Intimate partner violence (IPV) is a global issue that is estimated to affect around 30 percent of women during their lifetime. In sub-Saharan Africa this problem is compounded by the risk of contracting HIV. Whilst these two issues have been addressed separately, they are known to be closely linked. Interventions targeting individuals have sought to address both of these public health burdens with varying degrees of success. Now in a study in BMC Medicine, Charlotte Watts from the London School of Hygiene and Tropical Medicine, UK, Lori Michau from Raising Voices, Uganda, and colleagues present the results of the first trial in sub-Saharan Africa to employ a community based intervention to reduce IPV and the associated risk of HIV, called the SASA! Study. Here Watts and Michau discuss why research into IPV has been neglected, the challenges and benefits of community based interventions in this field, and what’s next for the SASA! Study.
Why is intimate partner violence (IPV) such a big problem in sub-Saharan Africa? What are the broader consequences?
Intimate partner violence is a big problem all over the world, not only in sub-Saharan Africa – global figures suggest that 30 percent of ever partnered women will be physically or sexually assaulted in their lifetime. What is striking in sub-Saharan Africa, and several other developing country settings, is that the past year levels of violence are also high. In our study community, for example, about a quarter of women had experienced physical or sexual violence from a partner in the past year.
The health consequences of violence are multiple, with both short and long term effects. As well as injury, women in violent relationships are more likely to suffer from poor mental health, and be at greater risk of having an unwanted pregnancy, sexually transmitted infections, and HIV. More broadly, violence within intimate partnerships can have profound impacts on children’s well-being and development, and limit women’s ability to engage in economic development activities.
Why has IPV received little attention from the research community?
For many years gender based violence, including intimate partner violence and sexual violence, were seen as being too sensitive and hidden an issue to be researched. People thought that violence was too private an issue to explore, and they were worried that population surveys could not be done without putting women at increased risk. Things are now changing, and we have strong guidance on how to conduct research in an ethically responsibly way, and we have learnt that if interviewers are carefully selected and trained, and interviews are conducted in a private setting, women do talk about the violence in their lives. There is now a fairly large body of population based evidence on the prevalence of intimate partner violence around the world, and a growing body of evidence on its health impacts.
The field of intervention evaluation is still in its relative infancy, although the evidence that does exist is promising, and shows that violence against women is preventable. It can be difficult to obtain funding for intervention trials – the SASA! trial had five donors, for example, and was small (with only eight clusters), due to constraints in available resources. As the field grows I hope that more mainstream research bodies will start to support research in this area.
What is the SASA! Study and what are its key findings? Were you surprised by any of the results?
The SASA study is a cluster randomised controlled trial, designed to assess the community level impact of SASA! – a community focused violence prevention programme. SASA! was designed to change the social norms and behaviours that underpin both violence against women and HIV related risk behaviours. Discussions and one-on-one teaching activities are delivered by community members (local leaders, and male and female activists) who are supported by staff at the Centre for Domestic Violence Prevention in Uganda, to support communities to think more about the causes and consequences of violence, the impact of this violence, and the ways in which individuals and communities can change.
The findings are exciting, suggesting that the intervention achieved significant impacts on a number of the primary outcomes over a relatively short timeframe. Impacts included significant reductions in community acceptance of violence, and significant reductions in multiple sexual partnerships among men. Women’s experience of physical violence were also 52 percent lower, although due to the small number of clusters, the findings were not statistically significant.
What I found surprising was that when we compared the intention to treat and per-protocol findings, in general the effect size estimates did not change substantially (although the physical violence outcome achieved borderline significance). For me it was interesting that even when people had not had direct contact with intervention activities, the same degree of change had occurred. To me this provides direct support of the value of the social diffusion model of intervention that SASA! uses.
What are the barriers to implementing community based interventions for IPV, and how can they be overcome?
Programming at a community level is complex – involving multiple stakeholders, institutions, opinions and opportunities. For a meaningful change in social norms, programming must be guided by a theory of change – a hypothesis of how change will happen over the timeframe of the intervention sequenced in an appropriate manner such that it does not get to the ‘ask’ regarding behaviour change, too early or too late. SASA! requires longer-term investment in communities through a systematic process of change; yet for many organisations, prevention programming is often thought of as a few community dramas or posters hung around town. Reorientation to a different way of engaging communities is needed.
IPV prevention activities are most effective when led by trusted others. Effective social norm change efforts require on-going exposure of ideas and are best delivered by a trusted and known persons rather than an external implementer. This is an unfamiliar way of working for many organisations. In the SASA! methodology activities are led by community members themselves and happen on a daily basis – not as part of a large public event but rather within the context of the daily lives of community members (while collecting water, waiting at the bus stand, shopping in the market, at the local drinking joints, etc). This often requires a large shift in how organisations think of programming; moving from non-governmental organisation (NGO)-led to community-led, sporadic activities to consistent presence in communities, message-based activities to encouraging critical thinking, phasing in ideas. Community mobilisation done well will take on a life of its own, with diverse individuals and groups making the issue ‘their own’. This is a marker of success but also not business as usual for organisations who are used to more controlled efforts.
Working in mixed sex groups is also important. It can be challenging to work with women and men together – rather than single sex groups – for a variety of reasons including women’s understandable reluctance to voice their opinion in the presence of men, and the need for small, single sex groups to work through identity and personal experiences. There is no hard and fast rule for this and when carrying out IPV prevention, care must be taken in how issues are raised and discussed. In SASA! there are both single and mixed sex groups. Many organisations find it difficult to ‘get men’ to activities. Often these activities are announced as being on violence against women – naturally, many men do not appear. With SASA!, community activists who are in equal number male and female go out to reach their own social networks and neighbours. Rather than calling people to activities, the community activists go to them, for example, meeting men at the local drinking joint, pool table, carpentry shop or garage (the same is done for women). In this way, the ideas come to people through their daily work and movements and while they are with their peer groups. This approach can be very powerful in shifting social norms as collectively, small groups of men and women discuss and work through these issues. This creates much more ‘stickiness’ for change as the peer group will reinforce and hold members accountable to new norms.
What are the long term benefits of community based interventions for IPV?
There are direct and indirect benefits for individuals, communities and the wider society when a community mobilisation, social norm change approach like SASA! is used. For individuals, the SASA! focus on power (power within, power over, power with, power to) opens often very marginalised people to new ways of thinking about their own power in their families and communities. Time and time again, we hear from community members that their perceptions of themselves have changed – from feeling powerless to recognising that while not all powerful, each of us can make decisions everyday with a variety of people about how we use our power. This has been transformative for many communities and individuals and has an effect on all spheres of a person’s life.
For families, women and men talk about a new respect, softness and kindness in their relationships – from where there was suspicion, control and fear, to the recognition that families and relationships can be stronger when the dignity and rights of the other are upheld. Men comment that there is more harmony in their homes and so they feel less compulsion to go out and look for other partners. Women talk about how they feel more respected and listened to and now can contribute to the family decisions and finances. Children share stories of relief about how their homes are much nicer places to be, with parents who they are not afraid of or afraid for.
In the community, when social norms about the value and worth of women shift opportunities open up for women’s participation, their voices can be heard in community meetings, they come together to work on issues important to them. We have also found that once women and men feel a sense of activism (taking action on an issue because it is important to you), this activism spreads beyond IPV to other community issues such as sanitation, water, political participation, etc. Feeling a greater sense of power means there is much more activism broadly in the community. This bodes well for positive change across the development spectrum.
SASA! is being rolled out across a number of countries in sub-Saharan Africa. What are the challenges in implementing the intervention in different locations?
SASA! is currently being used by over 30 organisations in the Horn, East and Southern Africa, including international NGOs, NGOs, community-based organisations and faith-based institutions. Each organisation is unique, as is each community. SASA! is being used in densely populated urban ‘slums’, rural villages, pastoralist communities, refugee camps and settlements, as well as within the structures of the Catholic church. SASA! can be adapted based on the skills, context and culture of a community and an organisation. We’ve come to learn that because SASA! works on the core driver of violence against women – power inequality – rather than the manifestations of violence, it is largely applicable in settings which are very different. However challenges remain, including:
a) Knowledge of community characteristics. Effective SASA! implementation requires deep understanding of the community – both geographically and socially. SASA! needs to be adapted to suit each community because every community is unique. SASA! is not a blueprint for an intervention but an overarching methodology that provides a basis for local adaptation. For example, in pastoralist communities the market days become important gatherings and spaces for dialogue in this otherwise diffuse community, in the Catholic church rather than setting up parallel structures the challenge is to tap into all the existing opportunities the complex infrastructure of the church provides. Likewise, in some refugee communities, rather than printing posters for hanging in homes and public spaces it can be more effective to print small, carrying size images as many homes are temporary structures with extremely limited wall space.
b) Anxiety, backlash and scepticism from the community. Raising issues of gender, power and intimate relationships is sensitive and requires care. Even though SASA! uses a benefits-based approach (i.e. demonstrating the benefits of non-violence for all rather than blaming and shaming those using or experiencing violence), there can be anxiety from women and men about what talking about these topics will mean. There can be backlash from both women and men who are resistant to changing the status quo. There can also be scepticism from women and men that programming will truly benefit them and include them – rather than the organisation leading the programming. All of these can be overcome through a variety of ways of working in the community including: a) ensuring that community members (women and men) are leading the activities, b) transparent processes of information giving and selection of activists when the work is beginning, c) quality training of the community activists before they begin activities in the community as well as ongoing support to help them expect and be able to constructively respond to possible community reactions, d) positive, benefits-based programming, e) commitment from the organisation that programming will be consistent and regular – this is essential in moving from project-based programming to fostering community activism.
What further research is required in this field?
We need research in many different areas. We need to learn more about how to achieve impacts at a population level. Much of the current intervention evidence focuses on assessing the direct impact of interventions on programme recipients. Given the high prevalence of violence globally, it is important to move to evaluating interventions that are seeking to achieve more widespread change, and to identify potential entry platforms that could be used to deliver effective interventions at scale.
Questions from Joanna Denyer, Senior Assistant Editor for BMC Medicine.