive valence

^*1 (Very difficult) to 7 (Very easy)

^**1 (Strongly disagree) to 7 (Strongly agree)

^***1 (Not at all) to 7 (Very)

^****1 (Very strong) to 7 (Very weak)

^*****1 (Very often) to 7 (Very rarely)

Two remaining domains assessed the training’s recommended practices. Innovation characteristics (i.e., compatibility, simplicity, and time to deliver the recommended practices; M = 5.63, SD = 1.24) garnered relatively high endorsement. Ninety-three percent “definitely” intended to recommend the training to others (the remaining 7% were undecided). Respondents generally rated the training’s practices to discuss anal health as simple to deliver (M = 6.15, SD = .89), but anticipated less compatibility with daily work practices and costs on their time (respectively, M = 5.83, SD = 1.20; M = 4.93, SD = 1.62). The single item for organization, also related to time available to implement recommended practices, garnered the scale’s lowest endorsement, reaching only 44% of the scale range, with the highest variability (M = 3.08, SD = 1.56).

Qualitative Inquiry

Table IV includes themes and illustrative quotations for acceptability, feasibility, and appropriateness, which we summarize below alongside additional qualitative evidence not quoted within the table.

Table IV.

Implementation themes from analyses of in-depth interviews (n = 30) and follow-up text entry responses (n = 49)

Domain	Theme	Illustrative Quotations
1. Acceptability	1.1 Participants perceived the content on anal health to be novel and professionally communicated	I think the knowledge [on physiology] was very comprehensive and in-depth. … It was brand new for me. Although I probably knew these things in the past, I didn’t know their exact structure or how they work. / 26 yo, Gay, VCT, CBO/CDC
		The approach and attitude this training adopted is all about sharing knowledge, instead of being shy about the topic or trying to hide. It made people feel that it was about knowledge. It made people feel comfortable and people benefitted a lot from it. And it was very eye-opening. / 31 yo, Bisexual, Other duties, CBO
		I think it really makes up a big blank area in China, I mean the topic of anal health. We didn’t really discuss it [before]. Or even if we did want to discuss it, we had no access to professional knowledge. / 36 yo, Gay, VCT, CBO
	1.2 Participants perceived the content on sexual pleasure as culturally relevant and necessary	The trainer talked about how to obtain pleasure from anal sex. I think this is a really good point as it comes from the needs of the community. So, the purpose of the training was to benefit community members. It's not only about “You need to be healthy, you should to this or that.” In this way, community members would really be willing to participate in the training. / 32 yo, Gay, Public health, CBO
		If you don't understand the pleasure [of anal sex], you won't be able to understand why people have anal sex. / 39 yo, Heterosexual, STI, Hospital
		I think regarding knowledge shared in the training, we always tell people how to protect themselves, but we don't know how to talk about pleasure or improving one's sex life. As far as I know, no one has done that. I understand that we are always able to tell people what they shouldn't do, and we’re not able to tell them what to do. / 33 yo, Gay, VCT, CBO
	1.3 Gradual exposure as a training approach engaged participants despite their initial reticence	It was the first time for me to participate in this kind of training, so, I was not quite at ease in the beginning, but I was able to engage. … I didn't feel it later. Basically, I felt better after the first morning and I didn’t feel ill at ease afterwards. / 26 yo, Bisexual, VCT, CBO
		Bringing up this topic and talking about it publicly made me feel both curious and embarrassed. But I think the discussion was quite open and I learned a lot. This part was really impressive. / 51 yo, Bisexual, VCT, CBO
		Even if someone was not very interested in the topic, being in the [training] environment where everyone is very active, this will affect them and make them more interested in this training … The trainer … was intentionally trying to make us not embarrassed. That is to say, he asked us to share things anonymously, asked us to exchange thoughts during small group sessions. He didn't ask us to reveal our private things. So I think there was nothing embarrassing. / 30 yo, Heterosexual, VCT, CDC
	1.4 Mixing different professional backgrounds facilitated understanding but also elevated social discomfort, serving gradual exposure for some and interfering with acceptability for others	The doctors were quite good. It's probably because they were interested in learning about this topic; they were quite good and participated in the training. This is different from CDC staff or doctors whom I have met in previous trainings. Or it's possible that it was due to the form of this training. Maybe they were a bit serious and distant in the beginning, but as we played the games and got desensitized, they felt that we could all participate in the training together. … In terms of my previous trainings with doctors, I sometimes felt awkward and could not talk freely. / 26 yo, Gay, VCT, CBO/CDC
		There was a session where we had a paper tag on the back and others came to describe it. And we were like, I mean at least for myself, we went to look for people who [also] work in CBOs or people we don't know. For officials from the CDC, I was quite embarrassed to go to them and to describe the phrase on their back. Because some of the phrases were indeed … I mean if you say it between CBO partners there is no problem at all. But it would be a little bit embarrassing if we describe it to the CDC officials. So we just avoided them intentionally. / 35 yo, Gay, VCT, CBO
		Everyone considered themselves as a student. This was very clear and also it is always hard to achieve this. For example, many people may think they are experts or very experienced, or they've studied this field. If people like this come to the training, they’d always feel they're professional and think about their status. … It didn’t happen in this training. … We were all equal and respectful to each other. / 33 yo, Gay, VCT, CBO
2. Feasibility	2.1 Information and information exchange practices are easy to use and compatible with existing approaches, but not sufficient for all client scenarios	There was not only theoretic knowledge, but also suggestions on how to relax, and different sex positions. The training also included knowledge about protection and lubrication, etc. I think this is all stuff that I might use in real life in the future. / 35 yo, Gay, VCT, Hospital/CBO/CDC
		You can only talk about this with people who are willing to talk with you. … There are people who don’t want to talk about anything with you. They just want to have the HIV testing done and leave immediately. The methods you mentioned would be limited in this scenario. Because you encounter so many different types of people when you are doing testing and counseling, it's impossible to solve all questions using one approach. … I think it can only be applied to a small portion of circumstances. / 37 yo, Gay, Public health, CBO
		I think the [elicit-provide-elicit] counseling method is great. It’s close to the audience, not complicated. You'll be able to understand it with a glance. … This is different from forcing participants with what I know. Instead, this approach is more gradual. It slowly shares the information with you with your consent and by feeding your curiosity. … It’s like slow cooking or the principle of treatment in traditional Chinese medicine. / 31 yo, Bisexual, Other duties, CBO
		I think in terms of feasibility, there is no problem. It’s easy to use. The critical point is when to use it. You cannot just copy the training materials. That is to say, we can't just repeat what's said in the textbook. It's just a recommendation. Everyone has different counselling skills. The important thing is when during the counseling to apply which method. I don't think there will be any problem or barriers for me to apply it in my practice. / 46 yo, Heterosexual, STI, CDC/Other setting
	2.2 Recommended practices suit existing practices, but adaptation and dissemination is unreliable	I’ve been talking to people briefly since I came back from the training, about knowledge of anal sex, etc. But I don’t think I could do the counseling as thoroughly as the trainer. … I believe it is necessary to receive concrete training in a classroom, to feel, and to summarize and come up with your own model for counseling. However … even if I repeat exactly what I learned, we probably couldn't achieve what the trainer modeled. / 29 yo, Gay, Other duties, Other setting
		When I returned from the training, the next day, I trained other counselors and staff in our clinic. But during my training … I compressed the training into a two-hour session and just provided a brief training. During this process, I could feel that people there were embarrassed. … They have many years of working experience. I think this embarrassment came from the fact that I didn’t help them get desensitized enough nor did I conduct the ice-breaking activities well. It was probably also due to my training skills. But I still think they have the capability to accept this. But you just need to give them more time to accept it. / 33 yo, Heterosexual, Public health, Hospital
3. Appropriateness	3.1 Information learned will help engage MSM	We need to better understand our body not only to have better anal sex, but also to protect it. We need to first understand it before doing it. / 26 yo, Gay, VCT, CBO/CDC
		If the training is open to the general public, I think gay men in general would be interested as well. This is very relevant to them. I mean gay men who are not working in health-related field, who are not CBO volunteers, nor participated in HIV-related work. I am saying that, just as a regular gay guy, they would be interested. / 30 yo, Heterosexual, VCT, CDC
		Everyone could see that CBO people were very interested in the training. So I believe that my patients should be interested in this content as well. / 41 yo, Heterosexual, VCT, CDC
	3.2 Information was linked to health and prevention	I think when people know more about this, it would be helpful for our future work on prevention and intervention. This is beneficial. Because anal sex is not simply about sex, it requires skills and knowledge. The point is how to have anal sex in a safer and healthier way. / 37 yo, Gay, VCT, Hospital/CBO
		The trainer didn't need to explicitly talk about HIV and STI prevention…I mean he didn’t need to specifically explain how to prevent these diseases. Instead, he achieved this through talking about anal health…If he exclusively talked about the negative consequences of HIV or STIs, it might have had some negative impact. So what he did is not specifically talk about HIV or STIs, but by educating everyone [about anal health] … we learned how to prevent diseases. This also achieved the outcome of disease prevention. / 36 yo, Bisexual, VCT, Hospital
		Clients rarely come for testing and ask us how to obtain more pleasure. This is rare, not many people. … I think pleasure is a less important point during counseling [than HIV]. … [But] if you couldn't open the conversation or they wouldn't share with you some critical information, maybe you could try to talk about pleasure to overcome the barriers, so that it would be easier to continue your counseling [about safety from HIV]. / 23 yo, Gay, VCT, CBO
	3.3 How to share information with clients is nuanced	It may make them have some negative reactions, they may think that you are trying to pry into his personal life. By the way, during the training, you could consider including this scenario: you are very eager to share this knowledge with a client, but the environment at that time makes it difficult to do so. In this situation, you are having difficulties and you don’t know how to start the discussion on this topic. In the training, you can teach participants skills on how to guide and elicit – how to gradually guide the conversation to this topic, so that you could better communicate the knowledge. / 28 yo, Heterosexual, Public health, CDC
		If a client comes to me and directly asks, I wouldn't feel any pressure … But if a client didn’t ask about questions related to this topic, and I need to initiate discussion about anal health or anal pleasure … it feels a little bit like sexual harassment. … I think it’s really hard actually to distinguish. … It wouldn't be any problem to talk about it in a group setting, but if it's one-one-one … It could be difficult to find the right tone to talk about anal sex in a one-on-one setting. If you talk gently, they may think you’re flirting. If you try to distance yourself and use a serious tone – which is very hard, to talk in this tone – even if you did it, it’d sound strange. / 29 yo, Gay, Other duties, Other setting
		I think the topic of HIV/AIDS is more difficult than this. I'm able to talk about that with my clients, why wouldn't I be confident about discussing this? … I haven’t tried to discuss this topic in this way yet. I don’t know if people will feel embarrassed at that moment, or don't know what to say, or they'd think it's too much and you're harassing them. I don’t know what will happen. … I don’t know if people will think it’s weird if I bring up this topic, like “Why would you talk about it?’ / 39 yo, Heterosexual, STI treatment, Hospital

VCT: voluntary HIV counseling and testing; STI: sexually transmitted infection treatment; CBO: community-based organization; CDC: Chinese Centers for Disease Control and Prevention

Acceptability: Satisfaction with the training and its recommended practices

Participants perceived anal sexuality to be a novel and important training topic, sensitively and professionally conveyed by the training’s emphasis on knowledge acquisition (Theme 1.1). They unanimously appreciated the comprehensiveness of the information related to physiology. Doctors, who knew anatomy but were less familiar with its involvement in anal sex, and gay CBO staff, who were familiar with anal sex practices but knew far less about anatomy, both characterized explications of sexual response as particularly revelatory. Participants also described the focus on sexual pleasure as culturally-relevant, acceptable, and necessary (Theme 1.2). They remarked that speaking about pleasure rather than disease could be less alienating and more compelling to clients than focusing primarily on harms associated with anal sex.

Participants reported that the training’s gradual exposure to public discussion of anal sexuality through facilitated, stepwise activities motivated them to engage in conversation with fellow training participants, despite their initial reticence (Theme 1.3). This engagement surprised some trainees, who remarked that these same training activities often elicited aversive feelings, like embarrassment, that mostly subsided with exposure. A second training approach, the deliberate inclusion of different professional backgrounds, had mixed emotional results (Theme 1.4). Training alongside colleagues from different disciplines (i.e., doctors, CDC officials, and CBO staff) increased acceptability for some by facilitating understanding and achieving a rare sense of equality between CBO workers and medical professionals. However, for a few CBO staff in particular, the prospect of discussing anal sex across this professional hierarchy elevated their anxiety. Flooded with anticipated discomfort, they simply avoided communicating with doctors or CDC staff during specific exercises.

Feasibility: Compatibility with current practices

Participants generally considered the content on anal physiology and pleasure, risk reduction, and information exchange skills to be compatible and easy to integrate with their existing approaches to sexual health counseling (Theme 2.1). A few characterized the MI-based elicit-provide-elicit approach to information exchange as useful only when clients are willing to talk, or when time was not a constraint. They requested additional skills to broach discussion with particularly reluctant rather than forthright clients, and tools like brochures or videos to invite or hint that discussion would occur during one-on-one visits.

Interviewees expressed strong interest in adapting and then disseminating the training (Theme 2.2). Several shared content with friends and colleagues and a few attempted to replicate the training upon returning to their work settings, even though no technical assistance was provided for dissemination. Their experiences were not smooth. One compressed the two-day training into two hours, and hypothesized afterward that the compressed focus on informational content rather than a gradual exposure approach promoted embarrassment rather than comfort, even among colleagues whom she expected would embrace the material. In general, participants wondered how to broaden the reach of the training topic, including beyond the time constraints of one-on-one counseling or medical visits. Ideas included community forums, online venues, and informal social gatherings.

Appropriateness: Fit with participants’ work mission and goals

Participants predicted that the informational content on physiology and pleasure would interest MSM clients, and therefore considered the content aligned with their professional mission to promote MSM engagement in health services (Theme 3.1). Participants also described that a better understanding of anal sexuality among HCPs could promote safer and healthier sex among clients (Theme 3.2), a goal of interest to the population. They reported that many men in China suffer from hemorrhoids, almost endemically, and that MSM had likewise presented with other anal conditions, like painful fissures. They wished the live training had discussed these topics in more detail, rather than relegating this content to the manual. A few voiced concern that promoting pleasure and harm reduction (e.g., reducing pain, eliminating bleeding) might dislodge the importance of HIV prevention, specifically condom use, in the minds of some MSM. However, these same participants also imagined leveraging discussion of pleasure to overcome barriers to discussing HIV.

Some participants reported that discussing anal health would require greater nuance and skill than the training provided (Theme 3.3). They wondered, when clients themselves were not the initiators, whether sharing the information they learned might be misinterpreted as flirtation, sexual harassment or prying. One noted that HIV itself is a more daunting topic to discuss than anal sex, but still wondered whether discussing anal sexuality more broadly than its relation to HIV might be misconstrued as inappropriate by clients.

Discussion

In our sample of Chinese HCPs from six cities, participants deemed the stigma-mitigation training and its recommended practices to be highly acceptable, feasible, and appropriate interventions. They particularly valued the information about anal physiology and sexual response. Motivation to share the content was also high and participants rated the training as having relevance and benefits to public health. These high ratings were coupled with two notable caveats from the qualitative data. First, attending to the gradual exposure process and not just the content of the training is essential to implementation of the training, because simply sharing information with HCPs could unintentionally provoke discomfort. Second, HCP willingness to broach anal health and sexuality with clients, particularly during one-on-one interactions, is limited by time and work constraints but also HCP concern about instilling discomfort in clients who do not initiate discussion themselves. Adapting the intervention with these aspects of implementation in mind could improve the likelihood that trainees share the training’s information with colleagues and implement its recommended practice, to both respond to and initiate discussion.

More broadly, high ratings for acceptability, feasibility, and appropriateness were driven by four factors: A) high perceived need for the knowledge presented, B) professionalism of the training team, C) the gradual approach that slowly normalized discussing the topic publicly and across health disciplines, and D) interest in dissemination to MSM clients themselves. Respondents reported lower and more variable ratings for contributing, participating, and engaging in the training. This may relate to trainees’ ambivalence about both the mix of health professions within the training room and the focus on pleasure, as reflected in interviews. Wariness to participate in the training activities across professions is particularly important to note, because any prolonged and enduring discomfort during or after the training could have an antithetical effect to the intervention’s intended purpose of promoting comfort. Specifically, if participants flood with personal discomfort within the training room, this could inhibit opportunities to approach a typically anxiety-provoking topic with clients outside of the training room. Likewise, while the focus on pleasure was welcomed by many, if participants become preoccupied with an internal debate that pits pleasure against prevention, this false dichotomy could detract from learning to synthesize both of these relevant aspects of human behavior in interventions with clients. Revisions to the training could lessen the likelihood of pairings across the hierarchy of health professions – or frame this as a gradual exposure itself. Likewise, the training could examine the purpose of focusing on pleasure through deliberate guided discovery [60], and more saliently model how health workers could employ the training’s materials and its approach – health promotion rather disease prevention – toward health workers’ own public health goals.

Participants provided creative examples of how to integrate what they learned beyond one-on-one interactions, including hotlines, videos, workshops, and outreach. Their request for client-facing materials along with the high degree of interest in dissemination suggest that the content and recommended practices are not so much incompatible or inappropriate, but that implementation will require additional materials to support bringing the knowledge they valued to their colleagues and clientele. This also identified another target for the training intervention, to increase the comfort of clients to discuss this topic with HCPs, consistent with previous mixed-methods research among MSM living in the United States [23,62]. Interviewees specifically identified the need for additional job aids and tools that communicate their willingness to discuss anal health and sexuality, even before clients broach the subject.

China is considered a high stigma social context [61]: HIV prevention services among MSM are a national priority, as a health issue, but the country discourages social organizing that advocates for legal equality and protection of rights among sexual minorities. In this light, HCPs may indeed require more structural and institutionalized conduits that communicate social acceptance of MSM in their clinical settings, rather than simply the deployment of individual-level communication strategies within medical and counseling rooms, as the training recommended. Other high stigma settings [61] might likewise benefit from these tools. Their usefulness would likely apply even in relatively more accepting healthcare contexts, as suggested in studies in the United States, where stigma toward anal sexuality continues to be an impediment to HIV prevention [23,62].

The inclusion of MSM within the Smarter Sex training intervention may be warranted, given participant recommendations and the participants’ stated lack of opportunities for MSM to acquire accurate and professional knowledge about anal health and sexuality. Our pilot intentionally trained gay and bisexual HCPs alongside non-MSM physicians and public health workers, with reasonably high acceptability. This suggests the possibility of adapting and testing the intervention in a more explicit mix of professional and non-professional target groups, HCPs and MSM who are not health workers. This combination could build comfort within the training room among both populations that might then generalize to interactions within healthcare settings. Intergroup contact has the potential to reduce stigma [50], and forms the basis of other HIV-related stigma mitigation interventions for health workers that have shown promising results [e.g, 63, 64]. However, it should be noted that intergroup discomfort lingered among the present study’s MSM trainees, who were publicly known as both MSM and HCPs. This was true even as they sat alongside colleagues dedicated to delivering MSM health care and despite the fact that the training never required disclosure of personal sexual practices. This highlights that discussion of anal sexuality may pose a greater threat to social interactions than discussion of sexual orientation, as suggested elsewhere [65], and bolsters an argument for training MSM more generally.

To this point, there is also an increasing need to increase MSM’s health literacy at earlier points in their sexual development, to promote health before the accumulation of habits that lead to harm and to lessen the effects of stigma on their hesitation to engage in care [8,66]. Online approaches to health promotion among MSM, including young MSM, are also increasingly common and empirically-supported [e.g., 67-69]. This could be particularly compelling within a Chinese context where sex education is needed [70] and millions reportedly already access online venues and smartphone apps to seek sexual partners [71]. Internet-based versions of the training could target the promotion of comfort among clients who are unwilling to attend in-person training, shaping an approach to disclosure with HCPs so that conversation becomes more feasible and appropriate within ‘brick-and-mortar’ healthcare facilities.

Our findings are limited. Selection bias could bend our findings toward more positive feedback, if both non-interviewed participants and those who did not respond to online surveys found the training to be less palatable than those who did respond. It is also possible that some aspects of the training content were lost in translation from English to Chinese, despite the presence of a dedicated Chinese facilitator and an additional four as-needed translators to facilitate understanding among trainees. All measures involved self-report which could further bias our findings, despite attempts at protecting anonymity online, given the relatively small number of participants and their existing and ongoing relationship with the GPP collaboration. That said, the quantitative and qualitative results were consistent. Finally, this pilot study sample was implemented among already MSM-friendly HCPs, some of whom identified as MSM themselves, and the training mixed health disciplines, a heterogenous group of doctors, public health officials, and CBO workers. This participant diversity likely limits both internal validity and the generalizability of these pilot study findings across all HCPs, particularly those who might benefit most from increasing their competence and comfort with regard to both MSM and anal sexuality.

Still, even within a select audience of HIV health workers who specialize in MSM services, knowledge about anal physiology was reportedly low and discomfort broaching the topic of anal health and sexuality was notably high, supporting the need for further intervention. The high acceptability, feasibility, and appropriateness of the training warrants refinement of the training as well as bringing its content to MSM themselves, either in tandem with HCPs or online. Evaluations of impact could also determine if the training indeed operates as a stigma-mitigation intervention to increase the likelihood that discussions of anal health and sexuality are included as a routine component of MSM competent care.