The promises and limitations of gender-transformative health programming with men: critical reflections from the field

Abstract

Since the 1994 International Conference on Population and Development, researchers and practitioners have engaged in a series of efforts to shift health programming with men from being gender-neutral to being more gender-sensitive and gender-transformative. Efforts in this latter category have been increasingly utilised, particularly in the last decade, and attempt to transform gender relations to be more equitable in the name of improved health outcomes for both women and men. We begin by assessing the conceptual progression of social science contributions to gender-transformative health programming with men. Next, we briefly assess the empirical evidence from gender-transformative health interventions with men. Finally, we examine some of the challenges and limitations of gender-transformative health programmes and make recommendations for future work in this thriving interdisciplinary area of study.

Desde la Conferencia Internacional sobre Población y Desarrollo de 1994, investigadores y profesionales han aunado esfuerzos para cambiar los programas de salud para hombres desde un enfoque de género neutro hacia uno más sensitivo y transformativo. Especialmente en la última década, los esfuerzos en esta última categoría se han utilizado cada vez más para intentar transformar las relaciones de género de forma que sean más justas en aras de mejorar los resultados sanitarios tanto para hombres como para mujeres. En primer lugar, evaluamos la progresión conceptual de las contribuciones de la ciencia social a los programas sanitarios transformativos de género para hombres. A continuación, valoramos brevemente las pruebas empíricas de las intervenciones sanitarias transformativas de género para hombres. Y para terminar, analizamos algunos de los retos y las limitaciones de los programas sanitarios transformativos de género y hacemos recomendaciones para el futuro trabajo en esta floreciente área interdisciplinaria de estudio.

Depuis la conférence internationale sur la population et le développement de 1994, les chercheurs et les praticiens se sont impliqués dans une série d'actions visant à transformer les programmes de santé destinés aux hommes, caractérisés par le sexisme, en programmes de santé plus sensibles à la dimension de genre et favorables à l'évolution des inégalités de genre. Les actions entrant dans cette dernière catégorie ont été menées de plus en plus fréquemment, en particulier au cours des dix dernières années; elles visent à rendre les relations de genre plus équitables au nom de l'amélioration de la santé, des hommes comme des femmes. Nous commençons par évaluer le progrès conceptuel de la contribution des sciences sociales aux programmes de santé ciblant les hommes et favorables à l'évolution des inégalités de genre. Ensuite, nous évaluons brièvement les données empiriques des interventions de santé dans ces mêmes programmes. Enfin, nous examinons certains des défis et des limites inhérents aux programmes de santé favorables à l'évolution des inégalités de genre et nous formulons des recommandations pour ce qui concerne les futurs travaux dans ce dynamique champ de recherche interdisciplinaire.

Keywords:

• gender norms
• masculinities
• HIV
• violence
• intervention

Introduction and overview

Over the past decade, global health programming with men has increasingly shifted from gender-neutral to gender-sensitive and gender-transformative (Barker et al. 2010a; Barker, Ricardo, and Nascimento 2007; Dworkin 2015). Just as researchers and practitioners have recognised that gender-neutral programming failed to take the gendered context that shapes women's health into account (Exner et al. 2003), they have also recognised that programming undergirded by gender-neutral assumptions limits an understanding of how gender relations drive men's health and can be intervened upon (Courtenay 2000a; Dworkin, Fullilove, and Peacock 2009). Geeta Rao Gupta (2001) first presented a now familiar conceptual framework to classify the extent to which health interventions engage with critical gender-related issues. While gender-neutral interventions do not take gender into account, Gupta defined gender-sensitive interventions as those that recognise the differing needs and constraints of women and men. Gender-transformative approaches, in contrast, seek to reshape gender relations to be more gender equitable, largely through approaches that ‘free both women and men from the impact of destructive gender and sexual norms’ (10).

Indeed, freeing men from gender and sexuality norms that negatively impact men and women appears to be a crucial step towards achieving gender equality and improving health. Given the rise of gender-transformative health programmes with men, now is a crucial time to step back, critically assess, and reflect upon the progress that has been made to date. In the sections that follow, we first delve into how theory and research from the social sciences have conceptually influenced global health programmes that are focused on work with men to improve sexuality and reproductive health outcomes. We then shift to an assessment of the empirical literature that focuses on the impact of gender-transformative programmes on health. Finally, we examine some of the challenges and limitations of gender-transformative health interventions and make recommendations for future work in this thriving interdisciplinary area of work.

A brief conceptual history of ‘gender’ in ‘transformative’ health programming

Global health programmes have increasingly targeted the social determinants of health in the form of economic and social factors (e.g. race, class, working conditions) that influence the ability of individuals, groups and communities to attain good health (Commission on the Social Determinants of Health 2008). Gender is a social determinant of health, and health-related interventions for men have progressively drawn upon various social science understandings of masculinity. In the mid-twentieth century, sex role theory (men enact male gender roles, women enact female gender roles) – and later the concept of ‘gender roles’ – offered an understanding that individuals are socialised to act according to the expectations associated with one's biological sex (Connell 1987, 1995). Although global health scholars have amply drawn upon the concept of roles in health programmes, social science scholars have critiqued roles as being too individualised, as implying a fixed notion of gender, and as leaving out the importance of power relations (Connell 1987, 1995; Connell and Messerschmidt 2005; Messner 1997). Nonetheless, the use of roles was important in global health programming given that it opened up a modifiable avenue for intervening on expectations about what it means to be a man (Barker et al. 2010a). In addition, both social scientists and global health scholars have empirically shown that men experience gender role conflict – and worse health outcomes – when they fear they cannot live up to masculine gender role norms (Fleming et al. 2014; Gottert 2014; Pleck 1995).

Role-based versions of masculinity in global health soon shifted to a broader understanding of masculinity as an ideology or a set of beliefs. In this shift, social scientists conceived of gender relations not only as roles but rather as a dynamic pattern of social relationships that are enacted by individuals – and are patterned into social institutions such as sport, bars, fraternities, the military, work, etc. Research has found that when men internalise masculine ideologies that link heterosexuality, the subordination of women and aggression, this can be harmful to both women's and men's sexuality and reproductive health (Bowleg et al. 2011; Crosset 2000; Harrison, Chin, and Ficarrotto 1995; Kalichman, Cain, and Simbayi 2011; Messner and Stevens 2002). Here, then, global health scholars who engage in gender-transformative programming have been influenced by a social science understanding of gender not as something that one ‘is’ but as something that one ‘does’ in a patterned set of social interactions within social institutions – a pattern that can be ‘undone’ (West and Fenstermaker 1995; West and Zimmerman 1987). Courtenay's (2000b) crucial work theorised the link between repeated enactments of masculinity and poor health beyond roles when he wrote:

From a constructionist perspective, women and men think and act in the ways that they do not because of their role identities or psychological traits, but because of concepts about femininity and masculinity that they adopt from their culture (Pleck, Sonnestein and Ku 1994). Gender is not two static categories, but rather ‘a set of socially constructed relationships which are produced and reproduced through people's actions’ (Gerson and Peiss 1985, 327); it is constructed by dynamic, dialectic relationships (Connell 1995). Gender is ‘some-thing that one does, and does recurrently, in interaction with others’ (West and Zimmerman 1987, 140) it is achieved or demonstrated and is better understood as a verb than as a noun (Kaschak 1992; Bohan 1993; Crawford 1995). Most importantly, gender does not reside in the person, but rather in social transactions defined as gendered (Bohan 1993; Crawford 1995). From this perspective, gender is viewed as a dynamic, social structure.

The above social science conceptions of masculinity are critical for two main reasons. First, they have added a relational dimension to an understanding of gender norms that global health programming could draw upon so as to not repeat the long-held erroneous conflation of gender with women (Dworkin, Fullilove, and Peacock 2009; Kimmel 1996). Second, with gender understood as a verb – as something that people do rather than an internal essence of what one is – therein opened up the possibility that gender relations could be intervened upon as patterns of behaviour and/or social practices. However, if gender relations are reduced solely to individualised roles and/or norms within global health, then some social scientists warn that this potentially ignores the ways in which women and men are differentially positioned in social institutions and thus may experience different drivers of and harms to poor health (Connell 1987, 1995). Here, global health programming with men specifically recognises that it is critical to transform power relations between women and men to effect lasting change in gender relations and health outcomes. The rise of work on ‘men as partners’ in women's health and on gender-transformative programming emerged out of the International Conference on Population and Development (ICPD) in Cairo in 1994. This conference spurred a paradigmatic shift on the role of gender in development outcomes and men were increasingly viewed as playing a key role in transforming gender relations to be more gender equitable. Delegates and their organisations recognised this as an important point of intervention for improving sexual and reproductive health outcomes, as has been noted by the United Nations Population Information Network (1994):

Changes in both men's and women's knowledge, attitudes and behavior are necessary conditions for achieving the harmonious partnership of men and women. Men play a key role in bringing about gender equality because in most societies, men exercise preponderant power in nearly every sphere of life, ranging from personal decisions regarding the size of families to the policy and program decisions taken at all levels of Government. (51)

One key conceptual turn at ICPD was not only to recognise the centrality of power relations and gender inequality to women's sexuality and reproductive health outcomes, but also to integrate a social science understanding of how men are also implicated in the gender order. That is, the health of women and men is shaped not only through women's structural-level or rights-based disempowerment, but also through men's adherence to narrow and constraining definitions of masculinity, known as the ‘costs of masculinity’ (Courtenay 2000a, 2000b, 2000c; Messner 2007). Indeed, research finds that men who adhere to dominant ideals of masculinity experience worse mental health outcomes, are more controlling of their sexual partners, engage in more high-risk sex, use violence to demonstrate power over others and avoid healthcare clinics more than men who challenge dominant notions of masculinity (Jewkes and Morrell 2010; O'Neil 2008; Santana et al. 2006).

It is this dual understanding of gender relations that has been so important to gender-transformative health interventions with men. That is, on the one hand, programmes have called upon a social science understanding of gender as relational and as built into patterns of social practices and dynamic social structures. Here, gender-transformative work intervenes with men to democratise gender relations between women and men. Simultaneously, gender-transformative programmes are influenced by social science theories that view masculinities as socially constructed, contested, fluid and locally and regionally determined. Using this view, because individual men are seen as participants in the construction of the gender order through the practice of masculinities, it is possible to challenge dominant norms by both encouraging positive aspects of masculinity in the name of improved gender equality and health and by disrupting the masculinity/femininity binary (Barker et al. 2010a; Connell 1995; Dworkin et al. 2013; Pulerwitz et al. 2010).

The promise of gender-transformative programmes

Do gender-transformative health interventions with men actually work to improve health? In 2007, World Health Organization colleagues applied this framework and published a systematic review of the evidence base examining health interventions targeting men and found that gender-sensitive and transformative programmes (defined earlier) were more efficacious than gender-neutral programmes at improving a variety of health outcomes (Barker, Ricardo, and Nascimento 2007). To update this work, Dworkin, Treves-Kagan and Lippman (2013) later published a systematic review focused on violence and sexually transmitted infection/HIV-related outcomes that included more recent studies and rigorous designs. In their review of programmes targeting outcomes related to HIV and sexually transmitted infection, they found that 9 out of 11 recent gender-transformative interventions resulted in statistically significant declines in at least one indicator of sexual risk. For programmes targeting physical or sexual violence against women, 6 out of 8 interventions showed statistically significant declines in the perpetration of violence. In addition, 11 out of 12 gender-transformative interventions revealed a statistically significant change in reconfiguring men's attitudes towards gender norms in the direction of more gender equality. Given the evidence, these authors concluded that gender-transformative programming with heterosexually-active men can play an important role in increasing sexually-protective behaviours, reducing HIV risks, preventing violence and changing attitudes towards gender norms to be more equitable.

It should be noted that more work needs to be done with regard to measuring the mechanisms of change in gender-transformative health interventions. Studies included in the recent systematic review (Dworkin, Treves-Kagan, and Lippman 2013) primarily examined whether changes in a validated ‘gender equitable men’ (GEM) scale led to positive changes in sexual and reproductive health outcomes. The GEM scale (detailed in Pulerwitz and Barker 2008) assesses the extent to which men agree with gender equitable and inequitable norms. Themes covered in the scale include: seeking relationships based on equality/intimacy or sexual conquest, equal rights for men and women, household labour and child care, and role of men and women in sexual health decision-making. Indeed, research has found that changes in GEM yield positive changes in sexuality and reproductive health outcomes (Pulerwitz and Barker 2008; Pulerwitz et al. 2010).

While the evidence base is still in a growth phase and it is urgent to improve the rigour and quality of existing study designs, there is some agreement that there is a ‘rapidly expanding evidence base … that has demonstrated that rigorously implemented initiatives targeting men can lead to significant changes in social practices that affect the health of both sexes’ (Peacock et al. 2009, S119). Despite the promising evidence, several questions still remain. Which dimensions of gender relations do gender-transformative interventions actually attempt to change? Programmes claim to attempt to democratise gender relations, but can gender relations actually be democratised solely through changing ‘gender norms’? While gender-transformative programmes have clearly demonstrated important successes and have drawn upon a merger of important social science and global health research, there are also numerous challenges and limitations that deserve further consideration in order to improve efforts in the future.

Challenges and limitations of gender-transformative work with men

We will now turn to identify and elucidate four main challenges and limitations of gender-transformative work based on a review of the literature and our own fieldwork in this area: (1) an overemphasis on harmful masculinities, (2) privileging a gender lens over an intersectional perspective, (3) struggle among some men with a newly democratising gender order and (4) lack of clarity on how to sustain changes in masculinities after programmes end. In discussing these challenges, we rely on data collected from our fieldwork with men in a health-related gender-transformative intervention carried out across numerous provinces in South Africa.

Focus on harmful individualised masculinities

Gender-transformative health programmes generally recognise that masculinities are constructed, fluid and modifiable. However, several scholars have underscored that work with men can reduce the complexities of masculinities to ‘problematic male attitudes and behaviors, such as violence and abuse of women and children, substance abuse, and risky sexual behaviors’ (Morrell, Jewkes, and Lindegger 2012, 3). A focus on the ‘problematic’ aspects of individual male beliefs, roles and behaviours places the onus on individual men to overcome complex and entrenched societal-level problems. While social scientists have pressed for an understanding of gender relations at the individual, interpersonal, institutional, cultural and societal levels (Connell 1995; Messner 1997), a focus specifically on norms within gender-transformative health interventions may unnecessarily limit the level of analysis to the individual, cultural and interpersonal realms (leaving out the structural realm). This emphasis implies that men's behaviours are solely about agentic choices and ignores a large body of research that shows that masculinities emerge out of cultural, historical and structural antecedents that shape the range and availability of masculinities that men select among and enact (Hunter 2004, 2005; Morrell 1998; Morrell et al. 2013). For example, research has shown how, instead of viewing South African men as enacting a problematic male role when they have multiple sexual partners, it is critical to consider how changing norms of masculinity emerged out of structural factors such as the system of apartheid and increased migratory needs due to shifting economic conditions and resulting long separations from partners (Hunter 2005). The above discussion highlights that while men do have agency that should be emphasised in gender-transformative programmes to make positive changes, it is equally important to recognise that this agency must also be considered within social, economic and cultural contexts that both constrain and enable men's individual and group-based choices.

We have already highlighted social science understandings of masculinities that press beyond individualised roles or norms and instead view masculinities as a collective set of practices that are shaped by economic, political and social contexts. Masculinities are not simply bundles of norms to undo at the individual and small-group levels. Rather, masculinities are shaped by poverty, migration, globalisation, racism and numerous other structural factors. Dworkin, Treves-Kagan and Lippman (2013), in their systematic review of gender-transformative interventions, found that only 3 of the 15 recent interventions included some aspects of community-level programming and community mobilisation activities to target these structural factors. Even fewer of these interventions are specifically focused on structural interventions.

Structural interventions for health are programmes or policies that target the contexts in which individual choices are enacted (Blankenship, Friedman, Dworkin, & Mantell 2006; Gupta et al. 2008). Such interventions seek to shift structural-level factors (e.g. poverty, migration, racism, gender inequality) with the explicit aim of improving health outcomes. Numerous governmental, non-governmental, policy-oriented and community-based organisations intervene at a structural-level to tackle a variety of issues. There is ample recognition in global health research with women that the range of femininities is constrained partly by structural factors. Structural interventions are therefore considered an important point of intervention for improving women's sexual and reproductive health outcomes (Dunbar et al. 2010; Grabe 2012). However, very few science-based global health gender-transformative interventions for men exist that attempt to shift the structural context in which masculinities and health are constituted (Bowleg and Raj 2012; Dworkin 2015). Given that gender-transformative programming for men currently includes few examples of structural interventions, this is clearly an area that needs to be bolstered in future work.

Why does it matter if programmes pin changes in health on the shoulders of individual men instead of helping to shift structures that shape masculinities? There are several reasons, one of which is that the selected level of analysis has implications concerning the extent to which the field can successfully and fully engage men now and in the future. For example, in our previously published work within a gender-transformative anti-violence and HIV-prevention programme in South Africa that conceptualised masculinities as a collective practice and drew upon community mobilisation strategies, men in our study still wondered:

… why [is this program] focused on men only? Men think that this is a form of making them culprits and they have shown their displeasure at the viewpoint that men are the castigators of abuse and violence … (Dworkin et al. 2013, 191)

In this case, focusing the intervention on men and violence unintentionally conveyed a message that only men are violent or perhaps that men are only to be understood as perpetrators of violence-related behaviours. Another man in our sample articulated the way that he understood the gender-transformative research of which he was a part: ‘I can say that they are researching men because they say men are lacking somehow in their behaviour.

The above quote does not mean that gender-transformative interventions should not intervene on the norms of masculinity that are found to be harmful to health. Still, without a structural approach that shifts the range of available masculinities and their social valuations, it can be expected that men across settings may feel that they are being asked to bear individual responsibility (e.g. changing gender norms) for massive social problems that influence masculinities and health outcomes (e.g. unemployment, poverty, violence).

Privileging gender over intersectionality

A related point is that gender-transformative approaches to improving sexual and reproductive health among men clearly privilege gender as the key axis of intervention. Such an approach does not adequately consider that there are differences and inequalities among men (Connell 1995; Messner 1997) that shape both health outcomes and the collective practice of masculinities. By focusing more on the health costs that occur when men adhere to narrow and constraining aspects of masculinities, gender-transformative work misses the intersectional nature of the identities and inequalities that shape men's health outcomes.

Intersectionality as a concept is credited to Kimberlé Crenshaw (1991), a critical race and legal scholar in the USA, and refers to the ways that structural and identity-oriented forces such as race, class, gender and sexuality are more useful to understanding inequality and health outcomes than any single axis alone. The concept has been used by numerous social scientists since the late-1980s to examine social inequalities (Baca-Zinn and Thorton-Dill 1993; Hill-Collins 1990) and was picked up by public and global health scholars in the mid- and late-2000s to better understand a variety of health outcomes, including sexuality and reproductive health outcomes (Berger 2005; Dworkin 2005, 2015; Gentry, Elifson, and Sterk 2005; Watkins-Hayes 2014).

Clearly, adopting an intersectional approach is urgent across a variety of health outcomes. For example, a brief look at the HIV epidemic in South Africa reveals that 31.4% of Black South Africans are infected, while only 1.1% of Whites are (Department of Health, South Africa 2012). An understanding of masculinities in South Africa is more nuanced when considering the role of class and socioeconomic status. For example, macro and local economic trends and deep, persisting inequalities in race, class and gender in South Africa have led to significant drops in formal employment and marriage rates and, consequently, a feeling that men no longer have access to the means to attain highly valued markers of masculinity (marriage and being a provider) (Hunter 2005; Morrell 1998; Morrell et al. 2013). This work suggests that masculinities are not solely about gender norms or some set of uniform gender privileges that men have, but rather that men's experiences and identities as men arise out of the post-apartheid context, the demise of the home economy and the crisis in the affordability of marriage, all of which clearly underscore how masculinities intersect with race and class relations.

An intersectional perspective is also needed to assess gender-transformative programming when considering the linkages between gender and sexuality. Gender-transformative programming privileges heteronormative masculinities and cisgender men. While some non-governmental and community-based gender-transformative programmes do address homophobia and transphobia, and also include men who have sex with men, science-based programming that is specifically gender-transformative largely ignores transgender and/or minority sexuality men. Gender-transformative programming in the future clearly needs to incorporate expanded notions of gender and sexuality to press beyond cisgender and heternormative understandings of men and masculinity.

The above points make it clear that men experience and enact masculinities differently depending on class, age, race, sexuality/gender identity/gender expression and other social locations and identities. Gender-transformative approaches do partly recognise this, and in the empirical literature there is some emphasis specifically on race and class-marginalised men's enactment of masculinities that are harmful to health. These populations are often the target population in gender-transformative work because evidence demonstrates that marginalised men disproportionately pay the costs of masculinity in terms of the impact on their health (Bowleg 2012; Courtenay 2000b). An example of an intersectional perspective is the Making Employment Needs [MEN] Count intervention by Raj et al. (2014) that tackles masculinities and structural factors that shape health, such as race and class inequalities. The MEN Count intervention provides Black men in the USA with case managers who provided ‘gender-equity counseling’ combined with assistance procuring or maintaining stable housing and employment (Raj et al. 2014). While the trends above are promising, the dominant emphasis in gender-transformative programming is on individual-level masculinities and, while this is important for effecting change, this emphasis can obscure other critical factors (e.g. economic inequalities, discrimination) that are intertwined with race, class, sexualities and other social locations.

Struggle among some men within a democratising gender order

Drawing upon points made in the previous section, gender-transformative programmes often call upon men who experience intensive and cumulative economic and social disempowerment to press towards the democratisation of gender relations between women and men. These are the same men who disproportionately experience the negative health costs of adherence to constraining definitions of masculinity (Courtenay 2000b; Dworkin 2010, 2015; Messner 1997). In our own and others' previous research, scholarship has underscored how heterosexually-active men can respond to requests to shift in the direction of more gender equality with backlash and masculinist attitudes and behaviours (Kimmel 1996; Messner 1998; Sideris 2004). While many men in gender-transformative programmes embrace women's rights and a shift in the direction of more gender equality, this can be extremely challenging for some men to accept and/or embrace due to their own feelings of threat and/or social disempowerment. This is not necessarily a problem of translation, where gender-transformative programmes struggle to communicate abstract principles of rights into concrete actions at the household or individual level (Dworkin et al. 2012). Rather, men's complex responses to shifting gender relations reflect the fact that they often have witnessed (or perceived) decreases in men's economic opportunities alongside improvements in women's status at the occupational, community, relationship and household levels, as well as enhancements in the protection of women's rights at the societal level.

Thus, in contexts of rapidly shifting gender relations where gender-transformative work is embedded, requests to shift men in the direction of more gender equality can be met by men with feelings of a decline in masculine status or household authority, feelings of uselessness and a destabilised sense of what it means to be a man (Messner 1998; Sideris 2004). For example, within our own published studies of gender-transformative health programming, men reported that while they agreed with the principles of women's rights in abstract terms, they also felt that in practical terms, rights were ‘an affront to manhood’ that made women ‘too demanding’ or that they were being ‘controlled’ in their relationships with women more generally. In addition, many men felt as if certain aspects of women's rights undermined ‘respect’ for men and diminished men's voice and decision-making power in the household (Dworkin et al. 2012, 2013).

A few men in our studies from South Africa stated that worsened health outcomes could result from pressing men in the direction of greater gender equality:

Men are violent against women as a direct action against equality. In some cases violence between women and men is about power. Men become powerless due to what the country calls for – change and equality. Some men will want to show women that they are powerful in a physical form. Some resort to sexual abuse and rape just to prove that they are more powerful than women and because they do not want to accept the changes.

In many ways, the sentiment expressed in this quote highlights the importance of gender-transformative interventions with populations of men undergoing shifts in gender relations. It is also important to highlight that these changes in women's empowerment and status would still occur – though perhaps not at the same pace – even if public and global health interventions did not promote gender equality.

There are a number of important issues to tease out here. First, it is important for interventions to consider conjoining content on gender equality with content on the costs to marginalised men of adhering to narrow constructions of masculinity (Dworkin 2015). A dual focus not only helps to successfully engage men on important questions such as how masculinity shapes their own and their partner's health, but also ensures that men do not feel attacked or blamed for equality and health issues in their communities. Second, because it is poor and marginalised men who disproportionately experience negative health outcomes (Baker et al. 2014; Courtenay 2000b; Dworkin, Fullilove, and Peacock 2009) and are the targets of gender-transformative programmes, these programmes risk reinforcing notions that dominant men are not in need of change, while subordinated men asked to carry the burden of increasing gender equality. While not the current emphasis of gender-transformative programming, dominant men (men who occupy positions of privilege at the top of social hierarchies and institutions) could also be targeted with policies and programmes and could be the target of advocacy-related efforts due to their positions of power and as policy-makers. For example, nationwide governmental programmes such as Chile's Cresce Contigo (engaging men in prenatal care and childbirth) or Brazil's National Comprehensive Healthcare Policy for Men (HIV-testing and violence prevention for all men in public health system) are implemented to men across societal divisions (Separavich and Canesqui 2013). In addition, the One Man Can programme based in South Africa has engaged multiple sectors of the society, including elite politicians, civil society, as well as poor and working-class neighborhoods. These efforts may counter feelings that gender-transformative programmes are pinning gender equality and health changes on the shoulders of already marginalised groups of men.

Additionally, it is important in programme content not only to consider the democratisation of the gender order in terms of women and men, but also how hierarchies of masculinities can be taken into account. It is therefore critical to highlight that men's disempowerment contributes to their increased likelihood of being victims of violence at the hands of other men, perpetrating violence against other men and perpetrating violence against women (Fleming et al. under review; Matjasko et al. 2012). When gender-transformative programming focuses solely on democratising gender relations between women and men – thus focusing on violence against women – the work often ignores the substantial violence that occurs between men (men are significantly more likely than women to be victims of grave physical violence and violent death) (Krug et al. 2002; WHO 2013, 2014). These facts also represent a missed opportunity in gender-transformative programming to simultaneously focus on violence against women and violence against men, and to test whether programme content is effective for reducing violence among both or is more effective for some groups over others.

The question of long-term change

While ‘changed men’ are the goal within gender-transformative programmes, it remains unclear how the maintenance of new patterns of masculine practice are continued after the close of programmes. This is partially due to the nature of intervention evaluations and research that typically lack long-term post-intervention assessments. More practically, however, it also remains unclear how, in the absence of broad-based contextual/structural changes, men will succeed in continuing to enact gender equitable practices beyond the short-run. In the words of one man from our previous studies, who characterised the sentiments of several men: ‘this program is good but it is not enough as we need more long-lasting solutions. It is not enough to have a few good men’. Other questions that remain relatively unexamined in the gender-transformative health literature include the following: How do women and men in communities where gender-transformative programmes are implemented respond to more equitable men at the close of programmes? Do they accept men who shift their behaviours and beliefs towards ideals of gender equality or do women and men use social control mechanisms to reinforce previously embraced and valued notions of masculinity?

In our work that focused on the impact of a gender-transformative programme on gender ideologies, violence and HIV risks, some men reflected upon the pushback they received from their peers after receiving the intervention:

What happened is that after the workshop, I met other guys in the village and had conversations with other men about what we had been taught. I told them that even men can wash the baby nappies and take care of children. I cannot describe the reaction but all I can say is that they were shocked and it was as if I was going to harm them. After that they then told me that I am a sissy boy, a softie, and some even suggested that maybe I am gay. They looked down on me and that really made me feel humiliated and feeling like I was less of a man. I then realised that in their way of thinking to them … I am not the real man.

This quote demonstrates that gender-transformative programming attempts to equalise relationships between women and men, but it is clearly less focused on equalising relations among men (e.g. between hegemonic and subordinated masculinities) – these too reproduce gender relations and impact masculine norms. To achieve long-term, sustainable change, community-level interventions need to be accompanied by broader policies and programming that are synergistic with the changes sought by these interventions. Without the accompanying societal-level change, men who adopt behaviours that are more gender equal (e.g. ‘wash[ing] the baby nappies’) may find themselves relegated to subordinated status by their peers (e.g. ‘to them … I am not the real man’). As other scholars have pointed out, masculinity is a valuable resource that some men use to construct status when they are otherwise marginalised (Courtenay 2000b; Majors and Bilison 1992) and without the reinforcement of more expanded definitions of masculinity through universal policies/programmes, it remains unclear how valuations of hegemonic masculinity will be dislodged among men who were not part of the intervention at the local level.

Earlier in this paper, we underscored the importance of relational definitions of gender, whereby the simultaneity of masculinity and femininity are taken into account in health programmes. Because women are not often the objects of study in gender-transformative interventions (though they are sometimes included), how shall new forms of masculinities be maintained if women also stigmatise and humiliate men and boys who deviate from traditional masculine norms (Connell 1987, 1995; Connell and Messerschmidt 2005; Jewkes and Morrell 2010)? What if changed men return home to practice increased gender equality only to find that women draw on ‘ideas of the legitimacy of male superiority’, ‘demonstrating complicity, constructing forms of femininity which accept male domination’ (Jewkes and Morrell 2012, 31)? Further, can boys grow to be gender-equitable men if their mothers (and fathers) socialise them to be unemotional, tough at all costs, sexually aggressive and strong? Overall then, it may be the case that gender-transformative health interventions, by primarily intervening with men, have reinforced not only a binary analysis, but a single-sex analysis and intervention model, replicating some of the limitations found when global health programming is only carried out with women (Dworkin et al. 2011; Dworkin 2015).

Transforming gender-transformative interventions: next steps and a call to action

In the two decades since the ICPD conference, the global health field has made tremendous strides in how to address relations of gender inequality and masculinities to improve health. Above, we have highlighted some of the challenges and limitations of gender-transformative interventions with the aim of thinking towards the future and improving our efforts. Below, we highlight four key areas to focus on in the coming decade.

First, it is critical that the field of global public health continues to discuss with, collaborate with and learn from social science researchers. Because of progress on the social science of gender in the past few decades, we now view and understand gender as relational and as a set of patterned and collective social practices. It is important that gender-transformative interventions capitalise on this understanding by incorporating both men and women into their programming. This can facilitate change by having women and men work together to meaningfully change the expectations that men and women have for each other. Additionally, in our own research, men reported that they benefitted a great deal when groups of women and men were joined together in gender-transformative programming. Men reported that they liked hearing the perspectives of women, that women's views were appreciated and that women ‘kept the men in the group honest’ about the changes that were – or were not – occurring in gender relations at the individual, household and community level. While allowing men safe spaces in which to openly discuss these complex issues is important, ultimately the renegotiation of gender relations will also require future programmes to test the impacts of simultaneously working with both women and men (Dworkin 2015; Dworkin et al. 2011; Pulerwitz et al. 2010).

Second, we have identified the importance of working towards structural and community-level change, rather than simply focusing on individuals. While we recognise that changing individuals has the potential to shift social structures, the field will also advance if it embraces an understanding that structural and community-level change can facilitate individual-level change. Intervention developers should further consider social institutions and policies that can promote community and societal-level shifts. For example, the national secondary school curriculum in South Africa has been modified to include concepts of power, masculinity, femininity, gender role stereotypes and gender inequality (WHO 2010). Targeting institutional policies – particularly masculinist social institutions such as military, the police force and sport – can more firmly establish an environment that is supportive of gender equality (Barker et al. 2010b). In Liberia, the national military has purposefully integrated male and female soldiers and trained them in ‘gender politics’ in an effort to overcome past abuses by the military (Blunt 2006). In additional to institutional and policy change, grassroots social movements facilitated by community organising have the potential to change social structures by pressuring elites and policy-makers to enact changes that diminish inequalities. These social movements can be transformative for gender relations and health and lead to larger societal shifts. Thus, these types of institutional policies and social movements, in conjunction with traditional gender-transformative programmes, have the potential to entrench more gender equitable norms in societies.

Third, while ‘gender-transformative’ has become the gold standard for many global health intervention programmes with a focus on gender, we hope that the next generation of health programming no longer limits itself to a focus on gender. It is increasingly evident that individuals are shaped by multiple identities and inequalities and that their experience of gender intersects with these social positions (e.g. class, race). To truly adopt this intersectional perspective, future interventions should not just aim to make parallels between racial and gender inequalities in programme content as was suggested in the last section of the paper, but also might aim to be transformative in race and class structures and identities. The MEN Count intervention (described previously) demonstrates that a combination of strategies addressing masculinities, race and class may be more effective at transforming men's lives, behaviours and attitudes towards gender equality. In addition, gender-transformative work of the future can work to be less focused on cisgender men and heteronormative notions of gender and sexuality, and make conceptual use of the full possibilities implied by the terms ‘sex’ and ‘gender’.

Gender-transformative interventions have taught us that public health programmes can successfully chip away at entrenched social norms and organisation. But, by limiting ourselves to gender-transformative we are failing to fully appreciate the intersection between other social identities and structures that pattern social interactions and health outcomes. By recognising the interconnectedness of these important social structures, interventions can synergistically tackle social problems whose roots lie in gender, class and racial inequalities.

Fourth, we need to improve the rigour of our evaluation of gender-transformative interventions. It is essential that future work utilises community randomised control trials, the gold-standard evidence for these types of interventions. Currently, programmes evaluate interventions using measures of men's attitudes towards gender norms (e.g. the GEM Scale) or a health outcome such as violence perpetration. To improve evaluations in the future, we need to continue improving measures of men's attitudes, behaviours and social practices, developing new scales that measure the full range of our understanding of gender and masculinity. We also need to recognise that a focus on a single health outcome is limiting due to the fact that masculinities and gender relations affect a broad range of behaviours and health outcomes (Baker et al. 2014; Courtenay 2000a). The global public health field would also benefit from the utilisation of qualitative research and process evaluations to more deeply understand how interventions are working, what the unintended consequences are (if any) and to more deeply understand what the mechanisms are that account for change.

Conclusion

The development and implementation of gender-transformative health programming with men has given the field of global public health unprecedented evidence-based tools to engage men and boys in working toward gender equality. The lessons learned from the field thus far can enable researchers, practitioners and programme implementers to modify gender-transformative work in ways described above so as to maximise the health and wellbeing of men, women, boys and girls, both domestically and globally.

Disclosure of conflict of interest

The authors have no conflicts of interests to declare pertaining to this work.

Acknowledgements

The content is solely the responsibility of the authors and does not necessarily represent the official views of the US National Institutes of Health.

Additional information

Funding

Paul J. Fleming was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant number T32 HD007168) and National Institute of Allergy and Infectious Diseases (grant number T32 AI007001). Chris Colvin was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health (grant number R24HD077976). Shari L. Dworkin was supported by a National Institute of Health grant to the UCSF Gladstone Institute of Virology & Immunology (Center for AIDS Research, grant number P30-AI027763) and by a UCSF School of Nursing intramural grant.