Optimising Maternal and Perinatal Death Surveillance and Response to prevent future deaths

Research question: Can perinatal and maternal deaths be reduced by implementation of a customised ‘Toolbox’ of interventions to optimise Maternal and Perinatal Death Surveillance and Response (MPDSR) in LMICs?

Background: When implemented optimally, MPDSR has been shown to reduce maternal mortality by 35% and perinatal mortality by 30%. However, documented barriers to implementation are frequent: - Lack of community involvement: under-reporting of community deaths, infrequent community-level investigation & review

- Fear of blame / litigation among health care providers, deterring engagement / learning - Poor quality data, hampering identification of modifiable factors - Insufficient training at all levels - Inadequate implementation of responses - Lack of monitoring / accountability Aims - To develop and optimise intervention components to improve functionality of MPDSR

- To evaluate the effect of a customised intervention (optimising MDPSR) on maternal and perinatal mortality. Specific objectives 1. To review existing tools, develop and adapt a harmonised toolbox of resources to optimise MPDSR implementation 2. Codesign customised district-level “intervention packages” using the toolbox

3. Evaluate effectiveness and cost-effectiveness of “customised intervention packages” to reduce perinatal and maternal mortality Methods and timelines

In Ethiopia, Ghana and Uganda, Community Advisory Boards and Stakeholder Committees will be established and meet regularly to advise on all aspects of the project (WP1.1, months 1-48). Political Economic Analysis will be conducted (WP 1.2, m1-12).

Phase 1 (WP 2-6, m1-12): Co-development of a comprehensive “toolbox” of resources for adaptation to specific country contexts, including tools for:

- Reviewing, then proposing amendment / development of laws and regulatory frameworks, to facilitate an enabling regulatory environment for MPDSR and legal accountability for implementing responses at all levels (WP2) - Improving data quality (WP3) - Training and mentoring health workers across specific MPDSR roles (WP4)

- Improving MPDSR response-tracking and budgeting (WP5) - Community death reporting, investigation and review (WP6) - Counselling bereaved families (WP6)

Phase 2 (WP7, m12-18): Developing customised “intervention packages”. Each country will hold stakeholder workshops to select, adapt and translate “toolbox” components to ensure they align with national policies, capacity, socio-economic context, and proportion of deaths in health facilities and outside.

Phase 3 (WP8, m18-48): stepped-wedge cluster-randomised trial to assess effectiveness and cost-effectiveness of “customised intervention packages” in 9 districts (3 in each country, implemented at months 18, 24 and 30), with perinatal mortality as primary outcome. To ensure that mortality data is reported reliably and consistently, we will also implement improved tools for death reporting in all sites from month 12. Outcome data will be collected up to month 42.

Anticipated impact and dissemination: The generic “toolbox” of interventions will be disseminated globally by the World Health Organisation (WHO). We hypothesise that its implementation would optimise MPDSR, address current challenges including action on recommendations and thereby reduce maternal and perinatal mortality. Lessons learned from implementing “customised intervention packages” will inform scale-up in each country and similar settings across the world.