Menu
Menu
Menu
Menu
Background: Between June 2011 and December 2016, the Saving Mothers, Giving Life (SMGL) initiative in Uganda and Zambia implemented a comprehensive approach targeting the persistent barriers that impact a woman’s decision to seek care (first delay), ability to reach care (second delay), and ability to receive adequate care (third delay). This article addresses how SMGL partners implemented strategies specifically targeting the second delay, including decreasing the distance to facilities capable of managing emergency obstetric and newborn complications, ensuring sufficient numbers of skilled birth attendants, and addressing transportation challenges. Methods: Both quantitative and qualitative data collected by SMGL implementing partners for the purpose of monitoring and evaluation were used to document the intervention strategies and to describe the change in outputs and outcomes related to the second delay. Quantitative data sources included pregnancy outcome monitoring data in facilities, health facility assessments, and population-based surveys. Qualitative data were derived from population-level verbal autopsy narratives, programmatic reports and SMGL-related publications, and partner-specific evaluations that include focus group discussions and in-depth interviews. Results: The proportion of deliveries in any health facility or hospital increased from 46% to 67% in Uganda and from 63% to 90% in Zambia between baseline and endline. Distance to health facilities was reduced by increasing the number of health facilities capable of providing basic emergency obstetric and newborn care services in both Uganda and Zambia-a 200% and 167% increase, respectively. Access to facilities improved through integrated transportation and communication services efforts. In Uganda there was a 6% increase in the number of health facilities with communication equipment and a 258% increase in facility deliveries supported by transportation vouchers. In Zambia, there was a 31% increase in health facilities with available transportation, and the renovation and construction of maternity waiting homes resulted in a 69% increase in the number of health facilities with associated maternity waiting homes. Conclusion: The collective SMGL strategies addressing the second delay resulted in increased access to delivery services as seen by the increase in the proportion of facility deliveries in SMGL districts, improved communication and transportation services, and an increase in the number of facilities with associated maternity waiting homes. Sustaining and improving on these efforts will need to be ongoing to continue to address the second delay in Uganda and Zambia.
We reviewed both quantitative and qualitative data to describe the implementation of SMGL interventions targeting the second delay and to evaluate collective outcomes. Partners collected program-specific data as well as indicators that were harmonized across a broader SMGL monitoring and evaluation strategy. In both countries, the U.S. Centers for Disease Control and Prevention country office oversaw partner-specific data collection and evaluation activities and each implementing partner had data quality control measures in place for data collection and data entry. We also reviewed partner-specific programmatic data and data from systematic evaluations. In both countries, SMGL partners carried out formative research to understand the country context and determine which specific factors influenced the practice of key behaviors before, during, and after delivery. Pregnancy outcome monitoring data in facilities, health facility assessments, and population-based data were the primary quantitative data sources used to assess programmatic outcomes related to strategies addressing the second delay across all SMGL-supported districts. Details on data collection methods have been published elsewhere.24 Implementing partners collected routine data on 31 SMGL indicators from the labor and delivery and other in-patient registers in facilities in SMGL-supported districts. Baseline indicators were calculated from data collected during Phase 0 between June 2011 and May 2012, and endline indicators were calculated from data collected between January and December 2016. We then calculated the relative change between the baseline and endline indicators. To compare the baseline and endline results for significant differences, we used the McNemar’s test, which is appropriate for dichotomous responses for matched pairs of data at different time points. Health facility assessments were conducted at baseline (late 2011 in Zambia and early 2012 in Uganda) and at endline (2017) in all health facilities in SMGL-supported districts. We used these data to document the status of health facilities and their availability of lifesaving emergency obstetric interventions at the time of the assessment. In this article, we present results compiled from facilities that maintained delivery capacity from baseline to endline—105 in Uganda and 110 in Zambia. The assessments were aligned with the WHO criteria for basic (BEmONC) and comprehensive emergency obstetric and newborn care (CEmONC)2 and included questions about facility infrastructure, staffing, ability to perform signal functions,2 stock-outs of key medications required for the management of complications, and referral system components including transportation and communication. We classified facilities as EmONC if, in the previous 3 months, they performed all 7 of the signal functions for BEmONC and all 9 for CEmONC at the time of the assessment, and non-EmONC if they were not capable of performing all of the BEmONC signal functions. Population-level household surveys (Reproductive Age Mortality Study in Uganda and SMGL Census in Zambia) were conducted in 2012 and 2017.24 We combined household data with the health facility routine monitoring data and health facility assessment data, to calculate the proportion of facility deliveries, stratified by EmONC capacity, at baseline and endline. The facility delivery rate was calculated using the number of deliveries verified to have occurred in an SMGL-affiliated facility divided by the estimated number of live births in the SMGL districts at each time point. The number of births was estimated by applying crude birth rates (derived from the age-specific fertility rates among women of reproductive age enumerated in 2013 in Uganda districts, and derived from the 2010 national census in Zambia) to the baseline and endline district population. We calculated the relative change in facility deliveries between baseline and endline, assuming some variation in error or measurement. To test for significance, z scores based on the normal approximation to the binomial distribution were used to calculate P values. We derived qualitative data primarily from population-level verbal autopsy narratives, programmatic reports and SMGL-related publications, and partner-specific evaluations that included focus group discussions and in-depth interviews. Following the population-based household surveys, retrospective verbal autopsies were conducted and used to measure the medical causes and delay-associated factors of maternal deaths. An open-ended narrative was captured, detailing the experiences of women prior to maternal death and offering context to health facility results. We reviewed implementation partner reports and SMGL-related publications and evaluations primarily to describe the intervention activities that occurred under each strategy.25,26 When necessary, we contacted implementing partners for clarification to resolve discrepancies and to provide more in-depth descriptions of program activities. Data were used to outline SMGL activities and contextualize findings related to the second delay. Partners provided data from focus group discussions with community groups and in-depth interviews with health systems staff at the district and health facility levels to understand the perceived impact of interventions. We analyzed focus group discussions and in-depth interviews using content analysis. From these qualitative data sources, we gathered information on the (1) description of strategies, (2) methods of implementation, (3) outputs (i.e., direct results of activities), and (4) outcomes (i.e., changes in knowledge or behaviors of the pregnant women/target population). Data were triangulated with the quantitative data to assess the implementation of strategies related to the second delay in the context of the SMGL initiative. The study protocol was reviewed and approved by the ministries of health in Uganda and Zambia and deemed non-research by the Human Research Protection Office of the Center for Global Health at the U.S. Centers for Disease Control and Prevention. Written informed consent was obtained for respondents in all households and among women for the census, Reproductive Age Mortality Study interviews, focus group discussions, and in-depth interviews.
N/A
