The effect of a community based health promotion intervention to change gender norms among women in a rural community in Sri Lanka

Gender Based Violence (GBV) means any act of violence occurring based on gender that results in, or is likely to result in physical, sexual, psychological or economic harm or suffering for women, including threats or such acts, coercion or arbitrary deprivations of liberty, whether occurring in public or private life [1]. Although, this definition is applicable to both men and women, the phenomenon of GBV mostly effects women [2]. This paper, therefore addresses the issue of violence against women subjected to due to gender by men, known as GBV. The most pervasive form of gender violence is abuse of women by intimate male partners, and this is known as intimate partner violence or domestic violence for this study.

GBV is a major public health issue around the world irrespective of income levels [2, 3]. Women are vulnerable to different types of physical, sexual and psychological violence at different stages of their lives within the family, community and society [4].Globally, 10–60% of women who have ever been married or partnered have experienced at least one incident of physical violence from a current or former intimate partner [5]. In a Sri Lankan study, women attending antenatal and gynecological clinics of a hospital in rural Anuradhapura were screened for domestic violence in 2002, it was found that 40% of them had experienced some form of GBV during their life [6]. Similarly, a survey found 40.7% of women who attended an outpatient department of the teaching hospital, Ragama, in a semi-urban area in the suburbs of Colombo were abused by their male partners [6]. As reported by World Health Organization (WHO), in Sri Lanka GBV ranges from sexual harassment in public places including public transport to acts of violence within the privacy of the home or work place [7].

Gender inequality and discrimination are defined as root causes of violence against women [8]. Evidence suggests that gender inequalities also increase the risk of violence by men against women and inhibit the ability of those affected to seek protection. Further studies have shown that violence against women and girls is related to social norms that prescribe male and female roles in society and condone abuse. Addressing of these gender norms is an important investment to combat the growing issue of violence against women [7‐13].

Changing gender norms has demonstrated positive results in combating the issue globally according to the literature [10‐19]. Most of these studies were aimed at young couples, boys and men. However, the methods of changing gender norms were limited to interventions such as educational programs, skill development of healthy relationships and public information campaigns. Commonly, these interventions were centralized around individuals. But there is evidence from previous studies that it is important to have coordinated actions at the community level for an effective primary prevention [20].

Many researchers have discussed GBV as a learned social behavior for both men and women [20]. This intergenerational cycle of violence persists because of the traditional social attitudes about women, men and violence. Social traditions have placed women as a housekeeper and a child bearer. Tragedy doesn’t end there, as for women deviation from such roles will bring violence as a punishment [21]. According to previous research, males are perceived as macho beings and aggression as an important macho trait [22]. Further, it is explained that, in South Asian societies, women consider themselves ineffective without males and males should dominate the family. And research has pointed out that husbands try to control wives through violence until they surrender [23, 24].

At present, there is no systematic mechanism for data collection in relation to the prevalence, causes and consequences of GBV in Sri Lanka; however, Sri Lanka has signed all important global declarations and created national policies on prevention of GBV. The Domestic Violence Act of Sri Lanka was passed in 2005, primarily as a response to domestic violence incidents. Also, there is a paucity of research evidence on the most common gender norms that trigger violence in Sri Lanka. However, in patriarchal Sri Lanka, motherhood and marriage are still social norms for Sri Lankan women [25]. And it represents the gender norms and perceptions instilled in Sri Lankan society, which perpetrates not only overt violence but also norms that promote men’s extreme control over many aspects of women’s lives. Gender norms related to motherhood such as caring and nurturing children and preparing food for the family are typical in Sri Lanka [26]. Hence, even minor deviations would trigger violent behavior by males. Existence of masculine attributes associated with money, problem solving, consuming alcohol and tobacco and engaging in household chores bring more value to male control over women [25].

In the context of Sri Lanka, there are preventive measures such as distributing information packages on GBV prevention for newly married couples and for migrant workers particularly travelling to the Middle East countries. In addition, improving the capacity of primary health care workers for detection and support was also implemented by the health sector. But the effectiveness of these measures on reducing GBV was not established due to the lack of scientific evidence. According to a study, the existing literature in the Sri Lankan context is limited to either prevalence studies or literature reviews on domestic and gender violence [27]. But most of the published global and local studies have recommended the need for community based interventions that ensure sustainable outcomes instead of only improving knowledge and attitudes [7, 27]. A key finding of these studies is that targeting women and girls is a sound investment. Women and their families are the main agents of perpetuating gender stereotyping to the next generation, and these agents determine the continuation and prevalence of the existing unhealthy gender norms that contribute to trigger violence when there is a transgression contrary to the norm [28]. In the context of Sri Lanka, women with children under five years are relatively easy to access as they attend field weighing posts conducted by public health field staff under the growth monitoring program of children.

The health promotion approach has been widely recognized as a process of enabling individuals and communities to increase control over their health and improve health status [29]. Health promotion works through concrete and effective community actions to set priorities, make decisions, plan strategies and implement them to achieve better health [30]. The sustainability of such an approach is largely due to increased community ownership.

Hence, it will be worthwhile to check whether, or how, women will take over this approach in preventing GBV. Even though the women are the most common victim of GBV, it remains unclear how this vulnerable group can be strengthened to prevent GBV despite the search for an effective preventive method. This paper presents a study conducted to assess the effectiveness of a community based health promotion intervention for changing women’s acceptance of gender norms in rural villages in Sri Lanka.

The intervention was considered a health promotion intervention as it has a design based on the principles of health promotion. Thus, this paper presents an arena to discuss results of such an intervention.

Our findings indicate that the health promotion intervention developed for this study successfully changed the acceptance of gender norms among women in this community. Prior to the intervention, acceptance on selected gender norms among women was high. But the post evaluation has shown that participants in the intervention group have less acceptance of most of the gender norms (5 out of 6), whereas no similar change was observed in the control group. This finding is consistent with other studies of gender norms changing interventions irrespective of the mode of intervention, whether it is a social, economic or combined [19, 37‐41].

This study measured the acceptance of prominent gender norms that were viewed to ‘trigger’ violence against women in Sri Lanka. Perceptions of being a good wife were identified by several past Sri Lankan studies [35, 42], and these provided solid support for the selected norms by explaining masculine attributes in a patriarchal society. According to these, the family expects wives to continue norms related to motherhood and marriage. Further, women also believe adherence to such norms will protect them against violence and at the same time will uphold the reputation of the family. Studies from throughout the world also identified events that represent transgressions of dominant gender norms [4], such as a wife not obeying her husband, talking back, not having food ready on time, failing to care adequately for the children or home, questioning him about money or girlfriends, going somewhere without his permission and refusing him sex. When considering the dominant gender norms that trigger violence, there were more than 70% of Sri Lankan women who totally agreed with both norms “Women are more responsible for nurturing and caring for children than men” and “Women are more frugal than men”. There were more than 74% women who agreed that “Women must know more household activities than men”. This acceptance denotes the traditional division of work among males and females in a Sri Lankan context, and the current study results are comparable with other research evidence in the world on incidents that trigger violence [35, 42]. Partner’s alcohol consumption was identified as a possible risk factor for violence. Behaviors associated with substance abuse are more acceptable in South Asian communities [43]. The majority of women in this study, in contrast, were taught that consuming alcohol or tobacco isn’t an attribute of masculinity.

The gender norm that “Behaving violently to a problem is a masculine way” didn’t change according to the findings. This norm is likely to be the most rooted gender norm among female communities as it is evident in the global studies as well [4]. According to a previous study conducted in Sri Lanka, there is an association between wives’ attitudes toward gender roles and intimate partner violence [42]. The wives were less likely to experience current abuse by husbands if they believed traditional gender norms. The influential nature of the norm may be a reason for not being successful in changing the above norm.

Apart from quantitative measures, we were able to explore changes from a qualitative perspective, which is one of the strengths of the study. Behavioral changes identified from the FGDs would be the strongest indicators of empowering women to put them in a position to be agents of change within the family and community. The claims of getting increased support from their husbands for domestic work and even for child caring, despite the negative remarks from in-laws, also indicated that these women could come out of the prescribed conventional gender role.

As an implication, these types of interventions can be modified by adapting validated ongoing monitoring systems to track changes in the process, in order to measure behavioral changes in a quantitative basis. But in reality there are no validated instruments to track such changes. So, further research is needed on this regard in a Sri Lankan context. Based on these results, further modified interventional research will be needed to change the unchanged norms as well to accelerate the process to change the behaviors regarding other norms.

GBV is a complex phenomenon shaped by antecedents operating at different levels, and gender plays a fundamental role. According to studies, this has important implications for the design of interventions to address GBV [4]. A recent review highlighted that it is easier to increase awareness and modify attitudes than to change violent behavior [16]. Similarly, this study also dealt with perceptions as the major priority while targeting behaviors as an indirect result. Furthermore, according to this review community wide interventions are required to prevent violence and these should address social and cultural factors at the community level. According to several studies, mobilizing communities by targeting both young males and females will be a key in reducing levels of violence [7, 8, 15, 17]. This study also focused on women who have children under five years of age. In Sri Lanka, this group of women are more likely to be enthusiastic of their well-being. Therefore, it is necessary to ensure the adaptability of the current intervention for other women groups in general. Also, many researchers have pointed out that rural sector is more prone to have GBV [16] and, hence, this study was conducted with a rural setup. So, applying the results to an urban setting might require modifications.

Though a number of successful projects have applied community based approaches for changing gender norms, there is limited evidence to date for utilizing a health promotion approach to mobilize communities to use the concept of community empowerment to change behaviors. According to the literature, most violence prevention programs for empowering women and girls in low and middle income countries (LMIC) consist of a series of educational workshops to ensure participatory group training [44]. Furthermore, these interventions were embedded in programs that aimed to improve sexual and reproductive health and livelihood programs, such as microfinance or vocational training. Irrespective of the aim, similar to our study, addressing masculine and feminine attributes was the goal. Moreover, violence prevention programs in LMIC also used approaches such as social communication and community mobilization. And many intervention models function in cooperation with HIV prevention programs. Hence, some studies used integrated approaches to combat gender norms as a step to reduce the prevalence of GBV. As a result, past interventions ranged from social, economic or combined interventions [40].

It was also evident from the literature review that most of the studies reported findings in relation to violence and rarely described changes in gender norms quantitatively, though gender roles were the targeted areas to change. Group training interventions for women and girls were conducted by several researchers and most of the studies were conducted quasi-experimental studies. These interventions consisted of life skills courses [45], empowerment and self-defense programs [46]. All studies demonstrated reduced violence against women after the interventions. In contrast, group training interventions were conducted with women and men through a 50 h training that included addressing gender norms [47]. But, results weren’t consistent for both women and men. Furthermore, there was a reduction in violence by community mobilized interventions for both women and men through educational programs [19, 48] and interventions based on psychological models [49]. According to our literature search, health was included in only one study as a theme [48]. Even though these studies presented results in relation to reduction in violence, the extent to which the gender norms were changed by interventions was not evaluated. In contrast, our study centralized around presenting the extent of changing gender norms via a health promotional intervention. As a secondary step we investigated the behavioral changes in relation to gender norms after the intervention. Therefore, there were considerable differences in between our study and past studies. Notably, our study was not a total educational program but an intervention based on a health promotion approach that was scaled up from knowledge improvement sessions to community planning sessions. Different health topics were used in cooperation to achieve the defined change in relation to each LFW step in our study.

Despite of the results achieved by most of the past studies, they lacked a proper theory and conceptualization or LFWs that could be used to explain the changes gained [14]. By contrast, this study was conducted according to the health promotion principles including a community based strategy involving women, community empowerment, community led mechanisms to maintain a high follow up rate, addressing UCD and incorporation of gender norms to cover all domains to address more than one type of GBV. Moreover according to the LFW, the acceptance of five norms among women were changed due to the change in their acceptance of the norms, starting question their basis and finally starting to question underlying factors of acceptance of gender roles. Development of the LFW made it easier for the PI to facilitate women to achieve the above changes in due time.

This community empowerment concept through a health promotion approach is novel to Sri Lanka. Hence, the findings were not directly comparable with previous studies conducted in Sri Lanka as no study had been done to change the acceptance of gender roles among women in Sri Lanka. This study found five easily changeable gender norms out of six prominent gender norms in Sri Lanka through an intervention that used the health promotion approach. In comparison, this is one of the first quasi-experimental studies to explain the behavior of each single gender norm after facing an intervention while other studies explained the change as a whole.

In addition to providing new data, our study also addressed several limitations identified by past literature related to interventional research on changing gender norms to date. The study included a comparison group, concern on gender norms related to various aspects and types of GBV, initiatives to maintain follow up and theory based designing of the intervention. Though the present study did not measure the impact of the initiatives on violence against women, this study collected information on supportive behaviors to reduce the problem. Furthermore, standardized or validated measures using a locally validated data collection tool was used. In addition, most of the past studies were not conducted in real communities, as past studies were conducted in external places such as in a training center [7, 8]. But the present study was conducted in real settings, which increased the ability to address real community determinants in their daily lives by improving skills among women to deal with them. A community initiative for males and females, designed to challenge gender norms and prevent violence against women and children in Uganda [37], reported that after two years following the program all forms of intimate partner violence had decreased in the community. Similarly, current research can also scale up towards a reduction of GBV incidences in a rural sector in Sri Lanka. Although evidence from evaluations in other contexts suggests that interactive group education is a key component of this type of intervention and may be necessary to sufficiently influence often deep-seated and complex gender related norms [15], the present study used more than interactive group education but also created a community driven process to convert knowledge into practice. Additionally, according to past studies economic empowerment of women was associated with risk of increasing violence. But, in our study such issues were not reported. The likely reason might be that in this intervention targeted to improve family well-being the active participation for the interventions was from women. It did not target individual level empowerment on challenging power relations but empowering women to include the family in the intervention [50, 51].

Limitations of this study include small sample size, inability to generalize findings to whole districts because of the selection of one village within it and selection of women who have a child under 5 years as the study population. Apart from the above, our study didn’t use the gender equitable men scale for gender interventions. In some similar studies they have use this scale to measure outcomes [19]. This study was too short to measure any long term changes in relation to behavior following the changes of acceptances of norms. Though, this study used only six norms, this study provides a methodology that might work with other norms as well.

Our study raises recommendations for the development of community based health promotion approaches and appropriate measures to determine their effectiveness. This study can be used as evidence for future research projects and policy measures that these kinds of initiations should be started at grassroots level health centers and grassroots level health officers should take the lead in reversing gender norms in the next generation.

The findings from this study concluded that a community based health promotion intervention targeting gender norms can be used effectively to address the acceptance of gender roles among women in rural villages in Sri Lanka. These results and methodology can be applied to other communities with similar socioeconomic and demographic characteristics.