Introduction: why men?
The 1990s have seen several shifts in the emphasis and direction of reproductive health programmes. However, very little of the refocusing of interest and output has been directed at the rights of men. Most concern is still focused on men's roles and responsibilities in relation to the health of their female partners. For women, the traditional focus on family planning programmes and determination of individual or couple-level contraceptive requirements has been replaced by a more holistic approach encompassing a broader spectrum of reproductive health concerns. This shift has arisen from separate directions which have combined to challenge the largely demographic imperative of lowering population growth rates towards a broader refocusing on human welfare and individual choice, and the public health goal of reducing reproductive and gynaecological morbidity (Collumbien & Hawkes, unpublished). Proponents of the wider conceptual framework have defined reproductive health to include both ‘family planning and sexual health care’ ( United Nations 1994). Putting this definition into operation at a programme level implies provision of ‘the widest range of services without any form of coercion’. In contrast, for men there has so far been relatively little output from this broader reproductive health agenda. Men tend to be seen as important only in respect to their female partner's health and use of services. Reproductive and sexual health care aimed specifically at men is still in its infancy.
The broadened reproductive health care agenda has stimulated an interest in sexually transmitted infections, including, but not limited to, HIV. Among the historically discernible influences resulting in this change three stand out. First, women's health advocates and other interest groups voiced dissatisfaction with the vertical contraceptive delivery systems designed to achieve demographic goals and which overlooked other issues in reproductive health ( Sen et al. 1994 ; Collumbien & Hawkes, unpublished). Second, the growing HIV pandemic raised awareness of other sexually transmitted infections, given the acknowledged interrelationship of the two epidemics ( Jones & Wasserheit 1991). Finally, the influential 1993 World Development Report disclosed the burden of ill-health caused by sexually transmitted infections (STIs) ( World Bank 1993), including HIV/AIDS.
The introduction of STIs, including HIV, into the framework of reproductive health care provision has necessitated the incorporation of men as potential participants, partners and ‘consumers’ of services. Policy makers, programme planners, researchers and health advocates alike recognize that men's reproductive health and their sexual behaviour have direct effects on women's health. This was codified during the 1994 International Conference on Population and Development in Cairo ( United Nations 1994). ICPD explicitly sanctioned signatory countries to develop reproductive and sexual health programmes which are comprehensive in scope and coverage and are accessible to both men and women. As the ICPD Programme of Action states: all countries should strive to develop ‘innovative programmes … to make information, counselling and services for reproductive health accessible to … adult men’. Moreover, the ICPD declaration targets men for STI control strategies. Countries are advised to ‘encourage and enable men to take responsibility for their sexual and reproductive behaviour … and to accept the major responsibility for the prevention of sexually transmitted diseases.’
A review of literature on involving men in reproductive health care programmes reveals that attention has focused on an interest in the knowledge and use of male methods of contraception ( Ringheim 1995; Hulton & Falkingham 1996); methods to increase support for the partner's use of contraception through joint decision making ( Mahmood & Ringheim 1997; Karra et al. 1997 ; Lasee & Becker 1997; Bankole & Singh 1998; Johansson et al. 1998 ); and preventing the spread of STIs by more responsible male sexual behaviour. However, in both the ICPD document and the published literature more generally, relatively little attention is given to men's own reproductive and sexual health concerns. Basu (1996) points to the overwhelming imbalance towards women's rights and men's responsibilities: a near-exclusive demand for female reproductive health since gender issues remain (rightly) at the forefront of concern.
Whereas ‘sex’ is used to refer to biological differences between males and females, assumed to be universal and unchanging, ‘gender’ refers to culturally constructed notions of masculinity and femininity, and these are therefore highly variable, depending on the culture or society where gender roles are expressed. Gender differences are rooted in sex differences but, as Oakley argued in an early text on the topic, whereas we are born male or female, we must learn to be masculine or feminine ( Oakley 1972). Despite this, the concept of gender has often been used to refer specifically to the situation of women rather than social relations between women and men that have resulted in gender inequalities, not least in the area of health ( Hunt & Annandale 1999). In relation to sexual and reproductive health, unless we include men in the simple but often overlooked social and epidemiological equation, we are going to continue along a path that will exclude men, and do harm to women.
The most recent International Conference on Population and Development was unequivocal in its support for advancing gender equality through support for ‘the elimination of all kinds of violence against women, ensuring women’s ability to control their own fertility and … eradication of all forms of discrimination on the grounds of sex'. The promotion of gender equality and ending of sex discrimination is long overdue. However, we need to ensure that the women-centred focus of programmes post-Cairo does not result in a potential discounting of the needs of poor and disadvantaged men in terms of health, equality and empowerment.
There are some key players in the reproductive health field who have questioned the wisdom of including men in programmes. They worry that men will come to dominate the arena of reproductive health, not only in their predominance as policy makers, programme managers, scientists, and researchers, but now also as consumers of reproductive health services ( Berer 1996). Concern is voiced that allowing men to enter into previously female domains (for example, family planning clinics now transformed into comprehensive reproductive health care centres) may discourage women from using such services ( Wegner et al. 1998 ).
This paper considers why researchers, health care providers and programme and policy makers should take into account the reproductive and sexual health needs of men as well as women; how programmes can be designed which acknowledge and are responsive to men's sexual health needs and concerns; and how by doing so, the outcome will hopefully be beneficial not only to the men accessing services, but will have a positive impact on the health of women too.
In undertaking research among men it is important to recognize from the outset that ‘men’ are not a homogeneous group with the same needs and concerns in all areas of the globe. Men are characterized not only by their sex and gender, but also by their age, ethnicity, sexuality, income, educational status, occupation, geographical location, their position within a family, access to information and their ability to put such information to use. When seeing men in these terms, it becomes clear that social and demographic factors must be taken into account when researchers wish to look beyond the prevalence of sexual (ill) health in men and seek to discover its determinants. Moreover, men's sexual and reproductive health needs are not static but will change over the course of a lifetime.
Surveys in India have shown, somewhat unsurprisingly, that men with higher levels of education, higher economic status, and those living in urban areas, have better knowledge of reproductive health matters, seek treatment more frequently and are more likely to protect themselves against STDs than other men ( Singh et al. 1998 ). Younger men know less about fertility or the female menstrual cycle than older men ( Drennan 1998). Similarly, young men in their late teens in a working-class area of Glasgow, United Kingdom, who were at various stages of sexual engagement with young women, held very different views as to the nature of sexual relationships, and were highly variable in their adoption of safer sex, even though their knowledge of the risks of HIV was good. Thus there was significant variation amongst these young men although they were recruited because they shared age, class and neighbourhood characteristics ( Wight 1993). Hence when looking at the issue of ‘men’ we need to be clear from the outset which men we are talking about. The issues to be addressed in both survey and intervention design will clearly differ according to a number of social, cultural, economic and demographic variables defining the ‘men’ in a target group.
What do we know about the sexual and reproductive health, sexual behaviour and sexuality of men?
Historically the sexual health of men in low- and middle-income countries has received very little attention, either from the research community or from public sector health care planners and providers. Similarly, in high-income countries, the sexual and reproductive health of men has, until relatively recently, remained hidden from academic, medical or policy scrutiny ( Pfeffer 1985). The HIV epidemic changed this somewhat, with many interventions aimed specifically at informing men about safer sexual practices.
The predominance of sexual and reproductive health interventions aimed at women is predicated on the fact that women bear a greater burden of reproductive mortality and morbidity as they shoulder the physical, and most of the social, responsibility for childbearing and childcare. When men have been included in reproductive health surveys and programmes, they have featured mainly as determinants of women's fertility. Very rarely have surveys been undertaken which concentrate on determining men's own sexual and reproductive health concerns and needs.
Many studies which have included men as participants focused on family planning, and more specifically, on how men determine the fertility choices of their wives. Demographic and health surveys from the 10-year period 1987–97 have shown that many men know, and approve, a variety of family planning methods ( Drennan 1998). Combined data from Demographic and Health Surveys (DHS) in 21 countries shows that the majority of men in most surveys can name at least one method of contraception – for example, in 15 of the 21 countries included in one analysis, more than 90% of men could name a contraceptive method ( Drennan 1998). Most men ‘approve’ of contraception and think that they should share responsibility for family planning with their wives. However, the level of approval for family planning is generally determined by a number of factors including country of residence, socioeconomic and educational status. Male methods of contraception (condoms and vasectomy) are generally the least used of all contraceptive methods, although this varies greatly between countries. Use of condoms as a couple's major method of family planning varies from less than 1% (e.g. in much of sub-Saharan Africa) to more than 40% (in Japan). Similarly, vasectomy rates range from 0% in many countries to more than 10% in some East Asian societies.
However, beyond knowledge of, and attitudes towards, contraceptive methods, men in general have been found to have very little knowledge of basic reproduction ( Singh 1998), often misinterpret the signs of possible infection in the reproductive tract ( Olukoya & Elias 1996), and may be unaware of danger signs during potentially complicated pregnancies or labours.
Other surveys indicate the strong relationship between men's actions and women's health outcomes. For example, studies of STI risk factors among married women have shown a consistently strong relationship between a woman's risk of having a sexually transmitted infection and the reported sexual behaviour of her husband rather than the woman herself. ( Hunter et al. 1994 ; Moses et al. 1994 ; Thomas et al. 1996 ). Researchers in Pune, India, for example, found that women who were not commercial sex workers who attended an STI clinic had an HIV rate of over 13% ( Gangakhedkar et al. 1997 ). The only significant reported risk for these women was sex with their husbands.
There are many reasons why married men are more likely to contract STIs compared to their wives: men may be more likely to be involved in sexual networks that include more than one partner; they are generally more mobile and more often involved in inter- or intracountry patterns of migration in pursuit of employment opportunities; and men are more likely to purchase commercial sex. In societies where sanctions against nonmarital sex are less harsh for men than for women, and where women are subjected to tight social and cultural control, women's noncommercial extramarital sexual relations may be less common than in other societies. Labour migration and the predominantly male demographic shift to urbanization in developing countries may further increase both the opportunity and the demand for commercial sexual relations.
The role of sexuality
Throughout the world, there is often a fixed notion of men's sexuality. ‘Categories’ of sexual behaviour are sometimes seen as immutable and somewhat rigid: ‘heterosexual’ (straight), ‘homosexual’ (gay), or ‘bisexual’. However, it is clear that there is fluidity and temporal variation in men's sexual behaviour and sexual identities. Studies indicate that, amongst ostensibly homosexual men, there are reportedly high levels of heterosexual experience, with up to 10% of samples of men recruited to investigate homosexual behaviour reporting sex with female partners in the previous year ( Fitzpatrick et al. 1989 ; Davies et al. 1993 ). Studies of the same sex behaviour of ‘heterosexual’ men are difficult to find, and it is often not clear whether reports of homosexual behaviour are by men whose primary sexual experience is heterosexual. Nevertheless, one of the few surveys of a representative sample of men and women in the UK – the National Survey of Sexual Lifestyles (NATSAL) – showed that 6.1% of men reported some lifetime homosexual experience ( Johnson et al. 1994 ).
In some regions, notably central and South America, and in countries such as Mexico ( Liguori et al. 1996 ), the Dominican Republic ( de Moya & Garcia 1996), Peru ( Caceres 1996) and Brazil ( Daniel & Parker 1993), bisexual behaviour amongst men is accepted much more readily than in other parts of the world, particularly amongst those who are the insertive partners in anal intercourse. Indeed, in many of these countries, the majority of cases of HIV infection occurring in women early in the HIV/AIDS epidemic were a consequence of the secondary transmission of the infection from male partners who had contracted the infection through homosexual sex ( Mann et al. 1992 ). It is also apparent that in many countries male prostitutes who receive payment for sex from other men do not primarily identify as gay, and may be married or have regular female partners ( Aggleton 1999).
In the UK NATSAL study, male respondents were asked about their attendance at STD clinics. Whilst this measure does not indicate morbidity per se, and to some extent is more specifically a measure of health-seeking behaviour, it is clear that a proportion of reported attendance was associated with a sexually transmitted infection. More than half of those men who had five or more homosexual partners in the last 5 years reported attendance at an STD clinic, compared to only one in seven of those men reporting five or more heterosexual partners. In a recent study of homosexual men in Scotland ( Hart et al. 1999 ), respondents were 4.5 times more likely to report ever having a lifetime experience of STD than heterosexual men who reported STD clinic attendance in the NATSAL study (36%vs. 8%), and 12 times more likely to have reported an STD in the last year (11%vs. 0.9%).
The importance of these findings is twofold. Firstly, any provision of sexual and reproductive services for men must take into account the range of expression of men's sexuality in different cultures, and not assume exclusive (or compulsory) heterosexuality in the entire male population. It is a cornerstone of sexual and reproductive health service provision that health care workers should be nonjudgemental with regard to the reported sexual behaviour of patients, as proscriptive attitudes are likely to result in delays in patients seeking treatment, or even avoiding services altogether. We would argue further that it is positively necessary to encourage and facilitate open discussion of sexual behaviour, including homosexual acts, as otherwise comprehensive service provision, including health promotion, will be compromised. If we consider that access to sexual and reproductive health services is a human right for men and women, then sexuality is an element of service delivery that must be included in protocols.
Secondly, homosexual behaviour can have direct and dramatic consequences for the female partners of these men. The high prevalence of reported STIs in homosexual and bisexual men indicates the potential for transmission of STIs, including HIV, to their female partners. Women are at risk of the infections contracted by their male partners, whether through heterosexual or homosexual sex, and this can therefore affect their own sexual and reproductive health (with such sequelae of untreated infection as PID and sub- and infertility), and, after conception, the health of the foetus and neonate.
Methods of studying male sexual behaviour in resource-poor countries
Since the identification of AIDS as a new disease in 1982, and of HIV as its causative agent in 1984, the study of sexual behaviour has been given a major impetus by this global epidemic of a life-threatening disease. However, the problems of studying human sexual behaviour have been recognized for at least the past 60 years ( Kinsey et al. 1948 ; Gagnon & Simon 1967; Coates et al. 1988 ; James et al. 1991 ; Wadsworth & Johnson 1991), with the primary focus on validity (the extent to which the data reported represent objective truth) and reliability (the extent to which a method generates comparable results over time or between populations).
The primary means of collecting self-reported sexual behaviour is through questionnaire-based surveys. That is, either self-completed questionnaires, interviewer-based surveys, or a combination of both. Beyond the issues of validity and reliability, however, there are a number of other problems. The first is cost. Random sample surveys of the sexual behaviour of entire populations are extremely rare, even in developed countries, because of the major resource and logistical implications. Although quota sampling is cheaper (whereby numbers of people are recruited in proportion to their demographic profile in the population at large), it remains an expensive option in developing countries, and is in any case problematic in terms of population representativeness (only those immediately available and willing to participate do so, introducing bias). The majority of studies of sexual behaviour are in fact opportunistic and group-specific: young people in schools and colleges, gay men in bars, prostitutes on the street, men and women recruited through work or a shared social/sporting activity.
A second problem is more particular to some resource-poor countries. Where literacy levels are low, the use of self-completed questionnaires is inappropriate, and other methods must be found of determining sexual behaviour. Although surveys administered by researchers can overcome this problem, it is considered inappropriate in many societies to discuss sexual behaviour with strangers and in others it may be considered polite to give a stranger the information that they appear to want (whether this is reporting relatively few or unfeasibly large numbers of sexual partners). Whilst survey methods should not be dismissed purely for reasons of the embarrassment of researchers, or because of undue political or other considerations, other ways of generating sexual behaviour information must also be considered.
Focus group discussions (FGDs) are a particularly appropriate method of data collection with some populations of men, as found in a study of men's attitudes to female-controlled methods of STD prevention in South-Western Uganda ( Pool et al. 2000 ). Apart from the frequently cited benefit of FGDs, namely that they succeed in producing data through interaction and discourse in a group setting that one-to-one interviews may fail to generate ( Kitzinger 1995), they can also be a culturally sensitive data collection method. The majority of men in the Ugandan study came from the Baganda ethnic group, the culture of which is primarily orally based and community-focused. Outside of the family, men and women are socialized and participate in a mainly homosocial environment, in which same-sex group discussions serve as a significant forum for reinforcing group identification and therefore function to increase social cohesion. Invariably this method is limited in that men are speaking in a group situation and therefore may be tempted to ‘present a face’, but as a means of accessing men's public discourse on topics of sexual and reproductive health, this method of data collection is to be recommended. It has also been employed in The Gambia in West Africa with young men from rural areas discussing their sexual health seeking-behaviours (Miles et al. unpublished); again it has proven to be an appropriate method to employ in this homosocial culture in which a significant amount of social activity occurs in same sex contexts.
Focus groups and cross-sectional surveys may be limited in the extent to which they can generate valid and reliable data on individual behaviour, but the anthropological method of participant observation and in-depth interviews undertaken repeatedly over a long period of time may result in improved data generation and recording. Once a researcher has been accepted into a community and is trusted by respondents, this essentially longitudinal method allows researchers to go back to key informants to clarify apparently contradictory findings or to confirm and authenticate discoveries and insights. As with much research in the field of human sexual relations, tact, delicacy and patience are all prerequisites if men are to provide researchers with the information on sexual norms, values and behaviours that they are seeking.
There have been attempts at novel means of data collection for nonliterate populations, most recently in a rural area of The Gambia, where secret ballots as a tool for evaluation of sexual behaviour change with men in same sex groups ( Shaw et al. 1999 ) were tested. Questions, for example, on sexual partners were marked by the appropriate number of ‘sticks’. For other questions in order to answer ‘yes’ the respondent had to mark a triangle, and for ‘no’ they had to mark a circle. It was explained that the ballot sheets were completely anonymous as participants were not to let anyone else see their answers. At the end of the session each participant folded their ballot sheets and put them into a hat. This method generated very high levels of ‘yes’ responses which did not accord with other methods of data collection employed (e.g. one-to-one interviews, known prevalence of HIV infection and other STIs), perhaps because in this culture to say ‘no’ outright is considered rude, but with further refinement the method may prove to be promising in this and other settings.
This brings us to a final methodological point with regard to data collection on sexual behaviour. Wherever possible, methods should be ‘triangulated’; that is, data inputs from a variety of sources are combined. For example, self-reports of STI history or other reproductive health problems in a population can be compared with health care attendance figures for these problems. Given that many men seek care outside the public sector (see below), both formal and informal health care providers should be included in such corroborative surveys. Epidemiological data on incident STIs, STI complications (such as infertility) and on perinatal outcomes can be factored in where these are available. These are essentially the methods of Rapid Assessment and Response, which has been used successfully to determine levels of drug use in resource-poor countries ( Power 1998; Rhodes et al. 1998 ), but have been less frequently employed in the field of sexual and reproductive health.
What are men concerned about?
Given the number and variety of reproductive health surveys among women to determine the prevalence and associated risk factors of gynaecological morbidity, and their associated reproductive health needs, we have seen above that very little is known about the sexual and reproductive health needs of men. However, if we are aiming to establish appropriate programmes, interventions and services which see men as full participants, we must begin to understand their own thoughts, feelings and concerns around the issue of sexual health. Whilst it is undoubtedly a step in the right direction to start encouraging men to access family planning services, this is unlikely to be a sufficient incentive for their full involvement in the reproductive health agenda. Men, like women, require holistic and comprehensive sexual health services which are responsive to their needs. To ascertain which services will be appropriate it is vital to determine the sexual health needs and concerns of the men in the community being served.
Data from population-based studies in Orissa (India) and rural Bangladesh illustrate some men's perceptions of their sexual health problems: more than 2000 men in two recent surveys indicated that their major sexual health concerns related to psychosexual disorders. Worries about semen loss were predominant in Orissa, whereas in Bangladesh men were more generally concerned with a wider variety of problems of a psychosexual nature – concerns about semen loss and ‘nocturnal emissions’ were volunteered by over 15% of the men questioned. Other commonly reported problems concerned sexual performance: premature ejaculation and inability to maintain an erection (Collumbien & Hawkes, unpublished). Such concerns were mirrored in concomitant clinic attendance figures for newly established male sexual health clinics in rural Bangladesh. Over 40% of attendees reported psychosexual dysfunction as their main reason for coming to the clinic ( Hawkes 1998).
The prevalence of reported psychosexual anxieties among the men in these two studies is not dissimilar to that from other published reports. Analyses of many studies in both Europe and the United States among community and clinic-based samples found a surprisingly consistent percentage of men who self-report psychosexual concerns of premature ejaculation (35–38%), male erectile dysfunction (4–9%), and inhibited male orgasm (4–10%) ( Spector & Carey 1990).
Thus, in designing survey instruments to study men's sexual health, and in thinking about the range of possible interventions suitable for improving men's sexual health, both researchers and programme planners need to consider the full range of concerns voiced by men. Whilst the international community of policy makers, health advocates, and colleagues may be concerned with, and focus upon, the problems of the sexually transmitted infections, men themselves may have a wider set of concerns. If men are to be encouraged to access services, services must reflect and respond to men's stated concerns.
Where do men go for treatment?
Although it has been suggested that policies to provide clinical services for men may, as a consequence, reach asymptomatic but infected women through partner notification strategies, such interventions are only likely to succeed if men initially seek care from providers who are trained to give appropriate and effective care. Such care would include not just determining the appropriate antibiotics to use for symptomatic men, but would also involve training health workers in other aspects of STI management: counselling, condom promotion, treatment for partners.
Concerns around provision of services for men often centre on how to integrate men's services into the existing health care system – for example, how to bring men into the family planning system or the primary health care system (especially if the latter is predominantly a site of maternal and child health services) (Wegner et al. 1998). It is clear, however, that men do already seek care for their sexual health concerns, and it may be a better starting point to try and determine where and why they seek such care. In this way, programme managers, for example, can decide which health providers need to be included in training programmes, and what needs and incentives govern these providers.
Surveys among men in both rural and urban areas of Bangladesh found that most men with symptoms do seek care, and that the vast majority (95%) of them do so in the multifaceted private sector ( Hawkes 1998). Given that the public sector primary health care services are mainly concerned with issues of maternal and child health, it is perhaps not surprising that men are forced to go to private providers with their sexual health problems. In contrast, a study among men in rural Tanzania found that most men with symptoms suggestive of an STI sought care in the ‘official’ health sector ( Newell et al. 1993 ). In designing an appropriate response to such a finding, policy makers and programme managers have to take into account (a) whether men can be persuaded to go to services traditionally seen as being a female domain; (b) whether the public sector can afford to now start seeing men as well; and (c) the costs and possibilities for training private providers in appropriate and effective care strategies compared to training public sector workers to treat more men.