Evidence Synthesis - Evidence Review Team 4: Gender Dynamics

Noting that the roles of both women and men in preventing child deaths (e.g., behavioural outcomes in family planning, health behaviours and service use for safer delivery, and immunising children) are well documented, this paper: defines key terms related to gender, describes gender-integrated interventions, and reviews evidence on the effects of gender interventions on health behaviours related to reducing preventable child deaths. It emerges from an initiative of the United States Agency for International Development (USAID), which in 2013 convened a team of experts to conduct a systematic review of the evidence on a range of topics. In order to conduct this review, evidence review team 4 (ERT4) explored 21 documents and descriptions of programmes from 14 literature reviews in order to develop a conceptual framework and provide a synthesis of the evidence.

The paper differentiates programmes along a gender continuum:

• from exploitative - interventions that "take advantage" of harmful norms as a means of achieving hoped for outcomes - e.g., a campaign seeking to increase women's uptake of prevention of mother-to-child transmission (PMTCT) of HIV services by incorporating messages that promote guilt ("What kind of mother would give HIV to her baby?")
• to accommodating - interventions that recognise but do not seek to change gender norms, dynamics, or inequalities in their implementation - e.g., an initiative that used video and in-person discussions with providers on the benefits of exclusive breastfeeding and encouraged men to help women with chores during the first 6 post-partum months.
• and, ultimately, transformative - interventions that actively examine and promote the transformation of harmful, rigid masculine and feminine norms and/or address gender inequalities such as imbalances in power and decision making in intimate relationships. "While increasing, evidence around this approach is still relatively scarce. This review illuminated a number of challenges related to implementing and measuring gender transformative interventions aimed at improving child mortality and development outcomes..."

Building on definitions of gender and the gender continuum for interventions, the paper presents results separately for interventions that:

1. seek to empower women - Of 16 interventions identified through this research, some were designed to reach adolescents and others reached out to older women (i.e., typically at least 20 years old and typically married). The meta-analysis of these efforts to address maternal and/or child health included 7 intervention trials, conducted in Asia (n=6) and Africa (n=1), which followed a 4-phase participatory action cycle that gathered women together to meet for several months, often led by a trained facilitator, to: (i) identify and prioritise problems; (ii) plan actions; (iii) implement locally feasible strategies; and (iv) assess activities. In addition, some of the interventions included peer counseling. Over the 7 trials, exposure to the participatory action groups was associated with a 37% reduction in maternal mortality, a 23% reduction in neonatal mortality, but no significant reduction in stillbirths. Because reductions in mortality were linearly related to the proportion of pregnant women in the groups, the authors conducted sub-group analyses, which showed larger significant reductions in maternal (55%) and neonatal (33%) mortality when at least 30% of pregnant women participated in the group. That said, the authors omitted half of the data from one of the trials, and effects on behavioural outcomes that contribute to mortality (e.g., clean delivery, institutional delivery, breastfeeding) were mixed - raising questions about the mechanism of effects. The participatory group intervention addressing malaria was associated with significantly larger increases in treated net usage in the intervention compared to the comparison villages at the study's conclusion.
2. specifically reach out to or include men - The literature reviews and literature search uncovered 27 different interventions that sought to engage men in women's service use or behaviour change to improve family planning (n=17), or maternal health or early child health (n=10). Sample set of findings: Two of the gender transformative interventions with positive effects and moderate to strong designs were rather intensive (e.g., workshops or visits with a male motivator/peer educator over 6 months) that provided information, addressed gender norms and male responsibility, and encouraged men to communicate with their partners and make joint decisions about family planning. One intervention was more focused on HIV, but described and promoted condoms as a dual protection strategy (this was the only measure of family planning use). Both interventions showed effects on condom use, and the study that measured family planning use showed a significantly larger increase in family planning use in the intervention group relative to the comparison group.
3. use a synchronised approach - The paper describes 5 interventions that incorporated activities designed to reach women, men, and wider communities to improve pregnancy care and maternal and early child health outcomes. Three of the interventions targeted married adolescents and two targeted older (i.e., 20 and older) men and women. For example, a long-term project in Ghana (The Navrongo Community Health and Family Planning project) had 2 main components: (1) modified the way services were provided (e.g., enhanced training for and incorporated community nursing) and (2) mobilised the community, including: establishing village health committees in collaboration with traditional male leaders, trying to incorporate women or women's views (e.g., about family planning) into leadership communications, appointing male community health workers to go door to door to talk about health issues, making referrals and providing some basic medicines. Although short-term evaluations suggested an increase in family planning use, a follow-up assessment after the project had gone to scale suggested that effects on family planning were not sustained - in part because community mobilisation efforts were not maintained.

Summary of the evidence:

• There are few well-designed studies to evaluate gender interventions, but a few moderate-to-strong quasi-experimental and randomised trials show that some approaches are effective. For instance, women's participatory action groups can reduce maternal mortality and morbidity if there is a threshold of pregnant women in the groups.
• Moderate-to-strong quasi-experimental and randomised trials show mixed effects on interventions encouraging men to support women's and children's health or attempting to change gender norms or increase communication/joint decision making.
• Male and/or couple education can increase family planning use, participation in pregnancy planning, and breastfeeding, but appears to have fewer effects on the number of antenatal care (ANC) visits and hospital delivery.
• Although few in number, there have been mixed effects of interventions that address gender norms, communication, and decision-making with men or with couples (both addressed family planning and both also worked to increase access to family planning).

Recommendations for policy and practice include:

1. Women's participatory action groups can reduce maternal mortality and morbidity if there is a threshold of pregnant women in the groups.
2. Male (or other) community-based distributors of family planning services to husbands and wives in the household can increase family planning use.
3. It would be helpful to assess replicability and determine parameters to go to scale for interventions that have a strong design and show positive effects (i.e., women's participatory action groups and community-based distribution).

Recommendations for research include:

1. More research on gender-transformative interventions is needed.
2. To capture the effects of multi-component interventions (e.g., synchronised interventions), different evaluation methods may be needed (e.g., multi-method designs with "triangulation" of data).
3. More research is needed to understand the ways in which norms (e.g., norms around the mother's and father's role in child health) and gender inequalities shape maternal and child health behaviours and service use.
4. More rigorous comparative analysis is needed to strengthen the assertion that programmes that include men and women in a gender synchronous manner result in more effective than reaching out to men or women alone.

Click here for the 23-page report in PDF format.