Attitude shifts and knowledge gains: Evaluating men who have sex with men sensitisation training for healthcare workers in the Western Cape, South Africa

Background Men who have sex with men (MSM) in South Africa experience discrimination from healthcare workers (HCWs), impeding health service access. Objectives To evaluate the outcomes of an MSM sensitisation training programme for HCWs implemented in the Western Cape province (South Africa). Methods A training programme was developed to equip HCWs with the knowledge, awareness and skills required to provide non-discriminatory, non-judgemental and appropriate services to MSM. Overall, 592 HCWs were trained between February 2010 and May 2012. Trainees completed self-administered pre- and post-training questionnaires assessing changes in knowledge. Two-sample t-tests for proportion were used to assess changes in specific answers and the Wilcoxon rank-sum test for overall knowledge scores. Qualitative data came from anonymous post-training evaluation forms completed by all trainees, in combination with four focus group discussions (n = 28) conducted six months after their training. Results Fourteen per cent of trainees had received previous training to counsel clients around penile–anal intercourse, and 16% had previously received training around sexual health issues affecting MSM. There was a statistically significant improvement in overall knowledge scores (80% – 87%, p < 0.0001), specifically around penile–anal intercourse, substance use and depression after the training. Reductions in negative attitudes towards MSM and increased ability for HCWs to provide non-discriminatory care were reported as a result of the training. Conclusion MSM sensitisation training for HCWs is an effective intervention to increase awareness on issues pertaining to MSM and how to engage around them, reduce discriminatory attitudes and enable the provision of non-judgemental and appropriate services by HCWs.


Introduction
http://www.sajhivmed.org.za Open Access contribute to the high HIV burden amongst MSM. Behavioural risk factors include condomless penile-anal intercourse, multiple sex partners and substance use in the context of sexual encounters. 6,11,14,19,20,21,22 Social and structural risk factors include stigma and discrimination towards MSM, 23 low levels of education and high rates of poverty amongst MSM from previously disadvantaged communities 6 and limited access to MSM-appropriate health services. 23 Discrimination and internalised homophobia are correlated with increased risktaking behaviour amongst MSM. 24, 25 Further, MSM often experience discrimination from public sector HCWs, 23,26 negatively affecting health-seeking behaviours and disclosure of risk practices. 27 For the purposes of this article, we use the word 'sensitisation' to refer to the process of increasing awareness and knowledge of an issue to instil empathy and effect the modification of negative attitudes and behaviour, with the intention of reducing discrimination and inequality. Sensitisation training assists people in examining their personal attitudes and beliefs. 28 Sensitisation training for HCWs around MSM is recommended by the World Health Organization 29 and has been employed by other organisations in South Africa and in African other countries (e.g. Nigeria and Malawi) 30 ; however, we were only able to find published data on the evaluation of the HCW training from Kenya (discussed later), 31 which was found to increase knowledge and support empathetic healthcare responses amongst HCWs towards MSM. 31

The training intervention
In 2009, the Desmond Tutu HIV Foundation (Cape Town, South Africa) and the KEMRI-Wellcome Trust Research Programme (Kilifi, Kenya) developed a training manual to equip HCWs with the knowledge, awareness and skills to provide sensitive and appropriate services to MSM. 32 HCWs were conceptualised to include clinical (e.g. nurses, doctors and counsellors) and support staff (e.g. clinic managers, outreach workers and social workers). Counsellors and nurses were the primary audience for the training.
In 2011, the manual was revised based on recommendations from previous training participants and included updates around HIV epidemiology, advances in biomedical prevention research and additional case studies and interactive exercises. 33 The training manual covered homophobia, stigma and discrimination, HIV and sexually transmitted infections (STIs), sexual orientation and gender identity (including a detailed explanation of the term MSM), common sexual practices, condoms and lubricants, mental health issues, substance use and risk-reduction counselling for MSM. The training did not include capacity development around the clinical management of health and related conditions affecting MSM.
We used the manual to provide sensitisation training to 592 public-and private-sector HCWs in the Western Cape province of South Africa, between February 2010 and May 2012. Training sessions were conducted over one or two days, depending on operational requirements, and included 10-20 trainees per training. The training programme was advertised by means of electronic communication and word of mouth. Training workshops were conducted in English and combined facilitator-led presentations, interactive group exercises, question and answer sessions, and opportunities for individual self-reflection. Training was provided free of charge and each trainee received a manual.

Research design
Quantitative and qualitative methods were used to evaluate the training programme.

Quantitative component
Trainees completed written pre-and post-training assessment questionnaires, before the start of, and at the end of, training. Pre-training assessments included previous training around MSM sensitisation, penile-anal intercourse and counselling MSM clients. Pre-and post-training assessments were taken from the MSM training manual and examined the sensitivity training manual curriculum. 34 In short, the trainees' baseline knowledge relates to common STIs amongst MSM; stigma, discrimination and homophobia and their effects; common sexual practices amongst MSM; common mental health problems experienced by MSM; and condom and lubricant use for penile-anal intercourse. Post-training assessments examined trainees' perceptions of the need for HCWs to be sensitive to MSM-related issues, MSM-appropriate service provision, personal beliefs and attitudes as well as comfort in counselling MSM around penile-anal intercourse. The assessments included closed questions with categorical variables (true or false; yes or maybe or no). A score was calculated based on the 15 knowledge questions included in the training manual. Trainees' demographic data were not collected, and participants were not requested to write down any identifying information on their assessment forms. Data from the assessment forms were entered into an Excel spread sheet and imported into Stata Corp v 11.2 (College Station, Texas). Pre-and post-training proportions for correctly answered knowledge-related questions were calculated and compared using the two-sample t-test for proportions. Changes in average pre-post training knowledge scores were assessed using the Wilcoxon rank-sum test. It was not possible to link pre-and post-assessments to individuals as participant details were not captured.

Qualitative component
All trainees completed a semi-structured, anonymous written evaluation form immediately after the training. The data from these forms were typed up verbatim and included in the qualitative analysis, alongside the data from focus group discussions (FGDs). Attempts were made to contact all of the HCWs who attended training in Cape Town between February and May 2010 (n = 167) via email or telephone. Trainees were then invited to take part in an FGD six months after receiving the training. Twenty-eight of the trainees (17%, 28/167) took part, and FGDs were scheduled to suit their availability. Four semi-structured FGDs were held in Cape Town during October and November 2010. FGD respondents included 18 counsellors, two nurses, four management-level HCWs, an outreach worker, one social worker and two people who did not stipulate their position. Topics of discussion included self-perceived attitude shifts towards MSM, experiences since the training in providing services to MSM clients, and providing services relating to penile-anal intercourse. The lead trainer conducted the FGDs in English. FGDs were audio recorded and accompanied by written notes. Audio recordings were transcribed verbatim, with identifying information omitted. An iterative open-coding analysis process was used, which involved comparing, contrasting and conceptualising the data across key emergent themes. One person conducted all of the coding. Qualitative analysis followed a grounded theory approach based upon the generation of provisional hypotheses and theories from the data, with subsequent elaboration and verification stages. 'Manifest' meaning was obtained through analysis of the key themes and 'latent' meaning from reflection on these and the findings from other components of the evaluation process, the context, and after discussion between the primary and secondary authors.

Ethical consideration
Approval for this evaluation was granted by the University of Cape Town's Faculty of Health Sciences Human Research Ethics Committee. Written informed consent to use the information collected was obtained from trainees. Personal details of the trainees were not captured on evaluation forms or procedures to protect their identity. Overall, there was a statistically significant increase in the composite knowledge scores, increasing from a median score of 80% at pre-training to 87% at post-training (p < 0.0001). An overview of pre-post training scores per assessment question is provided in Table 1. There was a statistically significant increase (p < 0.05) in the number of trainees who correctly answered the questions around stigma, HIV and MSM, anal sex practices amongst non-MSM, the relative risk of condomless anal sex, awareness of STIs affecting MSM, substance use and depression after the training. After the training, 75% of the trainees reported that they felt comfortable asking future clients about penile-anal intercourse, and 96% reported that they were aware of the potential risks associated with common sexual activities that MSM may engage in.

Qualitative component
The following section presents qualitative data collected from the post-training evaluation forms and the four FGDs. The qualitative data fall into four thematic areas: (1) reflections and perceptions of changes in knowledge and shifts in attitude, (2) implications of these on service provision and care, (3) barriers to service provision and (4) recommendations for implementation.

Effect of the training on knowledge and attitudes
Religious and cultural beliefs were cited by many respondents as sources of their own prejudice about homosexuality being The perception of feeling more comfortable in discussing penile-anal intercourse and same-sex practices after receiving the training was a commonly felt attitude shift amongst trainees. Respondents used phrases like 'more at ease', 'more comfortable' and 'more empowered' to describe their feelings towards having MSM clients and talking about anal sex generally after the training: Trainees who work directly with clients in the healthcare setting remarked on their realisation that they had been making heterosexist assumptions about their client's sexual partner or partners or practices without asking the client themselves: 'It's not only with the MSM clients, it's with the clients who come there in general. That you go into the room not just assuming that… because of their gender, they're going to choose the opposite sex… so you go in there with that in the back of your mind. So the kinds of questions you ask now is a little different than before. Because before you were thinking, ok so this is a woman client, so you're always asking her about her husband or Those respondents who felt that they had been relatively well informed and non-judgemental before the training said that their new knowledge gains served to increase their motivation to provide MSM-sensitive services: 'It's not necessarily that I've changed, but the training motivated me more… having that manual and applying that and the knowledge, I have become more motivated.' [Male, FGD] There was recognition from some respondents of the need for the training even though the South African Constitution protects the rights of all people regardless of sexuality or gender: 'This training is very important and covers almost every aspect on working with MSM. This is really needed as our country is now sensitive and acknowledges MSM, so as means of our longfighting battle of reducing the growing number of HIV, this MSM training is needed.' [Anonymous post-training evaluation form]

Experiences providing services to men who have sex with men after the training
A general feeling amongst FGD respondents was that since the training, they had been able to provide better services to MSM clients, as a result of being more informed, equipped with appropriate skills and knowledge, and prepared to ask appropriate questions of their clients: 'I think it's almost a relief for a lot of people to actually be able to tell someone who's not going to pass judgement. They go off and tell you just about everything you want to know… Once you ask the questions the person is comfortable and then they come out… the person is relaxed now.' [Female, FGD] Respondents reported that they also felt more comfortable in the company of MSM clients and discussing penile-anal intercourse; as a consequence, clients felt more comfortable to disclose their sexual behaviour: 'I've had two MSM couples, and they were free and talking, they felt the environment was free, they were talking about everything, so I'm so glad also that the couples are also coming in, and the MSM can feel free to express everything.' [Female, FGD] Some of the HCWs reported that after receiving the training, they had initially been concerned about having to ask clients about anal sex: 'I thought when we initially asked the anal sex question that people would be affronted, but surprisingly not… those who don't [have anal sex] will just say 'no' [sounding surprised] and then you just move on. I was quite surprised… I do warn people that it's quite personal the questions, and they have to see it in a broad context, because we have to speak to everybody, and this is just to see people's sexual behaviour and so on… they seem to sort of take to it OK.' [Female, FGD]

Barriers to service provision
In terms of barriers to service provision for MSM, one respondent stated that even if HCWs are sensitised about anal sex, the risk assessment forms used in the public sector do not include questions on anal sex, and therefore, the HCWs will not ask about it: 'We've had people saying to us that they won't ask the questions if they are not on the forms, because they don't know how to ask them.' [Female, FGD] Overcoming HCWs' personal beliefs and opinions was identified as important to provide necessary services to clients:

Recommendations for implementation
A commonly occurring recommendation was that the existing tools used for risk assessment and screening by clinics needed to be revised to include questions on penileanal intercourse, anal STIs and MSM behaviour.
Another commonly cited issue was the lack of informative and educational materials relating to MSM, sexuality and penile-anal intercourse available in health facilities. Respondents felt that the visibility of such materials would make clients feel comfortable and would encourage disclosure: 'People will sit there and read (posters and pamphlets)… it will encourage them to talk about it. But if they don't see it anywhere, they won't now come and ask the sister.' [Female, FGD] In terms of training requirements, respondents suggested that follow-up refresher trainings would be useful, combined with online reference and referral resources. It was also suggested that sensitisation of all levels of staff working in health facilities needs to take place: 'We need to bring all levels of staff in order for the changes to happen. Even starting from the security, if the security is discriminating, they (MSM) won't even come inside. The receptionist, everyone who works inside, the doctors and nurses too.' [Male, FGD] http://www.sajhivmed.org.za Open Access

Discussion
The findings of this research suggest that a two-day or less, low-cost, in-service sensitisation training for HCWs can reduce homo-prejudicial attitudes and increase relevant knowledge and relational skills. This training intervention increased knowledge and awareness of the health risks, social vulnerabilities and specific needs of MSM amongst those who completed assessment forms. However, the quantitative elements of the assessment forms did not assess shifts in values or moral views. Increased HCW skills and comfort in providing services to MSM was identified through the qualitative methods. Focus group participants reported shifts towards being less judgemental and discriminatory towards MSM in their healthcare settings. It is likely that changes such as these will help to reduce barriers impeding MSM from accessing health services.
The relatively low number of trainees that had counselled MSM prior to receiving the training may indicate that MSM had not been accessing HIV services at that time. Another potential explanation is that MSM who were accessing health services may not have disclosed same-sex behaviours to HCWs. Furthermore, very few trainees had received previous training around penile-anal intercourse, despite penile-anal intercourse being the most efficient way of transmitting HIV sexually. 35 The baseline knowledge of participants appeared to be higher than the baseline knowledge amongst HCWs in Kenya who were trained using the same curriculum. 24, 31 Constitutional protection against discrimination on the basis of sexual orientation and gender 36 ; government policy around non-discriminatory public service provision (Batho Pele Principles) 37 ; inclusion of MSM in the National Strategic Plan for HIV, STI and TB 38 ; and existence of an established lesbian, gay, bisexual and transgender 'community' within the Western Cape's major metropolitan area (Cape Town) 9 may have contributed to the higher levels of knowledge amongst HCWs in the Western Cape compared with the Kenyan counterparts.
Healthcare workers admitted to previously making heterosexist assumptions when conducting sexual risk assessments, an issue that has been previously identified as limiting disclosure of same-sex sexual behaviour. 39 General knowledge around the social, structural and health vulnerabilities of MSM was also low, highlighting the need for improvements in HCW knowledge around MSM-specific sexual health information. The training and evaluation exposed judgemental, moralising and homo-prejudicial attitudes amongst some HCWs towards MSM, demonstrating the challenges involved in creating an enabling environment for MSM to access healthcare, as well as the limitations of once-off sensitisation training interventions. The qualitative evaluation revealed some shifts in attitudes; however, on several occasions, HCWs referred to MSM in the third person, suggesting that much work is needed for sexual minority groups to be embraced by them.
Religious and cultural beliefs play a key role in informing people's attitudes towards homosexuality. 40,41 It is imperative that healthcare providers distinguish between their right to hold personal values and beliefs, and their professional obligation to render services free of prejudice and discrimination; it is a requirement of professionalism that personal beliefs be set aside when they conflict with professional duty. 40,42 One element of sensitisation training involves making HCWs aware of the distinction between their right to 'hold personal values, beliefs and prejudices', with 'their professional obligation to render services free of prejudice and/or discrimination'. 42 The evaluation of the MSM sensitisation training programme for HCWs in Kenya found similar results regarding the effects of training. 24 The two-day training intervention in Kenya, which included computer-based sessions followed by group discussions, significantly increased trainee knowledge on issues relating to MSM. 24 Notable knowledge gains related to the following areas: the effects that stigma and discrimination towards MSM can have on their vulnerability to HIV; penile-anal intercourse is practiced by men and women regardless of their sexuality, rather than only between MSM; the physiology of penile-anal intercourse, and related STI and HIV transmission risks; and increased awareness around mental and sexual health issues affecting MSM. 24 The evaluation highlighted how training alone is unlikely to shift practices and that it needs to be accompanied by institutional changes, including provision of tools that HCWs can use to implement the new knowledge (e.g. sexual reproductive health assessment forms that are gender neutral and can capture a range of identities, genders and sexual practices). Sensitisation training is an important step towards appropriate health services for MSM. In addition, skills to appropriately manage health conditions of particular relevance for MSM need to be developed and has started to take place in South Africa. 30,43,44 Limitations of the study Limitations to this research included the following: (1) we were unable to link pre-and post-assessment questionnaires to individual trainees, and as a consequence, limited reflections can be made on changes in individual trainees versus the group; (2) trainee demographic information was not recorded on the assessments, preventing further analysis; (3) response bias and the willingness of trainees to frankly discuss the programme and provide criticism without concern for causing offence are likely as the lead trainer also facilitated the FGDs; (4) FGD respondents were trainees who agreed to return after the training, which may have led to sampling bias -the trainees who agreed to participate in the qualitative aspects of the evaluation may have been more positive about the training; and (5) the evaluation did not assess the longterm effects of the training, and it is not possible to say if changes identified in the evaluation were durable.
Another potential limitation, unrelated to the research methods, but to the training itself, was that the facilitator was an English-speaking (not isiXhosa or Afrikaans, the two predominant first languages in the Western Cape) woman. Notably, the trainings were not facilitated by an openly gay man, as such, the female facilitator may not have been able to accurately relate the concerns or priority issues of MSM.
Arguably, the positioning of the trainer as a neutral 'outsider' may have made it easier for trainees to be open and honest about their prejudices. 45 The term 'MSM' groups men together with different sexual practices and risk, social identities and gender expressions.
Although it is useful, the term and approach have limitations. Future training programmes and subsequent service delivery would be improved if greater attention were paid to the various sub-groups that the MSM term encompasses, and how sexual practices, gender identity and sexual orientation are influenced by a range of factors, which may also change over time.

Conclusion
The findings of this study evaluating the MSM sensitisation training for HCWs in the Western Cape province of South Africa suggest that this training intervention increased knowledge around MSM issues, reduced judgemental and homo-prejudicial attitudes, and improved self-perceived skills and capacity in providing appropriate and relevant healthcare to MSM. The training was particularly beneficial in increasing awareness of MSM and knowledge relating to the issues of importance for MSM, specifically HIV, STIs, social stigma, mental health, substance use, sexual risk behaviours, anal health, and correct condom and lubricant use.
These findings suggest that this type of sensitisation training should be scaled up nationally and integrated into health service provider training. MSM sensitisation training for HCWs is an effective intervention to enable the provision of non-judgemental and appropriate services by HCWs and to increase awareness of unique issues pertaining to MSM and how to manage them. However, as seen from these results, even if HCWs are sensitised, without improved reporting systems and tools that are more inclusive and disaggregated, questions relating to sexuality and anal sex will not be asked, thus hampering effective service provision.
Focus should remain on ensuring that training and reform, which support the implementation of training content, contribute to enhanced access to quality care for MSM. Efforts to increase the efficiency of sensitisation training and to measure the long-term effectiveness thereof are also needed. Training should include other stigmatised groups at risk for HIV and should include additional focus on clinical aspects, including the management of anal STIs amongst patients. To be sustainable, training around MSM and related issues should be included in HCW pre-service training.