Despite significant improvements in maternal, neonatal, and child health (MNCH) and family planning (FP) status in Bangladesh, there were unfinished agendas in terms of achieving the MDGs targets related to newborn health and fertility.
Our assumptions were that – adverse newborn health outcomes and undesired fertility rates are clustered in high-risk groups and in the hard-to-reach communities. Therefore, differential attention to these high-risk groups is likely to improve the overall newborn health and family planning status of the country, and we termed it as ‘Targeted Approach. In the Targeted Approach, we planned to deliver targeted services to high-risk, disadvantaged, and hard-to-reach populations that contribute to a substantial portion of excess but preventable births and deaths. Although there was a growing discussion on targeted approaches in different forums and meetings, to our knowledge, no study was carried out on a broader scale to assess the effect of targeted approaches on MNCH and FP services. This study
This USAID-funded study under the Translating Research Into Action (TRAction) project at icddr,b aimed to develop a method for targeting high-risk newborns, and eligible couples with specific family planning needs, and to assess the sets of differential interventions to increase the utilization of MNCH and FP services which will eventually lead to improvements in newborn survival and family planning status. It had two separate components – newborn part, and family Planning part.
Newborn health
We designed and evaluated a combination of community and facility-based interventions interlinked by an active and responsive referral system. Our intervention included community health worker-assisted pregnancy and birth surveillance, improved identification of sick neonates, appropriate referral for facility-based care, and establishment of a newborn stabilization unit at the first level referral health facility. We hypothesized that this integrated intervention would improve care-seeking for sick newborns and quality of care at the first level facility, with ultimate improvement in newborn survival.
The study was carried out in Sylhet district (200 kilometers north of Dhaka with a population of around 4 million) where newborn-death rate was high. This was a quasi-experimental study. Five unions from Jaintapur sub-district were allocated to the intervention and four unions from Gowainghat sub-district to the comparison arm. In the intervention arm, Save the Children’s MaMoni program was implementing MNCH interventions, whereas the comparison arm received usual health services. We evaluated all women who delivered 12 months before the baseline and end-line surveys.
The outcome indicator was neonatal mortality rates, and the key coverage indicator was newborn care-seeking from qualified provider. However the study also assessed other indicators that included antenatal care from qualified provider, facility delivery, deliveries by skilled birth attendant, postnatal care from qualified provider. These indicators were compared between the intervention and comparison areas across baseline and endline surveys. Both outcome and coverage indicators were measured through cross sectional household surveys, adequacy survey and from longitudinal record keeping of the program offices. An incremental cost analysis was also done. All operational issues were systematically documented in both the intervention and comparison areas. Details of evaluation methodology are described below.
Family planning
This quasi-experimental study was a collaborative effort of icddr,b, MOHFW, and EngenderHealth. EngenderHealth (in partnership with BCCP and the Population Council) was undertaking USAID-supported project to improve the quality and use of LARC/PM. One sub-district was selected each from Brahmanbaria and Sylhet districts for the intervention arm. Similarly, two sub-districts from the two districts were also selected as comparison.
Baseline and endline surveys were conducted in intervention and comparison sub-districts among married women of reproductive age. The baseline was immediately before the implementation of interventions and the endline was after 15 months of intervention, in both intervention and comparison sub-districts. A follow-up survey was also conducted with IUD and Implant clients both in intervention and comparison Upazillas.
The effect of the interventions were measured by comparing indicators before and after the intervention and between the intervention and comparison areas. The key indicators were: LARC/PM use rate, use rate of short acting methods (pill, injectable, condoms) and discontinuation rate of IUD and Implants. Difference-in-difference (DID) estimation was used to track longitudinal changes in the difference of the above indicators. STATA statistical software (version 12) was used for data analyses.
The study findings were published on a report. A dissemination event was organized at icddr,b where relevant government, non-government, development partner, academic stakeholders were present.