It aims to evaluate the effective ness of Mãe Coruja interventionin reducing low birthweight and preterm birth. The study will use the Cidacs dataset combined with the data from Mãe Coruja program to carry out the quasi-experimental study. With the support of machine learning techniques, it will also identify social, economic, geographic and environmental conditions that are associated with the outcomes. An index of perinatal health risk will be built to inform improvements in targeting populations and the deployment of similar strategies and programs elsewhere in Brazil.
Municipal subsets of national databases on live births (SINASC), mortality (SIM), notifiable diseases (SINAN), and social programs (CadÚnico) were gathered. Additionally, two locally relevant databases were obtained: the Mãe Coruja Program registry for Recife (MCP-R) and the Recife Atlas. Prior to creating a probabilistic data aggregation algorithm, variable dictionaries were developed.
Initially, the first two databases (SINASC and SIM) were combined. All addresses were georeferenced by district, street name, and number, using commonly available software for visualization and mapping. Geographic coverage of the Family Health Strategy (ESF) and MCP-R was also included as information layers. This process resulted in the prototype of the Recife Georeferenced Health Database (RHGD).
This version of the RHGD enabled initial analysis of the risk factors for low birth weight (LBW) and preterm birth (PT) over time and across different areas. It also allowed for extensive and detailed data mapping and visualization related to diseases or health outcomes. Since the coverage information from ESF and MCP-R is integrated, the RHGD serves as a powerful tool for targeting interventions at municipal, district, or neighborhood levels, as well as for monitoring high-risk children and pregnant women individually. It facilitates the immediate identification of which health worker or team is responsible for follow-up care in primary health settings.
To assess the impact of the MCP-R initiative on birth outcomes, a quasi-experimental analysis was conducted focusing on the prevalence of low birth weight (LBW) and preterm births (PT) before and after MCP-R implementation in various districts of Recife. The analysis compared data from babies born in Recife between 2011-2013 (before the first MCP-R unit started) with those born between 2016-2018, a period during which 13 of the 96 districts had operational MCP-R services for at least one year. The results revealed statistically significant reductions in combined LBW and PT rates in the 13 districts where MCP-R was implemented, aligning with overall trends observed in Recife, where LBW rates decreased from 8.82% to 8.29% (p=0.0004).To assess the impact of the MCP-R initiative on birth outcomes, a quasi-experimental analysis was conducted to examine the prevalence of low birth weight (LBW) and preterm births (PT) before and after the implementation of MCP-R in various districts of Recife. The study compared data from babies born in Recife between 2011 and 2013, before the first MCP-R unit began operating, with data from those born between 2016 and 2018, during which 13 of the 96 districts had operational MCP-R services for at least one year. The results showed statistically significant reductions in the combined rates of LBW and PT in the 13 districts where MCP-R was implemented, reflecting overall trends in Recife, where LBW rates decreased from 8.82% to 8.29% (p=0.0004), and PT rates dropped from 12.36% to 11.48% (p=0.0001).
However, no consistent and statistically significant changes were identified when analyzing each district individually. Evaluating the effectiveness of community health interventions poses challenges, particularly when randomization is not possible. Additionally, various social, economic, and biological factors can significantly influence LBW and PT rates, which may obscure the effects of targeted interventions. This evaluation focused solely on birth outcomes; further research on long-term child outcomes at ages 1 and 5 would be beneficial.
Given the political momentum to expand the program due to its intangible benefits, such as empowering women and minimal risk of harm, the project provided the management team with data-driven insights, resource mapping, and visualization tools. This information facilitated the program’s extension to other neighborhoods, prioritizing local disparities instead of district-wide generalizations, and underscored the need for earlier and improved prenatal care. This led to the development of the web application prototype “Me Siga! (Follow Me!).” and PT rates fell from 12.36% to 11.48% (p=0.0001).
However, no consistent and statistically significant change was identified when analyzing each district individually. Evaluating the effectiveness of community health interventions presents challenges, especially when randomization is not feasible. Additionally, various social, economic, and biological determinants can significantly influence LBW and PT rates, potentially obscuring the effects of targeted interventions. This evaluation focused solely on birth outcomes; further research on long-term child outcomes at ages 1 and 5 would be beneficial. Given the political momentum to expand the program due to its intangible benefits, such as empowering women, and the minimal risk of harm, the project provided the management team with data-driven insights, resource mapping, and visualization tools. This information facilitated the extension of the program to other neighborhoods—prioritizing local disparities rather than district-wide generalizations—and highlighted the need for earlier and improved prenatal care. This led to the conceptualization of the web application prototype “Me Siga! (Follow Me!).”
The Mãe Coruja Program was designed to offer a comprehensive set of interventions aimed at pregnant women and children up to five years old through an integrated approach encompassing health, education, and social development. Over the past decade, the program has expanded significantly, reaching coverage in most municipalities of Pernambuco, including the capital, Recife. Existing evaluations have primarily focused on maternal and infant mortality rates. This project stands out as the first to assess the program’s impact on low birth weight and preterm birth rates, using innovative, data-driven methodologies. A central aspect of the project is the use of geospatial computing, which serves a dual purpose: facilitating both data integration and statistical inference, and acting as an exploratory tool to reveal contextual factors influencing the program’s effectiveness. This innovative approach aimed to provide deeper insights into program outcomes and guide future interventions.
This research significantly contributes to the planning and implementation of health interventions that are timely and that optimize limited municipal resources while addressing the contextual factors underlying disparities in perinatal health. Evaluating the impact of the Mãe Coruja program on perinatal health outcomes is essential for understanding the environmental and social determinants that influence these results and for providing specific recommendations for program improvement. Furthermore, obtaining more detailed, comprehensive, and up-to-date data will enable in-depth analyses that can guide the potential scalability and implementation of the Mãe Coruja program in regions beyond Pernambuco.