Rapid assessment and response adaptation guide on HIV and men who have sex with men

WHO/HIV/2004.14

RAPID ASSESSMENT AND RESPONSE ADAPTATION GUIDE ON HIV AND MEN WHO HAVE SEX WITH MEN

World Health Organization

Department of HIV/AIDS

RAPID ASSESSMENT AND RESPONSE ADAPTATION GUIDE ON HIV AND MEN WHO HAVE SEX WITH MEN

World Health Organization Department of HIV/AIDS

2

© World Health Organization 2004 All rights reserved. Publications of the World Health Organization can be obtained from Marketing and Dissemination, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel: +41 22 791 2476; fax: +41 22 791 4857; email: bookorders@who.int). Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to Publications, at the above address (fax: +41 22 791 4806; email: permissions@who.int). The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. The World Health Organization does not warrant that the information contained in this publication is complete and correct and shall not be liable for any damages incurred as a result of its use.

Authorship This manual was prepared by Paul Boyce and Peter Aggleton with Anne Malcolm of Thomas Coram Research Unit, Institute of Education, University of London. Gundo Weiler, Monica Beg and Andrew Ball of WHO edited the document under the supervision of Isabelle de Zoysa, Director Prevention, Department of HIV/AIDS, WHO For further information please contact: World Health Organization; Department of HIV/AIDS; Treatment and Prevention Scale up; CH-1211 Geneva 27; Switzerland FAX: +41 22 791 4834; E-mail: hiv-aids@who.int Acknowledgements

An expert panel reviewed this document and provided valuable inputs. WHO wishes to acknowledge their contribution in finalizing the document.

3

Chapter 1 Introduction Contents • • • • 1. Introduction RAR, HIV/AIDS and men who have sex with men What is an RAR? Audience Introduction

This adaptation guide for work with men who have sex with men offers guidelines on how to conduct a Rapid Assessment and Response (RAR) focusing on lifestyles, behaviours and HIV/AIDS concerns. It outlines a series of simple and practical activities that may be used to explore the circumstances, experiences and needs of men who have sex with men across a variety of settings. It is designed to be used either in conjunction with the WHO Rapid Assessment and Response Technical Guide (TG-RAR) or as an independent resource. Chapters in this manual offer guidelines on carrying out an RAR with men who have sex with men. Specific guidance is offered on the following issues: ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ 2. Understanding HIV/AIDS prevention among men who have sex with men Scope and focus of an RAR Planning of an RAR on HIV/AIDS and men who have sex with men A model for research and response with men who have sex with men Community participation and advocacy Training for RAR staff and volunteers Conducting of an initial rapid assessment Identification of methods appropriate for an RAR Analysis of RAR research findings Development of an action plan Strategies for monitoring and evaluation

RAR, HIV/AIDS and men who have sex with men

Recent UNAIDS and World Health Organization (WHO) estimates acknowledge that maleto-male sexual transmission is a major route of HIV transmission in many regions of the world (UNAIDS/WHO, 2002). However, the absence of systematic data on male sexual health in many countries means that it is difficult to estimate how many men may have been affected. Additionally, the stigmatization and/or marginalization of men who have sex with men makes it especially hard to develop even basic estimates of male-to-male sexual risk. Male-to-male sex is often secretive or hidden. Only occasionally do the individuals concerned have an explicit ‘homosexual’ or ‘gay’ identity. Given the widespread secrecy surrounding male-to-male sex in many countries, ‘evidence’ of male-to-male sexual health risks, especially those associated with HIV/AIDS, has mostly emerged from the campaigning activities of community groups of men who have sex with men (Parker, Khan & Aggleton, 1998; Parker et al., 2002). Work of this kind has raised awareness that men who have sex with men do indeed exist in cultural contexts where their presence is denied, and that many such men are at risk of HIV infection. 4

Men who have sex with men should not be thought of as a single, contained group of people but rather as a diverse population of men from across the social spectrum, with various ways of thinking about and experiencing their sexuality. An RAR should always aim to take account of the diversity of men who have sex with men within any given setting. It should focus on the varied lifestyles and sexual behaviours of such men, in order to facilitate a better understanding of the HIV/AIDS risks faced by men who have sex with men, and to highlight how male-to-male sex is intimately linked to HIV/AIDS epidemiology more broadly. 3. What is an RAR?

Rapid Assessment and Response (RAR) is an approach that has been developed to provide an assessment of complex health issues and behaviours within a short time frame. On the basis of the data collected, it offers a means for responding quickly with appropriate programme measures and interventions. Definition: Rapid Assessment and Response (RAR) Rapid Assessment and Response (RAR) is a way of making a comprehensive assessment of a particular public health issue. It involves focusing on the characteristics of the health problem, the population groups affected, key settings and contexts, health and risk behaviours and social consequences. It identifies existing resources and opportunities for intervention, and helps plan, develop and implement interventions and programmes.

The guidelines for conducting an RAR promote a community-based approach to research. This builds on methods developed by WHO over a number of years, and which have proved effective in conducting research and developing responses with various populations in many different settings. 4. Audience

This adaptation guide is intended to be useful to: • policy-makers and programme planners, to help them decide on how to proceed with conducting an RAR with men who have sex with men; • researchers who may be seeking particular tools to use in working with men who have sex with men; and • members of community-based organizations seeking to develop local responses to issues affecting men who have sex with men. The guide may be used on its own to assist programme planners in deciding on whether to conduct an RAR and how to prepare for it. However, for those who may be implementing an RAR, this guide is best used in conjunction with the Technical Guide (TG-RAR) itself.1 The structure of this RAR manual closely follows that of the TG-RAR. Each chapter of the RAR manual links to the corresponding chapter in the TG-RAR through section headings and numbering.

1

This can be downloaded from: http://www.who.int/docstore/hiv/Core/Index.html.

5

Chapter 2 HIV/AIDS prevention and men who have sex with men Contents • • • • • • • 1. Why conduct an RAR on HIV/AIDS and men who have sex with men Why ‘men who have sex with men’? Men who have sex with men, HIV/AIDS and the ‘general population’ Men who have sex with both men and women Scope and focus of an RAR Issues to be explored in an RAR Sexuality and social context Why conduct an RAR on HIV/AIDS and men who have sex with men?

The term ‘men who have sex with men’ can evoke strong reactions. This response can vary from rejection, shock and prejudice to curiosity and acceptance. In some countries (particularly Australia and countries in Europe and North America), men who have sex with men have been identified as especially vulnerable to HIV/AIDS. This has sometimes provoked prejudicial reactions against the easily identifiable groups of men who have sex with men—such as men who identify as gay. However, the association between gay men and HIV/AIDS has also helped to bring about more supportive responses. For example, as a consequence of community-based activism, in a number of countries health promotion, care and support programmes for HIV-positive men who have sex with men have been developed. In other countries, HIV/AIDS has been more strongly associated with heterosexual activity. This has been particularly so where there is little or no public acknowledgement of male-to male-sex. Denial of sex between men in these settings has meant that for the most part there has been little or no attempt to collect relevant epidemiological information on HIV transmission through male-to-male sex. In a context where male-to-male sex is denied, it may also be difficult to access men who have sex with men for the purpose of needs assessment and other health-focused research (McKenna, 1996). Despite these difficulties, the HIV/AIDS-related needs of men who have sex with men in developing countries are gradually receiving more attention. For large part, this has been due to the pioneering efforts of community-based organizations run by and for men who have sex with men (Parker, Khan & Aggleton, 1998). In some regions of the world—particularly in parts of Asia and Latin America—programmes and interventions for men who have sex with men have become well established. In Bangladesh, for example, the Bandhu Social Welfare Project offers a successful example of community mobilization and health promotion by men who have sex with men, funded and supported by international donor agencies.2 In many parts of the world, however, and particularly in many African countries, health interventions for men who have sex with men remain rare or poorly developed. Although recent work in South Africa (Donham, 1998) and Senegal (Niang et al., 2002) has begun to document the lifestyles and sexual health needs of men who have sex with men, work of this kind is still quite rare.

2

http://www.unaids.org/publications/documents/specific/men/msmpve.pdf

6

Most men who have sex with men conduct their sexual lives in secrecy and fear of being exposed. Vulnerability to HIV/AIDS is compounded by a number of factors, including the following. Marginalization and stigmatization means that in many countries men who have sex with men have been poorly served by HIV/AIDS prevention and care. In these circumstances, many individuals may lack understanding of HIV transmission and sexual health risks associated with unprotected sex. Research in India conducted in the early days of the HIV/AIDS epidemic indicated that because of media discussion associating HIV transmission with heterosexual sex, some men believed that sex with other men was completely safe (Oostvogels & Menon, 1993). Such beliefs can persist where health promotion around maleto-male sex is less than adequate. Stigma and discrimination can make it hard for men who have sex with men to talk with doctors and other health professionals about sexual health issues. Health workers may hold prejudicial attitudes about men who have sex with men. This means that not only are men who have sex with men less likely to receive accurate health information, they may also avoid going for treatment for sexual transmitted infections (STIs), particularly those associated with anal sex. The presence of concurrent STIs can mean that HIV is more easily transmitted, thus placing men who have anal sex with other men at increased risk of infection. For some men who have sex with men, decision-making about safer sex is based on assessments of a sexual partner’s personal qualities, rather than a more rational assessment of risk. Recent ethnographic work on sexual decision-making processes among men who have sex with men in Beijing, China found that men decided whether or not a male sexual partner was safe on the basis of whether or not that person appeared to be ‘trustworthy’, or of clean appearance (Choi et al., 2002).3 Studies also suggest that younger males who are not self identifying as men who have sex with men are more vulnerable to HIV infections and may be less self accepting of their sexual identity. Men self identifying as bisexual rather than gay are reported to have less factual knowledge about HIV and less intentions with regard to safer sex (Kelly et al., 2002; Waldo et al., 2000). In other situations, sex between men may have little to do with choice. This is most obvious in situations where men are physically coerced into having sex with other men. Accounts of this type of sexual activity are common in many countries. For example, in research in Nepal men who have sex with men in Kathmandu told of incidents in which they had been forced into having sex with other men against their will (Boyce, unpublished). In prisons too, as well as in predominantly male populations such as the military, sex between men may be frequent (Connell, 1995; Simooya and Sanjobo, 2001). Transactional or paid sex may also involve coercion and force, although men who sell or trade sex may have more immediate control over choosing the men they sell sex to, and the form of sex that takes place, than women do (UNAIDS, 2000; Aggleton, 1998). Finally, a variety of structural factors may enhance vulnerability in sex exchanged for money or other types of reward. Men who sell or trade sex often do so because they have few other

It should be noted here that it is not only men who have sex with men who make decisions about sexual safety based on criteria of this kind. Most people make choices about their sexual activities based on how they think or feel about their sexual partners rather than on purely rational assessments of risk.

3

7

choices about how to make a living. In this kind of context, it may be hard to resist the demands of some clients for unprotected sex, particularly if higher payment is offered.

8

Anal sex and condom use Unprotected anal sex presents one of the most significant risks of HIV transmission, if one of the sexual partners involved is HIV positive. While anal sex is often not the only sexual activity in which men who have sex with men engage, and some men may never engage in it at all, it is the activity that presents the highest risk of HIV transmission during male-to-male sex. The proper and consistent use of condoms during sex is the most effective barrier to HIV transmission for both anal and vaginal intercourse. For anal sex, it is advisable to use good quality strong condoms together with a water-based lubricant. Water-based lubricants minimize the friction between the condom and the anus. Oil-based lubricants on the other hand can cause condoms to break. In many countries, condoms suitable for anal sex and water-based lubricants are not readily available. This creates special challenges for condom promotion work.

2.

Why ‘men who have sex with men’?

While in some parts of the world, words such as ‘homosexual’ and ‘gay’ describe some men who have sex with other men, in other contexts these terms may not be well known, meaningful or translatable. Therefore, the phrase ‘men who have sex with men’ has become popular as a way of talking about sexual behaviour between men, irrespective of how individual men understand their sexual identities and/or feelings and experiences. Importantly, the term ‘men who have sex with men’ includes men who do not identify with any term describing their attraction to other men, and who appear conventionally ‘heterosexual’ from mainstream society’s perspective. In many cultures, there is no firm division between men who have sex with men and men who have sex with women. In some, perhaps many, men do both. Male-to-male sex, therefore, does not exist in isolation. Rather, it takes place in sexual networks that are connected to the so-called ‘general population,’ of which men who have sex with men are a part. Throughout this guide, we use the term ‘men who have sex with men’ in preference to other descriptions such as ‘gay’, ‘homosexual’ and ‘MSM’. We have avoided such terms because, as indicated above, these terms are culturally specific and may not meaningfully describe the types of male-to-male sex found in many parts of the world. The phrase ‘men who have sex with men’ is not perfect, however. Not all men who have sex with other men think of their behaviours as explicitly sexual. Describing such men as men who have sex with men can, therefore, be problematic as it imposes a sexual definition onto acts that may be understood in different ways by men themselves. Additionally, not all men necessarily think about their gender in the same way. For a few men who have sex with men, a feminine or ‘female’ identification may be an important attribute of sexual subjectivity. Simply describing such men as men who have sex with men may be insensitive to their understanding of themselves. Finally, it is important to note that the term ‘men who have sex with men’ has itself evolved so as to allow us to talk about sex between men in public health settings. It is not a phrase that will necessarily be understood all over the world, or be readily or meaningfully translated into all languages. The shortened version of the phrase ‘men who have sex with men’—‘MSM’—has become a convenient shorthand way of talking and writing about men who have sex with men in some contexts. We prefer the longer phrase ‘men who have sex with men’ though, since it encourages us to think about issues more precisely: not just as a convenient phrase, but also as a way of describing a diverse population of men who, for the purposes of an RAR, are labelled together because they all have sex with other men. Throughout this guide, therefore, 9

the full term ‘men who have sex with men’ is used to encourage clearer thought about the diverse cultural contexts, lifestyles and health needs of men who have sex with other men. 3. Men who have sex with men, HIV/AIDS and the ‘general population’

While the emphasis in this guide is on HIV/AIDS as it relates to work with men who have sex with men, it is misleading to think of men who have sex with men as an easily identifiable ‘separate group’ in society. While in some places, men who have sex with men have formed community groups, and some of these men may share similar lifestyles, such outwardly visible and ‘organized’ groups of men comprise only a small part of a far larger population of men who have sex with men, who may be less visible within the general population. It is important to keep this in mind when planning data collection, since most RAR work will want to identify HIV/AIDS-related issues relevant to a broad population of men who have sex with men. At the planning stage, remember that in thinking about men who have sex with men it is important to bear in mind that many of the men who are eventually worked with may not belong to an obviously identifiable group. Perhaps the majority of men who have sex with men are not easily distinguished by the way they dress or behave. Indeed, as discussed above, due to prejudice and stigma many may not want to draw attention to their sexual orientation. 4. Men who have sex with both men and women

It is essential to think about men who have sex with men as a part of the general population, rather than as a separate group. This is especially so because in some countries it has been found that there is no clear division between men who have sex with men and men who have sex with women. Across large parts of Africa, Asia and Latin America, many men who have sex with men also have female sexual partners, and substantial numbers may be married (Aggleton, 1996). HIV/AIDS among men who have sex with men cannot simply be thought of as a contained and separate epidemic among a distinct population of men who can only infect one another. Instead, it is more accurate to think of HIV as an epidemic that is neither ‘heterosexual’ nor ‘homosexual’ but as something that affects the general population, of which men who have sex with men are themselves a part. 5. Scope and focus of a "RAR on HIV and men who have sex with men"

A "RAR on HIV and men who have sex with men" consists of two principal components.

Rapid assessment of HIV/AIDS and men who have sex with men

Response through programmes and interventions to reduce transmission and impact of HIV/AIDS among men who have sex with men

For further information about the RAR process, its origins and links to programme development, see Chapter 2 in the TG-RAR. RARs vary dramatically in the geographical area focused upon. At a very broad level, they may focus on a particular country, state, city or region. More usually, they explore needs and experiences within a neighbourhood or a particular social space such as a park or community group. No matter what its scale, in carrying out an RAR it is important to examine cultural 10

attitudes, along with policy and legislation at the national level, as they affect attitudes towards male-to-male sexuality. More specifically, it is valuable to focus on specific locations where men who have sex with men socialize or look for male sexual partners or both. Such places—sometimes called ‘cruising areas’ in English—can include certain parks, public squares, cinema halls, transport hubs or any social space where people congregate. It is a good idea to find out about the location of such places and to explore how men who have sex with men utilize such environments. Are some places only used after dark, or are some places popular at all times of day? Although ‘cruising areas’ can offer a useful location in which to conduct an initial assessment, they are not the only places where men who have sex with men spend time. Additionally, many men who have sex with men do not socialize or look for sexual partners in such localities. Because men who have sex with men are part of the general population, not a separate group in society (see previous chapter), it is important to take this into account when thinking about the scope and focus of an RAR. Try not to focus just on ‘cruising areas’ and other easily identifiable spaces such as bars and night clubs, but consider other localities too, such as the work place, the family home, and indeed any place where men who have sex with men may be located. Consider how men manage their presentations of who they are in these different contexts by asking men about the differences between places in which they can openly present themselves to other people as men who have sex with men, and the places where they conceal their sexual orientation. Such an approach offers a good way of exploring how geographical and social settings can help to conceal the presence of men who have sex with men within the general social context. Importantly, this concealment can often be beneficial to men, as it can mean they are able to make contact with other men for sexual purposes without drawing attention to themselves. A good RAR should try to explore the different dimensions of male-to-male sexual life, to offer an analysis of how men who have sex with men locate themselves differently in different social and geographical settings. 6. Issues to be explored in an RAR

RARs vary with respect to the type of issues examined. In conducting an RAR with men who have sex with men, issues important to focus on include: • • • • • • • • • stigma and prejudice as they affect the lives and sexual activities of men who have sex with men; violence and harassment and their relationship to policing practices; emotional and mental well being of men who have sex with men and their effects on sexual behaviour and lifestyles; economic restraints faced by men who have sex with men; the experiences of men who sell sex to other men; knowledge of safer sex, STIs and HIV/AIDS; prevalence of STI symptoms; cultural beliefs about sex, sexuality, sexual health and HIV/AIDS; understandings of risk and safety in men’s lives, and their relation to perceptions of HIV/AIDS; and 11

•

attitudes towards and practices relating to drug and alcohol use and the potential affects of these on sexual behaviours and HIV infection.

The issues outlined here overlap. It is also likely that other issues will emerge as important during the course of the work. 7. Sexuality and social context

When conducting research on sexual behaviour, it is important to take into account the context in which behaviours occur. Sexual life is not purely individualistic but is informed by cultural values and attitudes. The way people think about their sex, their bodies and sexual activities is shaped by national and local social attitudes and cultural beliefs. However this does not mean that the individuals’ feeling about their own sexuality will always be shaped in the same way in any culture. Each individual will interact with these local cultural and societal influences to develop a personal understanding of their own and to others’ sex and sexuality. This is why, in addition to enquiry into sexual behaviour and more obvious sexual health issues, an RAR also needs to explore general social issues and personal experiences. This will facilitate the development of more sensitive and appropriate programmes and interventions. Recreational drug use seems to increase sexual risk taking behaviour, particularly unprotected anal intercourse among men who have sex with men. Clearly using substances intravenously also puts men at risk if HIV transmission through unsafe injecting as well as sexual exposure (Strathdee et al., 1998).

12

Chapter 3 Applying RAR principles Contents • Key features of an RAR

Chapter 3 of the TG-RAR outlines the principles and key features of rapid assessment and response. The advantages of an RAR over other approaches include: • • • timely results, by bridging the gap between assessment and response; creation of productive partnerships and alliances, by involving local communities; and initiation of action that can be sustained, by strengthening local capacity for assessment and response. Key features of an RAR

1.

The particular features of an RAR that distinguish this approach from other methods of working with men who have sex with men are listed below. • • • • • • Speed, timeliness and cost effectiveness. An RAR can usually be completed within 12 to 16 weeks. Practical relevance. An RAR’s purpose is to facilitate programme development, not to increase scientific knowledge for academic purposes. Strengthening local responses. An RAR fosters the involvement and participation of men who have sex with men. Use of multiple methods. This facilitates a comprehensive approach to gathering data to develop an understanding of social and health issues relevant to men who have sex with men and the situations in which they live. Inductive approach. This allows for building of ideas and conclusions from the data collected, not from pre-existing theory generated in other contexts. Multi-level analysis. This uses an analysis of individual, community and societal contexts and needs to understand the influence of each of these factors on the risk and vulnerability of men who have sex with men, and to develop responses that are relevant and able to be implemented. Reliability, validity and triangulation. This uses a number of methods and data sources to collect and analyse information.

•

13

Chapter 4 Organizing an RAR Contents • • • • • • Realistic goals Basic planning Anonymity and confidentiality Keeping to a schedule Where do you stand? Choice of staff

The key steps for planning and implementing the RAR can be found in Chapter 4 of the TGRAR. 1. Realistic goals

In many cases, an RAR is likely to be conducted in a setting where little or nothing is already known about men who have sex with men and HIV/AIDS. This makes careful planning important, since there may be little or no pre-existing information to help. Before describing the stages of the planning process, it is important to emphasize that an RAR is unlikely to be able to answer questions such as “Why does male-to-male sex occur?” or “How many men who have sex with men are there locally?” If faced with questions of this kind, remember that these issues are far beyond the scope of what you could be reasonably be expected to answer in a rapid assessment and response. Remember too that effective approaches to HIV/AIDS prevention and care among men who have sex with men have been developed in the absence of clear-cut answers to such questions. We do not need to know why some men pay for sex in order to make sex work safer; and we do not need to know how many heterosexual young people there are in a neighbourhood before we can initiate an effective education programme to meet their needs. Similarly, we do not need to know how many men who have sex with men there are in an area in order to develop a health intervention for them. Many men who have sex with men conceal their maleto-male sexual behaviour, which can make estimates of their numbers especially unfeasible and misleading. When planning an RAR, it is always important to keep realistic goals in mind. The emphasis should be on exploring the experiences, circumstances, lifestyles and needs of men who have sex with men locally as they relate to HIV/AIDS. This is exactly the kind of information needed to develop an effective response. 2. Aims The previous chapters offered guidance on thinking about some of the key issues in an RAR. When planning an RAR in more detail, it is a good idea to choose three or four key issues to focus on and to develop aims related to these. Typical aims could include: • to explore sexual decision-making processes among men who have sex with men in relation to the risk of HIV infection; 14 Basic planning

• • •

to investigate violence and harassment experienced by men who have sex with men and to consider this in relation to men’s vulnerability to HIV infection; to study beliefs about the body, sexuality and sexual health among men who have sex with men as they inform attitudes towards HIV and safer sex; and to explore the role of drug use, including alcohol use and injecting drug use among men who have sex with men and in their relations with other men who have sex with men as it relates to HIV transmission risk.

Writing down aims early on will help the RAR team develop a clearer idea of the scope and focus of the study. As the assessment is carried out, refer back to these aims to assess whether the RAR has stayed on track. Aims can be modified as the study progresses. Try to be flexible and allow ongoing findings to contribute to the development of the work. 2.1 Target population

It is important to be clear also about the target population to be investigated. This may be a particular population of men who have sex with men, defined perhaps in terms of age, socioeconomic background and/or race/ethnicity. Decisions about the population to be focused on may be influenced by the programme it is intended subsequently to develop. They may also be influenced by the men it is possible to access easily, and the resources available. 2.2 Geographical area

Do not to make assumptions about the places where men who have sex with men socialize. Remember too that men who have sex with men are part of the general population. Because of this, try not to restrict yourself to work in well known ‘cruising areas’ and meeting places, because this may limit your understanding of the diversity of male-to-male sexual life. In specifying a geographical area to investigate, it is a good idea to list some of the places to be focused on. However, it is important to be open to adapting this focus during the course of the work. More generally, however, it is a good idea to specify geographical scale. Thus, for example, if the RAR is being conducted in a city, it may be decided to focus on particular neighbourhood, or perhaps in various research sites around the city. It is a good to make an initial plan of the geographical area that will be covered in the RAR, so that everyone involved has a clear understanding of where the work will be carried out. 2.3 Time involved

Before beginning an RAR, think carefully about how long the work is to take. When will the assessment begin and end? This may depend on the budget available. Generally though, the initial assessment stages of an RAR should not last more than two months. 2.4 Budget

The budget will also affect the scope and focus of an RAR. It is a good idea to prepare a detailed budget for the resources available, indicating how these will be allocated to different aspects of the RAR process. Specific sums of money should be allotted for payment and/or expenses to those carrying out RAR fieldwork, data analysis, paper work and printing costs. 2.5 Methodology

Finally, think carefully about the data collection methods to be adopted. Generally, more than one method will be used to collect data. By combining different methods together, it is 15

possible to obtain a more complete picture. This issue will be returned to in Chapter 9 of this guide and is covered in more detail in Chapter 9 of the TG-RAR. 3. Maintaining confidentiality and assuring anonymity

When conducting an RAR, it is important to respect confidentiality, especially since the focus is on sensitive issues such as sexuality and sexual health. As already discussed, many men who have sex with men wish to conceal their sexual behaviour and orientation. In these circumstances maintaining confidentiality is particularly important. To maintain confidentiality, try not to write down the names of the men who have sex with men who have been talked to while conducting the assessment. Use false names instead or avoid using names at all. When writing up findings, take care not to include details that may identify a respondent. It can be a good idea to discuss issues of confidentiality openly with the men participating in the RAR. Ask them what level of confidentiality they feel comfortable with. Informed consent is a vital part of any assessment process. Informants and interviewees should know what is expected of them and to what use the data will be put. There are a variety of ways of obtaining informed consent. Some of these are described in Chapter 9.0 of the TGRAR. Do not take literature describing the subject of the RAR into areas where fieldwork is being carried out. This can make field researchers easily identifiable, which could be off-putting to potential respondents. Try to keep written materials such as questionnaires and field notes in a secure location if at all possible. RAR participants should be made aware that that they do not have to take part in any aspects of the research with which they feel uncomfortable. Explain that they are not required to answer any questions that they do not want to and that they may withdraw from the study at any time, without explanation. 4. So where do you stand?

Recognizing that men who have sex with men are often mistakenly viewed as a separate group in society rather than as part of the general population raises some important questions about where a team is ‘positioned’ in the RAR process. When it comes to work with men who have sex with men, perhaps the majority of social researchers, epidemiologists and other health experts will be (or will feel) somewhat removed from that which they propose to study and understand. They may have considerable research experience, but few links with men who have sex with men locally, and little familiarity with their needs and lifestyles. This will present some immediate practical difficulties in contacting respondents and getting an RAR started. More significantly, conducting an RAR from this standpoint will mean that the research team remains forever on the outside of the communities and networks of men who have sex with men whose needs it is hoped to assess. This can present major problems since one of the aims of a good RAR is to find out about HIV/AIDS issues from the point of view of potential service users. Thus, while an RAR team’s research methods may be well developed, its capacity to speak for men who have sex with men within a particular study area study may be less convincing. 16

A different scenario can arise when an RAR is being conducted by members of a community support or health-promotion organization run by and for men who have sex with men. Here, the team is likely to have easy access to a range of respondents and may be in a good position to develop an ‘insider’ perspective on the needs and social problems of men who have sex with men. This may give RAR findings a greater sense of authenticity. However, in comparison to a ‘professional’ research organization, the team may lack research expertise. As a result, any assessment undertaken may be seen as methodologically weaker and will perhaps not be so highly regarded in formal public health contexts. Overcoming issues of this kind can be challenging and frustrating. While these RAR guidelines cannot explain with certainty how to overcome barriers of this sort, they offer advice on how to go about work so as to have maximum impact on future policy and programme development. A community organization’s reasons for conducting an RAR may be quite different from those of a research group, perhaps being more concerned to assert the rights and needs of men who have sex with men. In this case, some of those who participate in the RAR may feel that formal policy environments are not the best places in which to advocate for their rights and needs. This may be particularly so in countries that have criminalized male-to-male sex. When planning the RAR it will be necessary to take account of both formal policy work and more ‘activist’ oriented approaches, and to decide where the RAR team situates itself in relation to these concerns. 6. Choice of staff

Recognizing the importance of ‘standpoint’ within the RAR process highlights the crucial difference between enquiry carried out by an organization with ‘insider’ access to networks of men who have sex with men, and that conducted by a group that is less familiar with the issues, and which may need the help of ‘gatekeepers’ and ‘sponsors’. This is relevant to the recruitment of RAR staff. If the RAR is being conducted by an organization working with and for men who have sex with men, it may be relatively easy to recruit staff who have some familiarity with the issues. They may themselves be men who have sex with men. This can be an asset in facilitating access to potential respondents. However, staff of specialized agencies working with men who have sex with often men often lack training in research skills. Research training will therefore need to be arranged during the early stages of planning and assessment. It may also be tempting for such staff to focus only on male-to-male sexual life in social contexts that are best known to them. It is a good idea to stress the importance of looking beyond ‘familiar’ contexts to explore male-to-male sexual relations in other settings. Alternatively, a research group or organization whose members are not men who have sex with men, may be conducting the RAR. Here, the individuals concerned may have good research skills but may lack sensitivity and understanding of issues relating to male-to-male sexuality. This can create difficulties in developing a sensitive ‘insider’ perspective and training on male-to-male sexuality should be arranged for such staff. Such training may not be enough to overcome any entrenched prejudicial attitudes towards men who have sex with men, but it should at least introduce awareness on these issues. An excellent compromise can be reached when social researchers work in partnership with community service providers. Here, honesty and openness are required concerning the strengths as well as the limitations that each party brings to the RAR process. It is a good idea 17

to discuss these issues early in the planning stage to help clarify different points of view within the RAR team and, if possible, to resolve any differences of perspective.

18

Chapter 5 Planning and implementing a response Contents • A health promotion response • Structural, community and individual levels of assessment and response • RAR, health policy and health promotion The TG-RAR identifies a number of key principles for intervention and programme development. Please look at these in Chapter 5 of the TG-RAR. 1. A health promotion response

Assessment by itself does little to promote health. Instead, responses are needed in the form of concrete programmes and interventions. Several different kinds of response are possible within the context of HIV/AIDS and men who have sex with men. They include actions at individual, community and structural levels. Definition: Response and Intervention A response or intervention is any action that is taken to help reduce a public health problem, help an affected population, change a particular setting, or change a risky behaviour. Responses include projects and activities that aim directly to help people change their behaviour (individual level); projects that aim to help communities and settings change (community level); and actions at a legal, political, economic, social, religious or cultural level that alter the environment in which the problem occurs (structural level). Interventions may take place in the areas of prevention, health promotion, treatment, policy responses and political activity.

Views differ on how best to deliver public health programmes and interventions. Programmes developed in one location often need to be modified for use elsewhere. For example, in some countries it may be appropriate to develop health promotion programmes that talk openly about the health needs of men who have sex with men. However, in other cultures it may be seen as inappropriate to talk publicly about such issues. Instead, it may be more suitable to develop a response that talks about promoting health among men in general, of which men who have sex with men are a part. A sensitive approach Sex between men is heavily stigmatized in many societies. In Botswana, the Men, Sex and AIDS Project set out to address the HIV/AIDS and sexual health needs of a wide range of men. Working from the capital city Gabarone, it first collected information about men’s interests and needs through conversations in bars, nightclubs and other informal settings. In the course of these conversations, it was possible to identify men’s fears and anxieties about issues such as HIV/AIDS in a non-threatening manner. While the initial focus of the work was not on men who have sex with men, the needs of such individuals rapidly became apparent. By approaching the issue indirectly, and by starting from men’s own experiences and needs, the project has been successful in reaching a wide range of individuals with health promotion messages, skills training and support.

2.

Structural, community and individual levels of assessment and response

The ‘individual’, the ‘community’ and the ‘structural’ signal three important levels at which a particular health issue, risk behaviour and/or a population group can be analysed. In order to fully understand an issue or problem, it is important to conduct an initial assessment at each of these levels. It is important too to take individual, community and structural considerations into account when analysing research findings and in using findings to inform a programmatic response. Figure 1 below illustrates these different levels of analysis. 19

Figure 1

Assessment at individual, community and structural levels Structural Social, economic, political, religious, cultural events

Community Local networks, settings, culture, norms

Individual Knowledge, attitudes, choices, preferences, behaviours

Using the model—a case study The following example of assessment and response focusing on condom use among men who have sex with men, comes from recent work in India. Structural, community and individual assessment and response on condom use among men who have sex with men Two local community organizations, Praajak and Integration, collaborated in conducting research on condom use among men who have sex with men in the Indian State of West Bengal. Few men locally used condoms when having anal sex with other men. Moreover, condoms available locally were too thin to offer protection during anal sex. They were also sometimes used with oil-based lubricants, which caused them to tear, posing a risk of HIV and STI transmission. Reasons for a low-level and incorrect use of condoms among men who have sex with men in West Bengal were assessed and responded to at structural, community and individual levels. Structural level assessment and response At the structural level, factors affecting condom use were related to a number of issues. One important issue was poor overall levels of information about the need to use strong condoms with water-based lubricant for anal sex. Condoms in India have been typically understood as a method of family planning. Additionally, the types of condoms available were not strong enough for anal sex in the absence of appropriate lubricant. Finally, water-based lubricants were not widely available and so could not be bought and used, even if health education recommended them. These issues were responded to with various structural level interventions. The community organizations that conducted the assessment began negotiations with a local condom manufacturer to explore the possibility of developing an affordable strong condom that could be sold as a profitable product by the condom manufacturer. Another company was approached to explore the possibility of manufacturing small sachets of water-based lubricant that could be easily carried in the pocket. Another option investigated was the possibility of importing sachets of water-based lubricant from Bangladesh, where a community-based organization had already developed a relationship with a manufacturer making sachets of lubricant. In addition to these responses, community organizations held discussions with the State Ministry of Health and Family Welfare in an attempt to counter legislation prohibiting the distribution of condoms in prisons. This response was linked to a broader national campaign to allow the legal distribution of condoms in prisons throughout India. Health promotion resources aimed at men who have sex with men were produced, with the intention of developing these over a number of years so that they remained up to date and relevant to current needs and issues. Community level assessment and response At the community level, low and incorrect use of condoms among men who have sex with men was related to a number of factors. First, because little information on condom use for anal sex was available locally, there was no culture of condom use for anal sex among men who have sex with men. It was also notable that men who have sex with men did not make up a ‘community’ as such. While some men did socialize together, many others did not. This meant they were less likely to be exposed to any health promotion interventions targeted at communities of men who have sex with men.

20

These community-level issues prompted various responses. The types of health-promotion resources described above were distributed among networks of men who have sex with men. The aim was to raise awareness on safe condom use and ways of practising this, given the difficulties in buying strong condoms and small sachets of water-based lubricant, along with many men’s perceived lack of choice in their sexual decision-making processes. In addition, outreach work in the places where some men who have with men socialized was used to promote safer sex. This involved outreach workers having informal conversations with them, including discussion on safer sex, but also on the many other issues in the men’s lives. This way, health promotion messages were integrated into the context of other needs. Because of awareness that many men who have sex with men do not belong to community networks of other men who have sex with men, health promotion activities were developed in specific locations where men meet, such as in prisons, remand homes, railway stations, parks, bars and work places. This work was aimed at reaching men in general, some of whom were likely to have sex with other men. Much of the work had to be conducted sensitively, because in some situations it was seen as unacceptable to talk directly about sex between men. However, the aim was to raise general awareness of the importance of condom use. Individual level assessment and response At the individual level, many men who have sex with men were found to have little personal interest in using condoms. This related in part to some of the structural and community-level issues discussed above. Given low levels of knowledge about condom use, and lack of community and peer pressure to use condoms, individual men who have sex with men were generally not in the habit of using them. Many men who have sex with men also talked about condoms reducing sexual pleasure, which was another personal disincentive. Ways of responding at the individual level included some of the community-level interventions already described. These included talking to people individually during the course of outreach work and other face-to-face activities. Often, individual level responses focus on issues in people’s lives other than direct sexual health concerns, since for many men who have sex with men, concerns such as selfesteem, prejudice and harassment are more important. However, talking to people about their personal and emotional concerns provides time and space to help develop their general sense of well-being.

3.

RAR, health policy and health promotion

In the initial stage of an RAR, it may be hard to know how to incorporate the findings of an assessment into health policy and health promotion activities. This is understandable. Indeed, part of the RAR process involves evaluating options and considering how research findings might inform response and/or intervention development—be it in terms of shaping health policy or applying a new health promotion strategy. As an RAR unfolds, it is likely that different ideas will be formulated about how to turn findings into policy strategies. Members of the RAR team may also establish connections with people who will be able to help develop the response and intervention component of the work. This is especially likely if an RAR is funded by or affiliated with a national or international health development agency or a national or local HIV/AIDS programme. Here, people supporting the RAR will have an interest in incorporating findings into ongoing work. It can also be helpful to develop relationships with others who may be supportive of your work. Seeking the backing of local health ministers, lawyers, doctors and journalists can be valuable. Such people may be useful as advocates, particularly in a potentially controversial field. Their involvement from the start can help to raise the profile of the issues being worked on. In some settings, it may be difficult to approach officials such as health ministers or senior health officials and talk to them about the focus of the work. This may be especially so in countries or regions where there is prejudice or denial about male-to-male sex, or indeed where sex between men is criminalized. In situations of this kind, it may be helpful for the work to be presented as focusing on HIV/AIDS and men in general, rather than male-to-male 21

sexual health in particular. Over time, it may be possible to become more explicit about the research and intervention focus. It will be up to the RAR team to strike an appropriate balance between being sensitive to prejudice concerning male-to-male sexuality, while also advocating for the sexual health needs of men who have sex with men. In developing a strategy for incorporating the findings of an assessment into a successful response, it may be helpful to refer back to the structural, community and individual level model of influencing factors outlined above. Think about each issue that needs to be examined in the RAR and its relevance at each of the three levels. This way, it may be possible to identify the policy and intervention issues that need to be focused upon. For example, in the case study of condom use among men who have sex with men in West Bengal it was possible to come up with a series of relatively distinct intervention strategies aimed at (i) condom manufacturers, (ii) policy-makers, (iii) community leaders and (iv) individual men who have sex with men. Planning a programme or intervention strategy in a similar way will help develop an appropriate response for each level.

22

Chapter 6 Social mobilization, advocacy and community participation Contents • • • 1. Social mobilization Community participation Advocacy Social mobilization

Social mobilization is key to a successful response. It involves encouraging people to take action on issues that affect their lives. When people are facilitated in taking responsibility for issues that affect their health, stronger and more sustainable health interventions will result. Social mobilization involves encouraging communities to recognize and act on their shared interests. However, in many circumstances it will not be adequate simply to encourage people to take responsibility for health issues such as HIV/AIDS without simultaneously addressing structural issues that affect vulnerability to HIV infection. This is especially pertinent when working with men who have sex with men, who, in many contexts, suffer from discrimination and social stigma. It will only be possible to bring about sustainable improvements in men’s lives by addressing the broader policies, practices and beliefs that affect life choices and affect the risks individuals face in relation to HIV/AIDS. 2. Community participation

Community participation is especially relevant to work on HIV/AIDS among men who have sex with men. Successful work to promote this population’s human rights and sexual health has often been initiated at community level. This has most commonly been through the pioneering efforts of community-based organizations of men who have sex with men. In conducting an RAR, it is important to recognize the value this kind of work and to either involve local groups of men who have sex with men in the RAR, or to initiate such organizational activity as a part of the RAR process. It is important to recognize, however, that community participation is not a straightforward process. Community groups may disagree with one another, especially about issues such as the rights of men who have sex with men. In some countries, it is common to find some men advocating a ‘gay’ rights agenda, while others speak of rights in terms derived from ‘local’ categories relevant to men who have sex with men. Moreover, many, perhaps the majority of men who have sex with men, will not participate in or identify with these types of political debates. With this in mind, it is important to remember that community groups may not speak for all the men who have sex with men in the RAR study area. Mindful of these concerns it is, however, a good idea to incorporate communities of men who have sex with men into the RAR process. 2.1 Setting up a community discussion group

A good way to learn about some of the problems and needs of men who have sex with men is to set up a community discussion group as an initial step in an RAR. Not only will this give greater credibility to subsequent work, it will also facilitate learning about key issues. A group 23

can be quite informal. It might involve inviting some men who have sex with men, who represent a social cross section locally, to talk about the RAR. How difficult it is to establish a community discussion will depend on many factors, including the local legal context, levels of stigma and discrimination, your skills as a researcher and programme planner, and the personal contacts members of the RAR team may already have with men who have sex with men. 2.2 Setting up a community advisory board

As planning progresses and the research strategy becomes more formalized, it can be a good idea to set up a community advisory board consisting of men who have sex from the study area. The role of such a board is to inform the ongoing RAR process, perhaps advising on how the research is perceived by men who have sex with men from within the town, village or region where the study is being carried out. This can be very helpful, both with respect to the legitimacy of the work and the quality of data obtained. An advisory board can be established via a similar process to that used to create a community discussion group. An advisory board is, however, likely to have a more active involvement in an RAR and may take a more formal role as part of the RAR process. This will probably include regular board meetings throughout the duration of the RAR. Difficulties linked to different perspectives on the RAR can, however, arise. For example, an advisory group may feel that it ‘owns’ the research and may want to direct it in ways contrary to the RAR team’s own goals. The RAR team will also need to think carefully about how much members are prepared to work collaboratively. This may be especially significant if work is being conducted by people who are not themselves men who have sex with men. The team will need to think carefully about including men who have sex with men as active participants in the work, rather than just as ‘subjects’. The RAR team may need to actively relinquishing some control over the RAR process. While possibly disconcerting at first, this can have benefits with respect to unanticipated outcomes and by allowing the genuine participation of men who have sex with men, something that may be empowering for the individuals involved. 3. Advocacy

Community participation alone will not be enough to bring about improvements to the HIV/AIDS-related risks experienced by men who have sex with men. Broader structural concerns must be addressed as well. Structural factors include such things as: • • • • • prejudicial legislation or policy on male-to-male sex; iniquitous policing practices towards men who have sex with men; social stigma concerning male-to-male sex; lack of lubricants suitable for men to use with condoms during anal sex; and ignorance about and negative attitudes to male-to-male sexual health issues in hospitals and health care centres.

Structural issues of this kind are brought about and reinforced by social, economic and cultural circumstances that are beyond the immediate control of individuals. As such it would be unrealistic to expect men who have sex with men to be able to address these issues through community participation alone. The involvement of a broader constituency of people within the RAR will be required in order to adequately tackle such concerns. Canvassing supportive partnerships of this kind is known as advocacy. 24

Partnerships may be formed with professionals and influential people such as: • • • • • • • • lawyers; politicians; police officials ; community leaders; counsellors and therapists; people who work in the media ; condom and lubricant manufacturers; and doctors and other health staff.

By involving such people, it may be possible to bring about broader changes in attitudes towards men who have sex with men. It may also be possible to develop policies and guidelines to promote sensitivity to issues relating to the vulnerabilities men who have men face generally, and specifically regarding HIV/AIDS. In many countries, concepts of human rights may have little salience, not only in relation to men who have sex with, but also people in general. Ideas emphasizing difference between people may have more cultural prominence than notions of equality. It can be a good idea to explore such beliefs in the RAR, as a way of understanding broader attitudes affecting social prejudice and considering how the needs of men who have sex with men may be best advocated for in such contexts. 3.1 Devising an advocacy strategy

In developing an advocacy strategy, it is a good idea to: ∗ Begin with realistic goals. There may be many issues to be addressed, but it will be impossible to do everything at once. In the first instance, select two or three key advocacy goals, such as working with local police officials or community leaders. Advocacy with such people early in the RAR can make the research process easier, perhaps helping to reduce any harassment RAR staff might otherwise experience. ∗ Be selective. Given the stigma and/or prejudice associated with male-to-male sex, it is a good idea to be cautious about the people with whom you conduct advocacy. Try to select those you think will be sensitive to the needs of men who have sex with men. ∗ Target the information. Not all the professionals will need the same information. Doctors, for example, will require specific material relating to male-to-male sexuality and sexual health. Lawyers will want literature that focuses on legislative and policy concerns. Community leaders will perhaps need information about the general experiences of men who have sex with men in the study area. Develop different advocacy messages for each group. ∗ Be sensitive. Professionals and influential people may sometimes hold prejudiced attitudes towards men who have sex with men, or they may even have limited awareness that such men exist. This can be frustrating, especially when actively campaigning for the rights of men who have sex with men. Try to raise awareness gradually rather than challenging people too much about their ignorance. This can help people of influence understand and eventually support arguments concerning the rights and needs of men who have sex with men. While an advocacy strategy cannot be expected to overcome all prejudicial attitudes concerning male-to-male sex and 25

HIV/AIDS, it can go a long way towards creating a supportive environment for an effective response.

26

Chapter 7 Training Contents • • Types of training Training exercises

Once a team of RAR staff and/or volunteers has been assembled, it will necessary to conduct training on the issues and methods involved in conducting the RAR. The training required will depend on the research skills and awareness in the RAR team. Training may be required in some or all of the following areas: • • • • • • • • knowledge of HIV/AIDS and sexual health; information on the life experiences and needs of men who have sex with men; legal issues associated with male-to-male sex in the study area; skills in interviewing, observation, focus group facilitation, surveying and mapping; skills in conducting research discreetly and sensitively in public settings; data recording and analysis skills; skills in managing sexual advances received during the course of fieldwork; and awareness of issues relating to confidentiality and anonymity.

Training is usually carried out over three to five days, but the precise time scale will depend on the needs of the people being trained and the number of issues to be covered. 1. Types of training

Types of training useful for an RAR include: Team training Team training involves the whole RAR team and provides background information on the RAR, as well as discussion on HIV/AIDS and men who have sex with men. It can help decide the scope and focus of the RAR. Team training should also be used to sensitize staff and volunteers to the data collection methods to be used. Short session training Here, training focuses on a specific issue (e.g. RAR aims and objectives, lifestyles and needs of men who have sex with men, information on HIV/AIDS in the study area). Short sessions can be held in addition to team training, or may be conducted as part of training or discussion with the RAR team and other relevant professionals and influential people such as doctors, lawyers and community leaders. Training of this kind can be an effective advocacy strategy. Capacity-building This aims to equip RAR coordinators with the skills necessary to train the RAR team, plan all the stages of the research and subsequent response and to generally manage the whole RAR process. This type of training is best facilitated by people who are already experienced in conducting RARs or similar pieces of work. 2. Training exercises

Different exercises can be used in conducting RAR training.

27

Discussion groups These can focus on general issues such as the lifestyles and needs of men who have sex with men, knowledge of HIV/AIDS and attitudes about the RAR. Discussion groups offer a good way of assessing the values and opinions of the group of people being trained. Role playing These can be used to enact scenarios that may be encountered while carrying out field work, such as conducting interviews in public places, managing sexual advances from interviewees or dealing with harassment from police. Group-work exercises These can include activities such as true or false games on knowledge of HIV/AIDS and safer sex. Participants are provided with a series of statements about HIV/AIDS and safer sex that have to be placed in ‘true,’ ‘false,’ or ‘don’t know’ categories. This can be a good way of exploring participants’ levels of knowledge on the issues and imparting correct information in a relaxed way Practice interviews These can involve one trainee using a provisional interview guide to interview another trainee who plays the role of someone being interviewed in the field. This is a good way to test how questions may or may not be understood by respondents, and for exploring the best way to record data while keeping the flow of the interview going. It is a good idea if other participants act as observers, watching practice interviews and commenting on interview technique once the interview has been completed Observation practice This can involve trainees going into the areas where they will be conducting fieldwork, watching what is going on and reporting back to the main group. This is a good way to learn about the variety of activities that may be observed and how to take field notes that contain sufficient information. For more detail on how to use these exercises and a suggested training agenda, refer to TGRAR Chapter 7.

28

Chapter 8 Wording the questions—the Rapid Assessment Modules Contents • • • • • • • Initial consultation Developing a profile of the study area Assessing contexts of HIV/AIDS prevention and care Assessing HIV/AIDS risks Assessing health issues Assessing HIV-related risk behaviours Assessing existing interventions

The assessment modules outlined here provide guidelines on how to explore the range of issues to be covered in an RAR. Early elements of assessment should aim to bring together existing data and information on both men who have sex with men and HIV/AIDS locally, where such information is available. In many countries, relatively little information about men who have sex with men may exist. Where no such information is available, it may be helpful to look at what has been learned about HIV/AIDS and men who have sex with men elsewhere. If no good quality information exists on HIV/AIDS as they affect men who have sex with men locally, information on HIV/AIDS should be collected as it affects the population more generally. This will allow the RAR team to contextualize their study of men who have sex with men within a more broad understanding of HIV/AIDS. Early in the RAR process, it will be important to elicit opinions from various people within the study area—not only influential people such as community leaders, politicians or doctors, but also other community members. This will help provide a useful background on attitudes towards and beliefs about men who have sex with men and HIV/AIDS. Such perspectives and opinions provide contextual information against which to plan more specific research activities. Finally, it will also be important to conduct initial assessment work with men who have sex with men themselves. Some RAR teams may not have much pre-existing contact with men who openly talk about having sex with other men, and so this may seem like a difficult task. Here, this initial stage in the RAR can be used to begin trying to make contact with men who have sex with men locally. Strategies for initiating contact are outlined later. Making contact early makes it easier to do more in-depth research later. If the RAR is being conducted by a team that already has good connections with men who have sex with men in the study area, or that primarily consists of men who have sex with men, contacting respondents may not be such a problem. However, try not to neglect broader issues concerning HIV/AIDS, and do not forget to enquire into more general views about male-tomale sexuality. This will provide important background information against which to contextualize other knowledge. 1. Initial consultation

An initial consultation should aim to be brief, perhaps taking place over the course of a week. The aim is to develop an overview of the situation by bringing together information from different sources. This can include published data as well as information elicited in preliminary interviews. Sources to be accessed include: 29

• • • • • • • • •

community organizations for men who have sex with men; people in social contexts where men who have sex with men are known to socialize; community based organizations working on HIV/AIDS and sexual health; hospitals, community health clinics and local health practitioners (such as healers or purveyors of non-allopathic medicines); government departments; counsellors (both professional and informal); researchers who have done work on HIV/AIDS and/or men who have sex with men; political and policy – relevant organizations; and lawyers.

Key questions to guide the planning of the initial consultation include: • • • • • • • • • • • What knowledge exists about HIV/AIDS and men who have sex with men? What beliefs and attitudes about HIV/AIDS are there among men who have sex with men? How do settings and behaviours influence risk of HIV/AIDS? How does drug use, including alcohol and injecting drug use influence risk of HIV/AIDS? What awareness is there of risk and vulnerability to HIV/AIDS among men who have sex with men? What individuals, organizations and populations need to be included in the subsequent RAR process? What should be the scope and focus of the RAR? Should the focus be on a particular group of men who have sex with men? If so which group and why? Or should attention instead be focused on a more general sample of men who have sex with men? What resources can be accessed to undertake the work? What methodological and practical considerations need to be taken into account? How will men who have sex with men be involved? How will the RAR findings be communicated and discussed?

See Chapter 8.1 in TG-RAR for more information on Initial Consultation. 2. Developing a profile of the study area

A study area profile should be developed early on as a way to document some of the main structural features of the study area. An area profile can include general information on geographical and environmental characteristics, political and government structures, the economic context, general infrastructure, social groups and community organizations, and the different types of health services and practitioners in the locality. Not all this information will seem directly relevant to HIV/AIDS. However, it will provide important contextual information as well as indicate key structural issues that might be explored more explicitly later on. Assessment of political, community, work and health infrastructures may also signal potential sites of collaboration in developing the RAR response. Key questions to be answered in developing a study area profile include: • • • What is the local population structure and what are the characteristics? What are the languages spoken? This may include knowledge of any specific code or colloquial speech used by men who have sex with men. What are the dominant religious and other beliefs? This should include reference to any specific influence such beliefs may have on attitudes towards male-to-male sex. 30

• • • • • • • • •

What are the key geographical and environmental characteristics? What is the prevalence of HIV/AIDS, main causes of morbidity and mortality, rates of STIs? What are the principal characteristics of the national and local economy? Are there economic factors that compel men to move into and out of the study area? What are the main channels of communication and access to the media? What are the main means of transport? What is national/local expenditure on health, and does this vary for the study area? Are there health or social welfare provisions especially for men who have sex with men? If so, what is the social, economic and ethnic profile of men accessing these? What are the main social problems faced by men who have sex with men? What form does national legislation on male-to-male sexuality and HIV/AIDS take? What are the levels of knowledge and enforcement of these laws in the study area?

See Chapter 8.2 in TG-RAR for more information on the Study Area Profile, including assessment grids that can be used. 3. Assessing contexts of HIV/AIDS prevention and care

Contextual assessment aims to builds on the study area profile by examining specific issues in more detail. The key to this component of the assessment lies in making links between national factors, what is happening locally, and how this affects the lives and circumstances of men who have sex with men. There are three key areas in any contextual assessment. Factors affecting the vulnerability of men who have sex with men to HIV/AIDS: • • • • • • • What economic factors (economic growth, restructuring, economic migration and mobility of men, urbanization, income differentials) affect the vulnerability of men who have sex with men to HIV/AIDS? What economic factors are of relevance to the lives of men who sell sex to other men? What political, policy or legal issues affect men who have sex with men’s vulnerability to HIV/AIDS? How do policing practices and other means of social regulation impact on men who have sex with men, and their vulnerability to HIV/AIDS? How do roles, status and power between men who have sex with affect vulnerability to HIV/AIDS? How do roles, status and power in relationships between men and women affect the vulnerability of men who have sex with men and their female sexual partners to HIV/AIDS? How do important cultural beliefs (religious, social, cultural, political) influence attitudes to both men who have sex with men and HIV/AIDS?

Factors influencing health and social issues relevant to men who have sex with men: • • • • • What, if any, systems of health care exist in the study area? What are the general treatment-seeking behaviours of people, especially in terms of STIs (including allopathic, ‘local’ and ‘alternative’ treatment options)? What are the attitudes of health-care providers towards male-to-male sexual health? Do laws inhibit the provision of heath care addressing male-to-male sexual health? Is there political representation for and/or by men who have sex with men, nationally or locally? 31

Factors influencing the development of effective programmes and interventions: • • • • • Is there political commitment to tackling HIV/AIDS at national and local levels? Is there political commitment to addressing the health needs and human rights of men who have sex with men at national and local level? What are the attitudes of local people of influence towards work with men who have sex with men? How explicit can an intervention with men who have sex with men be? Are there local health services with which an intervention for men who have sex with men could collaborate? Are there specific policies or practices locally and nationally that might impede or facilitate the development of an intervention aimed at men who have sex with men?

See Chapter 8.3 in the TG-RAR for more detail on contextual assessment, methods and data sources, and assessment grids. 4. Assessing HIV/AIDS risk

It is a good idea not to make prior assumptions about what populations or settings present the most risk when starting the RAR. Use the RAR as an opportunity to explore the experiences and potential vulnerabilities of all men who have sex with men. As we have discussed, many men who have sex with men are not readily visible in society. A good RAR will emphasize the HIV/AIDS risks faced by such men, as well as the risks experienced by more obviously vulnerable populations. Two sets of questions are of special relevance to risk: those focusing on the concealment of male-to-male sexual orientation, and those relating to the specific circumstances of men who sell or exchange sex. The types of questions suggested here can be adapted for use in other circumstances. Concealing male-to-male sexuality • • • • • • What are the social circumstances in which men can be open about male-to-male sex? What are the circumstances in which male-to-male sex is either concealed or not seen as relevant? What are the emotional and/or psychological consequences for men who do speak about their male-to-male sexual desire and orientation? What are the potential advantages to men of not being open about their male-to-male sexuality in all areas of their lies? Do men locally comprehend a concept of explicit identity based on male-to-male sex (such as ‘gay’ or ‘homosexual’)? Or is this seen as an unfamiliar or irrelevant idea? What might be the potential issues involved in advocating for the rights and needs of men who have sex with men in context(s) where perhaps the majority of such men do not consider having sex with other men a significant attribute of social identity?

Selling or ‘exchanging’ male-to-male sex • • How does male-to-male sex for payment (or other reward) take place? For example, dos it take place through bars, brothels, nightclubs or organized street work? Does it also take place more individualistically and/opportunistically? If male-to-male sex work is organized, by whom is it organized and how are men who sell sex recruited into such systems? 32

• • • • •

What are the range of payments? Do men exchange sex for rewards other than money (e.g. food, drink, clothes, etc.)? To what types of men do men sell sex? What are the age ranges, social, economic and ethnic backgrounds of men who buy sex? What sense of choice do men who sell sex experience in relation to their work? Do some men feel that they have no choice other than to sell sex? Do some men see selling sex as something they choose to do? How do men who sell sex understand HIV/AIDS and safer sex? Does sex work put men at any special risk of HIV infection? If so, how and why?

See Chapter 8.4 in the TG-RAR for more detail on population and settings assessment, methods and data sources, and for assessment grids. 5. Assessing health issues

The aim of the Health Issues Assessment module is to help the RAR team assess the extent of health issues as they relate to HIV/AIDS and men who have sex with men. This includes issues that relate directly to HIV/AIDS as well as more general factors such as beliefs and social attitudes. It will often not be feasible to develop comprehensive estimates of the extent and nature of HIV infection among men who have sex with men locally. This is because there is little or no relevant information. In other cases, data will be more readily available. Where information is available, it should be utilized in the RAR. Sources could include: ∗ community organizations working with men who have sex with men; ∗ national or local health information systems; and ∗ international health and development organizations. Always remember that much existing information is likely to pertain to the most identifiable networks of men who have sex with men, rather than the more extensive population of less visible men. Because of this it will be wise not to generalize. An RAR can also be helpful in exploring attitudes and beliefs about HIV/AIDS. It can also shed light on other health issues such as emotional and psychological well-being. Key questions that may be asked in a health assessment include: • • • • • What are men who have sex with men’s sources of information about HIV/AIDS? What are the different beliefs about and attitudes towards HIV/AIDS among men who have sex with men in the study area? What are men’s treatment-seeking behaviours for STIs? What emotional or psychological issues do men who have sex with men report, particularly with regard to stigma, shame or prejudice experienced? How do emotional and psychological issues affect sexual behaviour and potential risk of HIV infection?

See Chapter 8.5 in the TG-RAR for more detail on health issue assessment, methods and data sources, and assessment grids.

33

6.

Assessing HIV-related risk behaviours

A health risk behaviour assessment aims to examine the extent and nature of HIV-related risk behaviour among men who have sex with men. In particular, the focus is on factors that potentially enhance the risks faced by men who have sex with men, and the factors that may make men more resilient to risk. Risk factors can be understood at structural, community and individual levels. Individual issues include feelings towards sexual partners as they influence perceptions of risk. Community issues include local norms and beliefs about specific sexual practices such as anal sex and condom use. Structural issues include economic factors that influence choices in sex work, or which affect men’s patterns of economic migration. Key questions to ask include: • • • • • • • • What different sexual behaviours are there among men who have sex with men? What is the prevalence of different behaviours and what behaviours present the highest risk for HIV infection? At what age do men start having sex with other men? Who, generally, are men’s first sexual partners and in what circumstances do first sexual experiences come about? Do men speak of their first sexual experiences as being by choice, or through coercion? What are the different contexts and settings in which sex between men occurs? What economic factors influence male-to-male sex? What economic constraints or choices are of relevance to men who sell sex? What are men who have sex with men’s perceptions of risk and safety in their lives generally? Do men experience violence or harassment? If so in what circumstances? How does risk relate to men’s understanding of HIV infection and safer sex? Do men who have sex with men use condoms and water-based lubricant for anal sex? What are the structural issues that might affect condom and lubricant use? Do drug and alcohol use affect the sexual behaviour of men who have sex with men?

See Chapter 8.6 in the TG-RAR for more detail on health and behaviour assessment, methods and data sources, and assessment grids. 7. Assessing existing interventions

An RAR will often be conducted in places where there have been few prevention and care responses aimed specifically at men who have sex with men. Where such strategies do exist, however, the RAR should take account of them. It can also be important to assess interventions aimed at other populations such as male migrant workers, young men in general or sex workers. Key questions to consider in undertaking the intervention assessment include: • • • • Are there existing interventions with men who have sex with men? Are these specifically targeted projects, or interventions working with other populations who have developed contacts with men who have sex with men in the course of their work? How relevant and accessible are local HIV/AIDS services to men who have sex with men? What understanding of the lifestyles and needs of men who have sex with men is there among staff working on local projects? Are there other services that might potentially support an intervention with men who have sex with men (e.g. counselling projects, sexual health clinics and community groups)? What are the attitudes towards men who have sex with men within such services?

34

Many of these questions can be explored by bringing together existing evidence from projects and interventions in the study area, as well as by conducting interviews with key informants. Ask men who have sex with men participating in the RAR about the types of services they access, and those they do not. Ask them also what kinds of programmes and interventions they would like to see as part of a well-founded RAR response. See Chapter 8.8 in the TG-RAR for more detail on intervention assessment, methods and data sources and assessment grids and summary grids.

35

Chapter 9 Finding out about men who have sex with men: Methods Modules Contents • • • • • • • Sampling Interviews Focus groups Observation Questionnaires and surveys Mapping Data analysis methods

The previous chapter discussed the types of issues that might be explored in an RAR assessment. Here, we will describe some of the methods that can be used, and how they can be specially adapted for work on HIV/AIDS and men who have sex with men. These methods and how to use them are described in detail in Chapter 9 of the TG-RAR. 1. Sampling

Sampling is the process of selecting the people who will participate in the research. There are several methods of sampling. These methods and techniques are described in TG-RAR Chapter 9.3. In the case of an RAR on HIV/AIDS and men who have sex with men, and because of difficulties of easy access, the most suitable sampling strategies are likely to be opportunistic sampling and snowball sampling. Opportunistic sampling Opportunistic sampling involves accessing members of a target population using whatever opportunities arise. It can often begin with meeting key individuals and accessing other informants through them. Opportunistic samples can be helpful in confirming that a particular behaviour exists, or is frequent in its occurrence. As such, it may be especially useful in countries or communities where there is scepticism or denial that sex between men occurs. Opportunistic sampling can be employed in such a situation to show that there are indeed men who have sex with men in a particular area, neighbourhood or nation. A potential drawback to this approach is that it usually accesses the most readily identifiable men who have sex with men. Other men, whose lifestyles and behaviours may make them less visible, may be missed out. This may have the effect of limiting understanding of the diversity of male-to-male sexual lifestyles and experience. Snowball sampling Sometimes known as ‘network sampling’ or ‘chain-referral sampling’, this approach is particularly suited to investigating populations that do not have an obvious presence locally. Snowball sampling involves working with intermediaries, who introduce the RAR team to respondents. This is particularly appropriate when working with people who may be vulnerable and/or highly stigmatized. 36

In work with men who have sex with men, snowball sampling will most likely involve: • Contacting a known individual with links to men who have sex with men in the study area. Such a person may be a man who has sex with men himself. He may perhaps be one of the respondents originally contacted using opportunistic sampling, or a member of the RAR team. This individual subsequently introduces other men who have sex with men to the team. These men are interviewed or invited to attend a focus group discussion. In turn, these men introduce other men to the research team. This continues until either no further sample members can be contacted, or sufficient contacts have been made.

• • •

One disadvantage of snowball sampling is that it is unknown what the samples represent. It may also be difficult to locate suitable intermediaries for certain populations. For example, working with a fairly young population of men who have sex with men may make it difficult to contact older respondents. A snowball sampling strategy does not necessarily aim to achieve a representative sample of the entire target population. The main aim is to explore key issues in the target population’s lives, based upon as diverse a selection of respondents as it was possible to contact. Other sampling strategies Time and place sampling, for example, involves conducting research in different research localities at different times of day, in order to identify differences in the social character of environments. This can be particularly suited to observation in some of the areas where men who have sex with men socialize. Block sampling involves deciding to conduct research in selected sites only. In research with men who have sex with men, block sampling might involve deciding to work in a specific ‘cruising area’, where it is known men who have sex with men meet as well as perhaps in a few more ‘general places’ such as street corners, bus stations or bars where men meet up. 2. Interviews

An interview involves talking, listening to and learning from people. Interviews can be used to explore people’s understandings of health problems and health behaviours, together with related attitudes and opinions. Interviews provide respondents with the opportunity to talk in detail about their lifestyles and needs. Chapter 9.4 of the TG-RAR guide gives guidance on issues such as who should be interviewed, and how to organize, prepare and conduct interviews. We suggest that you consult this section as a way of helping you to develop your own interview strategies and guidelines. One-to-one interviews can provide the opportunity for respondents to talk in confidence about personal issues. Group interviews, on the other hand, can provide an opportunity for respondents to share different attitudes and opinions. They can also enable the identification of areas of disagreement and ‘tension’ between respondents’ perspectives and accounts. Several different types of interview are possible. 37

Unstructured interviews involve the interviewer asking respondents to offer their views and opinions freely. An unstructured interview is similar to an informal conversation. The interviewer may have some questions in mind beforehand, but the order of questioning will be determined by the flow of ideas between the interviewer and respondent. Unstructured interviews require the interviewer to be skilled in listening carefully and asking questions that keep the conversation natural in style, but also relevant to the themes and issues under exploration. To the inexperienced eye, the information gained from unstructured interviews can sometimes appear confusing and chaotic, and this may mean that researchers find it difficult to analyse the data obtained. However, with practice, the RAR team should become more confident and skilled in using unstructured interviews, especially as interviewers become more familiar with the issues being focused upon. Semi-structured interviews often follow a similar conversational style to the unstructured interview, but here the interviewer uses a pre-structured list of questions to guide the flow of discussion. The aim, however, is not to follow the list of questions rigidly but rather to use the list as guide to the important topics to be focussed on. This ensures that important issues are covered but also allows for a flexible response, particularly in relation to matters that were not anticipated in the question guidelines. Structured interviews use question guidelines more strictly, with the same questions being asked of all respondents, in the same order. The strength of this type of interview is that the common format makes data easier to analyse and compare. Additionally, an inexperienced interviewer may find a structured interview schedule relatively easy to follow. On the other hand, some interviewees may find a structured interview a rather formal and artificial experience. This may be a major disadvantage, particularly when sensitive issues are being discussed. An inflexible format may also prevent the collection of unexpected but relevant information that arises during the course of an interview session. Interview topics In advance of any type of interview, it is necessary to decide on the particular topics that will be focused upon. Within the context of an RAR, these may include: • • Local terms for sexuality and/or sexual identity. phrases that may be used to talk about sexual behaviour and sexual roles. For example, particular terms may be used to describe the role taken in anal sex (activo and passivo, for example in some Latin American countries). Alternatively, there may be terms used locally to describe men who cross-dress and/or act in a feminine manner. The relationship between ways of talking about sex and actual sexual behaviour. For example, do men who are designated as ‘passive’ always take the receptive role during anal sex, or is the relationship between talk about sex and actual sexual life more complicated than this? Do the places that men meet other men for sex dictate what activities occur? Understandings of sexual health, including local terms used to describe STIs. Treatment-seeking behaviours in relation to sexual health, including the use of allopathic and other remedies. Availability and affordability of condoms, particularly condoms suitable for anal sex. Patterns of condom use. Availability and affordability of water-based lubricants for use with condoms. Paid or compensated sex among men who have sex with men. Harassment, violence and prejudice. 38

•

• • • • • • • •

• • •

Social mobility and migration for work. Issues of relevance to men who have sex with men who also have sex with women. Drug and alcohol use, and its relationship to sexual behaviour.

This is just a list of suggestions. There may be other subjects to be explored. A community advisory board can help to identify these. Community advisors may also be used to pre-test interview questions. That way, feedback can be obtained on whether questions are clearly worded and easy to understand. 2.2 Keeping a record

When conducting interviews, it is important to keep a record of what is said. One way is to use a tape recorder. This can be useful if a word-for-word transcript is required. However, there are a number of issues that must be considered if using this method. First, when interviewing stigmatized people such as men who have sex with men, many interviewees may not want what they say recorded. Also, the public contexts in which many interviews take place may not be appropriate for making tape recordings. Finally, the costs and time involved in making a detailed transcription of interviews must also be taken into account. Another approach is to take detailed interview notes. These can be prepared during the course of an interview itself, although doing so may disrupt the flow of the interview and make the person being interviewed feel uncomfortable, especially if he is talking about sensitive topics such as sex and sexuality. An alternative is to write interview notes after an interview has occurred. It is best to prepare those notes as soon as the interview is over, as this will make it easier to remember the information elicited. 3. Focus groups

A focus group consists of a number of individuals who are interviewed collectively because they have had a shared experience, come from a similar background, or have a particular skill or knowledge. Focus groups are useful for identifying and exploring people’s beliefs, attitudes and behaviours, and providing ideas for further investigation. Chapter 9.5 of the TG-RAR guide provides detailed information on the use of focus groups, including how to set up and manage them, the number of people who should be included and the skills required by facilitators. A range of topics can be explored in focus group work. The topics listed in the interview guidelines above may help structure the choice of issues to be explored. They may be supplemented by questions to elicit shared and different perspectives on situations and events. Examples of these include: • • • • • • • What are the main social problems faced by men who have sex with men? What are family and societal attitudes towards male-to-male sex? What were group participants’ first male-to-male sexual feelings and experiences? Have group participants encountered prejudice, violence or harassment in relation to their sexuality? If so can they talk about this? What social and economic factors influence involvement in male sex work? What are the main sexual health concerns of men who have sex with men? Can group participants talk about experiences of STIs, including treatment-seeking behaviour? What do group participants believe and know about HIV/AIDS? 39

Setting up a focus group A focus group should normally involve a group of people with shared experiences and interests. This will provide the basis for group commonality. How many focus groups are undertaken will largely depend on the number and diversity of men who have sex with men with whom the RAR team is in touch. When setting up a focus group, remember than it will probably be best if only men who have sex with men participate. This will offer a more secure environment of like-minded people within which to express opinions and share experiences. 4. Observation

In the context of an RAR, observation can be valuable in allowing researchers to better understand the lives and circumstances of men who have sex with men. Observation can also enable the RAR team to examine some of the tensions and contradictions between what people say, and what they do. This is not to suggest that observation is used to catch out people who are lying about their behaviours. Rather, it is to emphasize that for everyone there may be tensions and contradictions between the way they talk about our lives and the way they act. Observation is especially good at exploring this, especially in relation to relatively hidden behaviours such as male-to-male sex. Chapter 9.6 of the TG-RAR guide offers a more detailed description of different styles of observation. Observation and sexual participation It is important to stress that the goal of observational research is not to observe sexual activities between men who have sex with men. Research of this kind is ethically problematic, illegal in many countries, and potentially exploitative of those being observed. Having said this, at least some members of the RAR team will be men who have sex with men themselves. Some may well have sex with other men during the time frame of the RAR, and others may even be part of the social scene of men who have sex with men in the study area. This can present some ethical dilemmas. By far the majority of health and development agencies will not look favourably upon researchers’ sexual participation in the work they are carrying out. Because of this, it is essential to adopt clear guidelines stipulating that researchers will not engage in sexual activities with subjects while they are working. During RAR training, it may also be a good idea to rehearse scenarios in which members of the team practise dealing with sexual advances received while conducting fieldwork, including during observation. Acting as a sexually unavailable researcher sometimes, and as a potentially sexually available participant in the same social spaces on different occasions, can convey mixed messages. Managing this type of situation is one of the challenges of observational research since, being a naturalistic method, it blurs the boundaries between what can be regarded as research and what is life experience. It is important to discuss these kinds of issues with other members of the RAR team, and to reach shared agreement on a way forward. Observation within a broader context It is important to extend the range of places in which observation is carried out beyond the male-to-male social and sexual spaces that may be most familiar. Members of the RAR team will need to keep an open mind in identifying the different sorts of settings in which men who have sex with men may meet. Government offices, department stores, shopping arcades, 40

cinemas, buses and street corners are just some of the everyday locations in which men meet other men for sex. Few areas are taboo in this respect, although initial encounters in any of these places may not be obviously or immediately sexual. Examining the connections between male-to-male sexual activities in these settings and other social practices is a key aspect of observation. This way, it may be possible to develop a better understanding of how men who have sex with men remain both visible (to one another) and invisible (to others) within a particular context. Recording observations Observation data is perhaps best recorded in the form of notes written up after observations have been made. Writing things down during the course of observation itself is likely to interfere with the naturalistic flow of the work. It may also not be wise for RAR team members to carry written notes on male-to-male sexual behaviour in the public environments where they will conduct observation work, as the potential loss of these in such an environment could threaten confidentiality. 5. Surveys and questionnaires

Conducting surveys can be a useful way of gathering information from a large sample of individuals. The subjects chosen to explore could include the types of issues outlined above. Thus, for example it might be decided to carry out a survey focusing on condom use among men who have sex with men. However, by asking about other related issues, a well designed survey could allow an exploration of the relationship between different social variables (e.g. age, economic status and ethnicity) and condom use within a particular setting. Chapter 9.7 of the TG-RAR guide offers detailed guidance on designing and organizing surveys. Whether or not surveys are used as part of an RAR will probably depend on the number of men who have sex with men it is possible to access. In contexts where sex between men is illegal and/or where men who have sex with men are heavily stigmatized, it may be difficult to recruit an adequate sample size. However, in places where well-established groups of men who have sex with men exist (e.g. in the form of formal or informal support groups and community organizations) conducting a survey may be easier. It may also be possible to carry out a survey of health-care workers knowledge of and attitudes towards men who have sex with men, as part of a larger contextual investigation. Designing a questionnaire In order to conduct a survey, a questionnaire or interview schedule will need to be developed. Questionnaires usually work best when they are limited in length and have a clear focus. Begin by identifying two or three specific, related topics, and work from these to identify related questions. For example, a questionnaire on condom use may explore men’s decisionmaking practices about condoms, the perceived availability of condoms in the study area, and knowledge about the use of lubricants. Itemizing sub topics Once the overall focus of a survey has been decided upon, identify a number of sub-topics within the overall subject area. Taking the example of condom use, these might include: • • How frequently do respondents use condoms in sex with other men? What sexual activities do men use condoms for, with other men? 41

• • • • • •

How do men decide whether or not to use condoms? Do respondents usually carry condoms with them? What is the local availability and affordability of condoms? Do men who have sex with men also use condoms in sexual activities with women? What factors influence whether men use condoms with female sexual partners? Do respondents know about appropriate lubricant use?

Wording the questions Once specific sub-topics have been identified, it will be necessary to decide how to word the questions asked. Wording should be specific and clear. Avoid using technical terms wherever possible and avoid leading questions. For example, it is often better to ask a fairly nondirective question such as, “Can you tell me about an experience of violence and harassment you have suffered in relation to your sexuality?” rather than a more directive and challenging question such as, “Why do men who have sex with men experience violence and harassment?” The latter question presumes that the informant agrees that men who have sex with men suffer from violence and harassment. It also asks the respondent to justify their opinion. The former question, on the other hand, gives the respondent room to talk about whether they have experienced violence. Open questions of this latter kind can be supplemented with more direct questions as the interview proceeds. It is also a good idea to ask sensitive questions in a non-personal way. For example it will often be better to ask, “Where do people go to get treatment for sexually transmitted infections in your area?” rather than, “Have you ever had a sexually transmitted infection?” The latter question asks for directly personal information and may make the interviewee feel defensive. The former question offers scope for the interviewee to talk about sexually transmitted infection without revealing any personal information unless they wish to. Every question should be simple and clear, and should address only one topic. It is often best to ask a series of short questions that suggest specific answers rather than using long sentences. For example, rather than asking a general question such as, “What do you think about the availability of condoms in your area?” it may be better to ask several separate questions such as, “Are condoms available in your area?” “Where are they available from?” “What is the price of condoms locally?” and “Can you afford to buy condoms?” Short, precise questions of this kind invite concise responses that can easily be recorded and analysed. Implementing a survey Once a questionnaire has been designed and tested, the RAR team will be in a position to conduct a survey. In carrying out a survey among men who have sex with men it may not be a good idea to take the questionnaire sheet itself into public areas. Paper questionnaires can be quite conspicuous and some men may not be willing to be questioned in such a public fashion. If possible, it will probably be better to conduct surveys with men more privately. With issues of privacy in mind, it is advisable that questionnaires do not ask for personal information of respondents such as name and address. Many respondents find such questions threatening and will be unwilling to answer. It is a good idea to reassure respondents that all information written will remain confidential and will not be identified with them in any way. It can sometimes be a good idea to use self-completion questionnaires in which respondents fill in a questionnaire themselves. This can be an effective means of conveying to respondents 42

that questionnaire data will remain confidential. However, this method presumes literacy, and not all respondents will be able to read. It is important to be sensitive to this. Questionnaires may be used in other aspects of the RAR research. A survey of health-care professionals could elicit their knowledge and opinions about male-to-male sexuality. Here, issues relating to confidentiality and sensitivity will be different since respondents are less likely to be asked to reveal information about their personal feelings and sexual experiences. 6. Mapping

Another research method is mapping. Mapping uses graphics such as maps, drawings and pictures to illustrate aspects of the environment in which people live. Maps can be drawn of spaces, networks, bodies, and the distribution of events or activities. Geographical or spatial mapping The map included below was drawn by a focus group to illustrate some of the popular places where men who have sex with men socialize and meet sexual partners in their town. The lines drawn between different places were used to indicate that men often moved between places during the course of an evening, sometimes meeting other men along the way. G.P S tation S tuart

P ark

Old T own Bus Station

Lakeside Plaza Com plex T rading bank

Central P ark

New Bus S tation

T . D. Cinema S outhern P ark New c inem a River

A geographical or spatial map of the places in which men who have sex with men spend their time is likely to reveal many everyday locations. Identifying these is important in illustrating that men who have sex with are not a discrete, socially separate, population that only spends time in certain locations (e.g. ‘cruising areas’). Including mapping in an RAR can also be a helpful way of challenging assumptions that men who have sex with men are set apart from the rest of society, and therefore have completely different HIV/AIDS-related concerns. Body mapping Body maps illustrate part or all of the human body. They may used to represent knowledge about health, illness and the body. In order to create a body map, a group can be asked to draw a picture of a whole body, or a body part. These illustrations can then be used as the basis for discussion. Body mapping can be used to help elicit perceptions of HIV by asking people to 43

draw what they imagine happens to the body of someone infected with HIV. This might include pictures representing any beliefs about outwardly visible signs of HIV infection or depicting what is imagined to happen inside the body. Another use of body maps lies in exploring feelings about sexuality and the body. Participants can be asked to indicate on a picture of the human body various potentially erotic sites. This can be particularly useful in illustrating the diversity of sexual acts possible between men who have sex with men. Because of the HIV risks associated with unprotected anal sex, there has been a much emphasis on the idea that men who have sex with men mostly engage in anal sex, often with men being designated as habitually penetrative or penetrated partners. While it would be wrong to suggest that men do not have anal sex with other men, this is not always the case, and indeed many men may never have anal sex with other men. A body map is a good way of exploring men’s ideas about other sexual possibilities and as such can be used to supply information useful to safer sex interventions. Chapter 9.8 of the TG-RAR provides more detail on different mapping methods. 7. Data analysis

Once the assessment component of an RAR has been completed, a range of information will have been collected. This information may be referred to as your findings or data. Among the different data sources will be interview transcripts, survey findings, notes from observation, charts and diagrams drawn in focus groups, and perhaps other types of data as well. All these findings will have something in common, since they are all likely to relate to issues relevant to HIV/AIDS and men who have sex with men. However, particular parts of the data will probably relate specifically to the different questions asked in the course of the assessment. Trying to make sense of all this information may seem confusing at first. Analysing all the data Once the initial assessment has been completed, there may be pressure to write up the findings and present them in a report as soon as possible. Pressure of this kind may be particularly acute when the work is being funded by an agency that expects the results of the RAR to be delivered as soon as possible. After all, an RAR analysis is supposed to be rapid and, as we have discussed, the information produced is intended to be good enough for developing programmes and interventions rather than for writing complex social science reports. However, it is a mistake to rush data analysis. Rapid research methods involve striking a balance between working quickly and working in sufficient detail to ensure the results are of genuine value and relevance (Nelson & Wright, 1995). In order to get the most out of the data, begin your analysis by thinking about all the different issues explored. Organizing a group discussion among members of the RAR team can be a good way to start. These discussions can be quite informal, offering an opportunity for everyone to describe their impressions of the research and what they learned from it. A discussion of this kind can be a good starting point for data analysis, facilitating reflection on the RAR findings from different points of view. This will help decide on some of the key issues that should be focused on later. In order to take analysis forward, it will be necessary to link together the different types of research data in a more systematic manner. This process of seeing how findings fit with one another is known as triangulation.

44

Triangulation Triangulation involves looking at the same data set from different points of view so as to identify patterns and trends, as well as differences and inconsistencies. For example, if one of the main themes explored is issues relating to condom use and anal sex, then keeping these issues in mind, data can be triangulated by looking at interview transcripts, field notes, focus group findings and other sources of information to see where (and in what ways) issues relating to condom use and anal intercourse arise. Triangulation also allows the different methods used in an RAR to support one another. For example, it allows findings from surveys of condom use for anal sex among men who have sex with men to be contextualized against other sources of information—e.g. in-depth perspectives elicited in individual interviews and findings from focus group discussions. In turn, findings from all of these sources may be complemented by data generated during periods of observation. By bringing different sources of information together, it will be possible to develop a multi-dimensional perspective on condom use. Organizing and coding data In order to make sense of findings, categorize the data under different headings. Ideally, this should be done in relation to the different subject areas or ‘headings’ that made up the subthemes within the RAR. These headings could be taken from the original aims that were set when work started, and/or from issues that have arisen during the RAR itself. It is a good idea to make a list of these headings in the form of a chart or a spreadsheet. Data can then be organized under these different categories. On the other axis of the spreadsheet, indicate the source of the information by listing the different methods used in the RAR. In this way, findings on each subject can be easily triangulated across the different data sets produced. Findings► Methods▼ Interviews Focus groups Observation Surveys Mapping Violence Condom use HIV Sexuality and gender Sex work STIs First sexual experiences

If a computer is being used to store and analyse findings, it may be possible to sort data into different computer files. Various software programmes can help with this. However, here we will assume that findings are being analysed by hand, probably using data and information that has been written down on paper. Findings can be organized by looking through the data and deciding which information belongs under which heading. In some cases, it may be found that one piece of data fits neatly under one heading. For example, a focus group discussion on awareness of HIV/AIDS can perhaps be categorized under the heading ‘HIV/AIDS knowledge’. However, in other situations the team may find that a data set contains information that is relevant to more than one heading. For example, an interview transcript or questionnaire may contain information about attitudes towards people with HIV/AIDS, sex work, violence and sexuality, etc. Here, it will be difficult to store all this information under only one heading. This is where coding comes in. Coding is simply a technique used in research to mark different parts of a data set according to the subject areas and themes that they are relevant to. Sometimes coding may be built in, in advance. For example, a structured questionnaire may include coding categories to 45

enable the data to be categorized as it is being collected. On the other hand, a tape recorded, semi-structured interview may elicit data that can only be coded afterwards. Most coding methods involve marking findings so that it is possible to pick out the parts that are relevant to each specific subject area straight away. Write a symbol (perhaps a letter, number or some other code) next to different parts of a piece of data to indicate the subject area that each section refers to. Thus, for example, to show that an entry in participant observation field notes is relevant to the subject of ‘violence towards men who have sex with men’ the text might be marked with the letter ‘V’ to represent this issue. Similarly responses in a questionnaire may be marked with the letters ‘Mrg’ to indicate the topic of marriage or ‘Asx’ to represent the subject of anal sex. The actual symbols used for coding (be they letters, numbers or some other form of code) should be chosen and agreed to by the RAR team. Individual, community and structural level analysis By categorizing information collected under different subject headings, trends, variations and main features can be identified. In order to take the analysis one step further, however, think about each issue according to individual, community and structural levels. To help do this, draw up a table, listing different research issues on one axis and individual, community and structural levels on the other. Such a table might look something like this: Findings► Level▼ Individual Community Structural Violence Condom use HIV Sexuality and gender Sex work STIs First sexual experiences

By analysing issues at individual, community and structural levels, it will be possible to identify how the different HIV/AIDS-related experiences and needs of men who have sex with men fit together and can be addressed at different levels of intervention. This is valuable for planning intervention and programme activities. Presenting findings and initiating planning Throughout data analysis, it is important to think carefully about how findings will be fed back to participants and funders. The needs of these two groups may differ considerably. While the former may be concerned to know that their views were treated with respect or may have specific advocacy objectives, the latter may have other needs in mind. They may, for example, wish to be reassured that findings from the RAR assessment have genuine value for the development of future programmes and interventions. It is often not possible to prepare one report that will satisfy these two groups simultaneously. This can present ethical and practical difficulties. Members of the RAR team may find themselves wondering for whom they conducted the work. They may feel loyalty to respondents and community advisory board and may want to present their findings in a way that reflects their priorities. But this may not please funders seeking a more dispassionate and seemingly objective stance. One way of addressing this concern is to prepare two versions of the RAR assessment report; the first for the agency that commissioned the work, the second written so as to be accessible to the majority of men who have sex with men. Alternatively, a research feedback session may be held with study participants—an important approach to adopt when levels of literacy are low. Disseminating findings in this way can be rewarding but takes time. The balance 46

between different types of dissemination must therefore be decided upon in relation to the resources available. When producing an assessment report, not everyone involved will be happy with what the report says. Some men who have sex with men may disagree with the way findings have been presented, perhaps in relation to issues such as sexuality, condom use, marriage and other concerns. It will be important to think carefully about how their views can be taken into account. This can be one of the most challenging aspects of data analysis. Hard decisions may have to be made about how data is presented. Remember your standpoint and, wherever possible, reference what you have to say back to this. At the same time, it is important that your reports reflect the actual data. Resist the temptation to simply fit findings into preconceived ideas about how programme development should proceed. RAR findings often generate challenging and difficult questions that disrupt established plans for project development. However this is precisely why an assessment was carried out; because we do not know all the answers already.

47

Chapter 10 Contents • • • • • • •

Developing an action plan

Bringing key findings together Developing structural, community- and individual-level response Setting priorities Specifying intervention aims and objectives Identifying key activities Assessing resources, costs and timescales Developing final intervention action grids

Having analysed assessment data, it is important to discuss the findings with the stakeholders involved. These include men who have sex with men in the study area, the community advisory board and representatives from organizations such as development agencies, health ministries and organizations working on HIV/AIDS. Feeding back findings in this way is an important part of the process of deciding how concerns identified in an RAR assessment can be addressed through different types of programme or intervention. Developing consensus around an action plan will help take this process forward. However always remember the following points. • • • • Developing an action plan should be carried out jointly with men who have sex with men locally, the community advisory team and other stakeholders. Consultation and discussion is best facilitated by face-to-face meetings between all the parties involved. Given the prejudice and shame associated with male-to-male sex, special effort may be needed to protect anonymity and confidentiality in these meetings. One person, normally the RAR coordinator, should take responsibility for facilitating and coordinating the development of the action plan. How long it takes to develop an action plan will depend on the amount of information collected, the problems addressed and the complexity of the programme or intervention planned.

In planning a response, it is important to focus on the most significant findings from the assessment. When developing an action plan, think about the following questions. • • • • • Which particular issues relating to HIV/AIDS and men who have sex with will be focused on in the programme or intervention? How exactly is it proposed to implement the programme or intervention? What are the strengths of the proposed programme or intervention? What resources are there locally to maximize programme/intervention success? What factors might inhibit the success of the programme/intervention?

Using these questions as a basis for discussion will help everyone involved think about the issues concerned in planning a response or intervention. It is a good idea to write down answers to these questions as a way of recording different people’s views. Throughout the action plan development process, different opinions are likely to be expressed. There may, moreover, be competing demands as to what the programme or intervention should focus on, and how it should be carried out. In this respect, the expectations of men who have sex with men may differ considerably from those of funding organizations. These views may, in turn, be at variance with those of local health ministries and/or HIV/AIDS 48

organizations. Sometimes, it can be difficult to address all these different points of view in a sensitive manner, and it is unlikely that an action plan can meet the expectations of everyone involved. Group discussion allows opinions to be shared and, where possible, compromises reached. 1. Bringing key findings together

In developing an action plan, is important to gather together key findings from the RAR assessment. Having followed the data analysis guidelines presented in this guide, the data should have already been brought together under different headings. These will identify key areas to focus on in the action plan. The next step is to develop several action grids. These are tables that help organize data and show how RAR findings relate to needs identified and the responses proposed. Action grid 1 is the first grid. It is used to: • • • • identify key assessment findings; list the information source for each key finding; assess the validity of the finding(s); and indicate the response or intervention proposed.

Here is an example of an action grid that could be drawn up on the issue of violence and harassment affecting men who have sex with men. Action grid 1 Key finding High levels of violence and harassment towards men who have sex with men. Information source Interviews with men who have sex with men. Observations and experiences of RAR fieldworkers. Focus groups with men who have sex with men. Validity In interviews, many men made reference to accounts of violence and harassment. Fieldwork staff have witnessed and/or experienced harassment. Stigmatization and/or criminalization of male-to-male sex in the study area reflects and supports a social climate of violence and harassment toward men who have sex with men. General response Develop a community-based response that offers space for men to talk about experiences of harassment and violence. Outreach work in areas where men who have sex with men socialize. Individual and group counselling to help men develop strategies to assert and protect themselves. Advocacy work with local police to raise awareness of the needs of men who have sex with men. Advocacy work with lawyers and local politicians on the rights of men who have sex with men.

2.

Developing a structural, community and individual level response

After describing the relationship between findings and possible responses in Action grid 1, it is helpful to consider just which responses might be feasible, and how these might address different needs. Follow the individual, community and structural level model used throughout this manual. Draw up a new table indicating individual, community and structural level categories on one axis, and list the possible interventions proposed on the other. For each of these possible interventions, identify more specific intervention proposals that may be directed at individual, community and broader structural issues. At this stage, proposals need not be too detailed. Rather, aim to identify some of the strongest intervention possibilities.

49

To illustrate, let us take the example of how the different general responses proposed to address violence and harassment in Action grid 1 may be further developed as individual, community and structural level interventions. Individual Develop a community-based response that offers space for men to talk about experiences of harassment and violence. Outreach work in areas where men who have sex with men socialize. Ask men who have sex with men for suggestions on strategies that would help them to address violence and harassment. Community Hold a discussion group with men who have sex with men on the issue. Structural Aim to secure a private location in which men can meet to hold discussions.

Consider some of the personal issues involved in work of this kind, such as risk to outreach workers. Conduct one-to-one counselling and support sessions with men who have sex with men.

Recruit and train community workers and volunteers to undertake outreach work. Discuss potential collaboration with other organizations working on men’s health. Implement training in counselling skills for project staff. Conduct awareness-raising training on the experiences and needs of men who have sex with men for counsellors and medical staff in the intervention area. Assess the possibility of conducting awareness-raising discussion groups with local police on the needs of men who have sex with men.

Aim to secure the support of local police (where appropriate) so that they do not harass outreach workers.

Individual and group counselling to help men develop strategies to assert and protect themselves.

Create awareness-raising campaigns on the needs of men who have sex with men aimed at counsellors and medical staff.

Advocacy work with local police to raise awareness of the needs of men who have sex with men.

Contact police officials and assess their attitudes towards men who have sex with men. If possible, develop collaboration with police who are sensitive to the needs of men who have sex with men. Contact lawyers and politicians in the intervention area and assess attitudes towards men who have sex with men. If possible, develop collaborations.

Assess the possibilities of addressing issues such as police harassment of men who have sex with men, perhaps in discussion with local politicians or partner health and development agencies.

Advocacy work with lawyers and local politicians on the rights of men who have sex with men.

Conduct training in legal and human rights issues for project staff and volunteers. Conduct awareness-raising training with lawyers and politicians on the human rights and legal issues affecting men who have sex with men.

If appropriate, conduct advocacy work on the human rights and legal issues affecting men who have sex with men. This could include lobbying politicians or the media. (The appropriateness of this work will need to be judged according to circumstances.)

3.

Identifying which responses to develop

Once the different aspects of the main proposed interventions have been identified and discussed using the model described above, it will be necessary to choose three or four of the most important proposed responses to develop into more detailed intervention plans. Everyone involved in developing the action plan should be included in discussing what should be the main areas for intervention. It can sometimes be difficult to decide which responses to develop. Try to be pragmatic. Some proposed responses may be highly desirable but quite impractical, perhaps because funding is 50

not available or because the intervention may be too controversial at an early stage of programme development. Different people involved in developing the action plan may prioritize different interventions. For example, programme planners may emphasize interventions more explicitly related to the prevention of infection, while community groups of men who have sex with men may prioritize social support. In deciding which proposed responses to develop, compromises will need to be made. Once decisions have been made, outline some of the key strategic issues for each response. Using Action grid 2, list the three or four responses to be developed along one axis. On the other axis, place the words ‘priority’, ‘relevance’, ‘feasibility’ and ‘acceptability’. Action grid 2 Priority Response 1 Response 2 Response 3 Response 4 Relevance Feasibility Acceptability

Ask the team to discuss the following questions • • • • Priority. How important is the response? How urgently is it needed? Why is it important to act now rather than later? Relevance. Is the proposed response appropriate? Could another response be more effective? If so, what could it be? Feasibility. Are there any obvious obstacles to the development of the response? Are there any obvious factors that might facilitate the response? Acceptability. How acceptable is the response to the target population, stakeholders, decision-makers and the public?

List the team’s responses to each question in the action grid.

51

Priority Response Developing outreach work aimed at men who have sex with men. The work is important as a way of conducting health promotion activities where men who have sex with men meet, socialize and/or have sex. Outreach work is a priority, since this is the best way to contact men who have sex with men. It is important to implement this work as soon as possible because men who have sex with men can be especially vulnerable to HIV transmission and have been neglected by many HIV/AIDS prevention programmes

Relevance Outreach work of this kind has been successful in work with men who have sex with men in many countries

Feasibility Obstacles to the development of outreach work may include violence and harassment of outreach workers. Outreach work can draw attention to the social and sexual activities of men who have sex with men in public places. This can create unwanted attention Consider the possibility of men who have sex with men conducting outreach work themselves. This may make outreach work more acceptable to the target population.

Acceptability Outreach work may be acceptable to many men provided it is conducted sensitively, and with respect to men’s desire for privacy. Because outreach work with men who have sex with men must be conducted discretely, respecting the privacy of men who have sex with men and without drawing public attention, it can be difficult to demonstrate success to stakeholders and decision-makers.

Once the interventions to be developed and key issues relating to the priority, relevance, feasibility and acceptability have been decided upon, it is time to plan each response in more detail. This will mean specifying: • • • • • • 4. aims and objectives; key activities (for each aim); costs and other resources (for each activity); time frames (for each activity); responsible persons or agencies (for each activity); and indicators (for measuring each activity and objective). Specifying intervention aims and objectives

Aims and objectives communicate what the intervention is seeking to achieve. They make it easier to plan an activity and to evaluate the effectiveness of a programme or intervention. For each proposed response, identify at least one aim and several related objectives, e.g. Aim: To promote access to sexual health services and greater HIV/AIDS awareness among men who have sex with men. Objective 1: To develop a sexual health drop in service for men who have sex with men. Objective 2: To raise awareness of the HIV/AIDS and sexual health issues of men who have sex with men among staff of local clinics and sexual health services. Objective 3: To raise awareness of STIs and HIV/AIDS among men who have sex with men and to promote appropriate treatment-seeking. 5. Identifying key activities

Once aims and objectives have been identified, specify activities that will help achieve these. Discuss different ideas within the RAR team. Initial ideas do not have to be too specific. Allow people to offer their views and to discuss the potential of different ideas. In the case of the example above, the types of activities that might be identified include: 52

• • • 6.

Activity 1: To develop a sexual health promotion and HIV awareness-raising drop in service for men who have sex with men. Activity 2: To conduct training for medical workers on the HIV and sexual health issues experienced by men who have sex with men. Activity 3: To develop sexual health promotion and HIV awareness-raising health promotion materials aimed at men who have sex with men. Assessing resources, costs and time-scale(s)

Having decided on aims, objectives and activities, make an assessment of the resources, costs and time scale of each intervention. Try asking the following questions: ƒ What resources will be needed?

We can think of resources in different ways. First, there are human resources. These include the people who will work on the RAR intervention. Workers may include outreach staff; people who can conduct advocacy work; health workers; lawyers and others. Counsellors may also be required, together with specialists in training on sexuality and sexual health. It may even be valuable to recruit a doctor, nurse or lawyer to help with specific medical and legal issues. Then there are infrastructural resources. These may include rooms or offices used as a base for the intervention or a location in which to hold discussion groups, health promotion workshops and other community-based activities. Other infrastructural resources include computers, transport, television, video and other kinds of equipment. Finally, there are informational resources. These include leaflets and other media such as videos or discussion group guidelines. It may be valuable to arrange access to publishing services in order to be able to produce information of this kind. ƒ What costs might be incurred?

The team conducting the programme or intervention will need to work out a budget for each resource. For example, decisions will have to be made about how much to pay outreach workers, or how much to spend on producing leaflets or buying condoms. When developing an action plan, try to put a cost next to each intervention component. Actual costs will depend on the overall budget available, and this will most likely need to be negotiated with the organization funding the programme or intervention. ƒ Time frame

Describe in detail each component of the programme or intervention to specify how much time will be spent on each activity. For example, three to four months may be needed to design, pilot and produce a range of health promotion leaflets in consultation with men who have sex with men. The next six to eight months may then be spent distributing the leaflets. The time decided upon for each activity will probably need to be negotiated with everybody involved in the RAR. Decisions will once more depend on the budget available. 7. Developing intervention action grids

The aims, objectives, key activities, resources, costs and time scale of each of the main interventions or programme components should now have been specified. As a final step, it is a good idea to draw up a set of action grids that will illustrate each of these different 53

components for each intervention within the overall response. This will help everyone involved in the RAR see how the different aims, objectives and resources fit together. Action grid 3 below outlines the components involved in developing an intervention designed to promote access to sexual health promotion and HIV/AIDS awareness among men who have sex with men.

54

Action grid 3 Aim: To promote access to sexual health promotion and HIV awareness-raising services for men who have sex with men Objectives To develop a sexual health drop-in service for men who have sex with men. Activities Find appropriate premises. Establish collaboration with medical staff. Advertise the drop-in among men who have sex with men. Costs and resources Cost of premises. Payment for medical staff. Cost of medicines. Cost of advertising materials. Time frame Begin looking for premises immediately. Begin developing relationships with suitable medical staff immediately. Open drop-in centre within two months of intervention. Begin developing training materials in two weeks. Begin training workshops in one month. Responsible person or agency The organization carrying out the intervention. Medical staff working in the drop-in. Indicator Number of men who have sex with men coming to the drop-in service. Feedback on the drop-in centre from attending men.

To raise awareness of the HIV and sexual health issues of men who have sex with men among staff at local clinics and sexual health services. To raise awareness of sexually transmitted infections and HIV among men who have sex with men and to promote appropriate treatmentseeking.

Develop training workshops on the sexual health needs of men who have sex with men aimed at medical staff Conduct training with medical local staff. To conduct informal awareness-raising discussions during outreach work. To hold sexual health discussion groups. To develop and distribute sexual health promotion material specifically aimed at men who have sex with men.

Costs of training materials. Fees for trainers. Costs for renting training premises.

The organization carrying out the intervention.

Number of training sessions conducted. Feedback on training sessions from medical staff.

Costs of producing health promotion materials. Expenses and salaries for outreach workers and discussion group facilitators.

Begin outreach work immediately. Begin discussion groups within two weeks. Begin designing health promotion materials within two weeks. Produce heath promotion materials within two months.

The organization carrying out the intervention. The company designing and producing health promotion materials.

Number of outreach sessions conducted each week. Number of discussion groups held each month. Number of health promotion materials distributed. Feedback from men being worked with.

Once Action grid 3 has been completed for each intervention or programme element, the action plan will be completed. Interventions can now be put into practice. Throughout implementation, the action plan should be referred back to periodically to assess how work is proceeding. Remember though that as a programme of work is implemented, things may not always proceed according to plan. This is only to be expected. It is impossible to anticipate every issue that might arise before initiating a response. Use the action plan as a guide, but try to be adaptable and flexible. For example, during the course of outreach work the RAR team may find that it is possible to contact men who have sex with men in places that were not anticipated when preparing the action plan. In this case, it would be entirely appropriate to adapt the action plan to include outreach work in new locations. 55

Flexibility is especially important when conducting work on HIV/AIDS and men who have sex with men because in many settings the RAR team may be conducting the first intervention of this kind. The team will have to learn which intervention strategies are successful and which are less successful as they go along. Such a process of trial and error or, put more positively, learning by success, is inevitable. A good response should be adaptable and flexible, while aiming to remain faithful to the overall aims and objectives of the RAR. Chapter 10 of the TG-RAR provides more detail on how to develop action plans.

56

Chapter 11

Evaluation

Contents: • Monitoring and evaluating programmes and interventions • Monitoring and evaluating implementation • Evaluating outcomes and impact 1. Monitoring and evaluating programmes and interventions

Continuously appraising the implementation of programmes and interventions is an essential aspect of any RAR. Most usually, this occurs through regular monitoring to keep track of a project’s progress and assess whether it is meeting its objectives, or whether objectives should be altered in the light of RAR findings. Evaluation on the other hand involves determining the value or worth of a specific programme or intervention. In an RAR with men who have sex with men, the work conducted is likely to be especially innovative. Evaluation is particularly important here, as a way of assessing what has been achieved. Monitoring is commonly used to: • record what has been done—project inputs, activities and expenditure; • check progress towards achieving objectives and assess impact; • identify problems encountered and how these have been addressed; and • identify barriers and constraints to implementation. Monitoring is useful for tracking activities of programmes and intervention activities and identifying trends in data. It supports evaluation by alerting to changes that may have occurred, but it is not able to produce causal evidence about the effectiveness of an intervention. Evaluation is used primarily to determine: • whether a programme or intervention has been properly targeted; • whether it is working in the way it was hoped to; • the extent to which a programme or intervention is effective; • programme or intervention costs; and • whether there were any unexpected problems or benefits. There are three levels or phases of evaluation (UNAIDS, 2000): i. process evaluation – assessment of the programme’s content, scope or coverage as well as the quality of programme implementation (e.g. number of groups conducted, number of condoms distributed, amount of money spent on activities; percentage of men who have sex with men using health-care services, number of men who have sex with men trained as peer educators, increased knowledge of HIV/AIDS). ii. outcome evaluation – assessment of the positive and negative effects of a programme or intervention on desired changes (e.g. increased awareness of safer sex among men who have sex with men). iii. impact evaluation – assessment of the longer-term changes in HIV infection that are attributable to a specific programme (e.g. reduction of HIV transmission rates among men who have sex with men in a country). Note however that this type of evaluation is beyond the scope of what can reasonably be achieved in an RAR, since long-term data acquired over a period of years will be needed to make such an assessment. Moreover, 57

because in many contexts men who have sex with men are not socially apparent, it may be especially difficult to evaluate the impact changes associated with male-tomale sex on the local epidemiology of HIV/AIDS. See TG-RAR Chapter 11 for more information on evaluation. Another very useful resource to refer to is A Guide to Monitoring and Evaluation: National AIDS Programmes (UNAIDS, 2000). In designing a programme or intervention, the RAR team will need to develop procedures for monitoring and evaluating the work to be undertaken. Indeed, action plans should be used to generate the monitoring and evaluation approaches adopted, in particular the kinds of indicators and measures that will be used to collect data. In doing so, it is important that those implementing the action plan understand why it is necessary to collect data for monitoring and evaluation purposes, and are clear about what specific data is to be collected. It is also important to be clear at the start about the type of monitoring and evaluation that will be undertaken, and who will assume these processes. 2. Monitoring and evaluating implementation

Process evaluation requires that data is collected on how, and to whom, the programme or intervention is being delivered. This form of evaluation is commonly used by projects to assess their progress and measure their achievements. Areas of evaluation Coverage

Questions to ask • • Who was reached? How many men who have sex with men were reached?

Data required Number of men who have sex with men contacted Proportion of men, by age, gender and sexual identities (if men speak in such terms), and also by geographical area in which they live Staffing and training Service procedures and activities Measures of service activity Number and characteristics of men who have sex with men using the service Client satisfaction surveys Number of materials and supplies distributed Project budgets Staff numbers Cost of training Materials produced or purchased and used Administration costs

Activities and service delivery

• • • •

How did men who have sex with men access the service? What services were provided? How were the services organized? Did the service prove acceptable to men who have sex with men?

Resources used

What resources were used to implement the intervention?

3.

Evaluating outcomes and impact

Outcome and impact evaluation, on the other hand, measure whether a programme or intervention produced the desired changes in the population of men who have sex with men. They might ask questions such as “Did a health promotion outreach strategy result in an increase in men’s knowledge and/or relevant changes in behaviour?”, “Did the establishment 58

of a service for men who have sex with men in an STI clinic lead to an increase in the number of men talking about male-to-male sexual health issues in the clinic?”, and “Did a peer education programme increase men who have sex with men’s skills and knowledge to protect themselves against HIV infection?” Many evaluations report relationships between specific interventions and particular outcomes. These may be associations such as, increased attendance at STI clinics and fewer STIs recorded, increased rates of HIV testing due to improved facilities, and, changes in injecting behaviour due to the effects of an outreach programme. Determining the causal links between the intervention and changes to behaviour is often difficult, or may take some time. If there are changes in behaviour that can be attributed to a specific programme or intervention, then it may be useful to look at HIV prevalence to get an indication of whether efforts locally are making a difference to this data as well. However, as discussed, this can be especially difficult in the case of men who have sex with men, as in most contexts there will be no data specifically relevant to this population. Where such data does exist, it most often focuses on the most easily identifiable groups of men. See TG-RAR Chapter 11 for further details on specific evaluation methodologies and research designs to assess the outcome and impact of interventions.

59

References Aggleton P, ed. (1996). Bisexualities and AIDS, London, Routledge. Aggleton P, ed. (1998). Men Who Sell Sex, London, Routledge. Boyce P (2001). Rapid ethnographic assessment of male-to-male sexuality and sexual health in Kathmandu, Nepal. Report produced for Family Health International (FHI). Choi K-H et al. (2002). High HIV Risk but Inadequate Prevention Services for Men in China Who Have Sex with Men: An Ethnographic Study. AIDS and Behaviour, Vol. 6 (3)255-266. Cleaver F (2002). Institutions, Agency and the Limitations of Participatory Approaches to Development’ In: Cooke B and Kothari U, eds. Participation: The New Tyranny, London, Zed Books. Connell RW (1995). Masculinities, Cambridge, Polity Press. Donham, DL (1998). Freeing South Africa: The “Modernization” of Male–Male Sexuality in Soweto, Cultural Anthropology 13(1):3-21. Joint United Nations Programme on HIV/AIDS and the World Health Organization (2002). AIDS epidemic update. Geneva, UNAIDS/WHO. Joint United Nations Programme on HIV/AIDS (2000). AIDS and men who have sex with men, Technical update, Geneva, UNAIDS. Kelly JA et al. (2002). HIV risk characteristics and prevention needs in a community sample of bisexual men in ST. Petersburg, Russia. AIDS Care, Feb 14 (1):63-76. McKenna N (1996). On The Margins – men who have sex with men and HIV in the developing world, London, Panos Institute. Nelson N, Wright S (1995). Power and Participatory Development: theory and practice London, Intermediate Technology Publications. Niang CA et al (2002). Meeting the Sexual Health Needs of Men who have Sex with Men in Senegal. Washington, DC, Horizons Project. (http://www.popcouncil.org/pdfs/horizons/msmsenegal.pdf). Oostvogels R, Menon S (unpublished). Men who have sex with men: assessment of the situation in Madras. Report produced for the Government of Tamil Nadu. Parker RG (1996). Bisexuality and HIV/AIDS in Brazil. In: Aggleton P, ed. Bisexualities and AIDS. London, Taylor and Francis. Parker R, Khan S, Aggleton P (1998). Conspicuous by their absence? Men who have sex with men in developing countries: implications for HIV prevention, Critical Public Health, 8, 4:329-346. Parker R et al. (2002). Reaching Men who have Sex with Men. In: Lamptey P and Gayle J, eds. HIV/AIDS Prevention and Care in Resource Constrained Settings: A handbook for the design and management of programs. Washington, DC, Family Health International. Patton C (1999). Inventing ‘African AIDS’. In: Parker R and Aggleton P, eds. Culture, Society and Sexuality, a reader. London & Philadelphia, UCL Press. Simooya O and Sanjobo N (2001). In: But Free—an HIV/AIDS intervention in an African prison. Culture, Health & Sexuality, 3, 2, 241-251.

Strathdee S et al. (1998). Determinants of sexual risk-taking among young HIV-negative gay and bisexual men. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology. 1998. Sep 1, 19(1):61-66. Vanita R (2002). Queering India – Same sex love and eroticism in Indian culture and society New York and London, Routledge. Waldo CR et al. (2000). Very young gay and bisexual men are at risk for HIV infection: the San Francisco Bay Area Young Men’s Survey II. Journal of Acquired Immune Deficiency Syndromes, Jun 1, 24(2):168-174.

61