Brazilian Maternal and Child Nutrition Consortium: establishment, data harmonization and basic characteristics

Area of Science:

• Maternal and Child Health
• Nutritional Epidemiology
• Data Harmonization

Background:

• Pooled data analysis in maternal and child nutrition often excludes low- and middle-income countries.
• Existing studies lack detailed methods for data harmonization and heterogeneity assessment.
• There is a need for robust datasets to address critical maternal and child health questions in under-resourced settings.

Purpose of the Study:

• To describe the creation of the Brazilian Maternal and Child Nutrition Consortium dataset.
• To exemplify data harmonization and heterogeneity assessment methods using gestational weight gain (GWG).
• To establish a foundation for future pooled analyses in maternal and child nutrition.

Main Methods:

• Longitudinal studies published between 1990 and 2018 were identified and invited for participation.
• Data underwent consistency analysis, outlier identification (considering longitudinal data), and heterogeneity assessment using multilevel models.
• A final sample of 17,344 women from 21 studies was included after data reception and retention.

Main Results:

• Fewer than 1% of 72,616 weight measurements were identified as outliers.
• Women with pre-pregnancy obesity exhibited lower gestational weight gain (GWG).
• GWG, birth length, and birth weight were consistent across included studies and comparable to a Brazilian nationwide study.

Conclusions:

• The Brazilian Maternal and Child Nutrition Consortium dataset provides a valuable resource for pooled analyses.
• Harmonized data and robust methods increase the potential for addressing key maternal and child health issues.
• Pooled analyses are crucial for advancing maternal and child health research, particularly in countries with limited research investment.