How Exclusive is Exclusive Breastfeeding?

Kristine A. Gonzalez

Objective: To determine the difference in prevalence of exclusive breastfeeding (EBF) as measured using different pre-defined time intervals and measurement frequencies: 24-hour recall vs. 7-day recall; and 24-hour recall vs. set of 24-hour recalls obtained weekly since birth.

Methods: This study incorporated data from two population-based studies – the first, a cohort study conducted between 1998-2001 in San Pedro Martir, Mexico City; the second, a cross-sectional survey conducted in 1999-2000 in Guatemala City. The cohort study analysis included: calculating the prevalence of EBF by week of age to 26 weeks; calculating the average percent breastfeeding (BF) by week of age to 26 weeks; calculating and comparing the Kaplan-Meier survival curve for EBF using different time intervals; calculating the percent of BF in the study; and calculating the percent of weeks EBF for each study child. With the cross-sectional study, analysis included comparing 7-day to 24-hour recall to characterize the prevalence of EBF at the measurement time. 

Results: In the cohort study, the prevalence of EBF and BF varied according to the age of the infant. As expected, the prevalence of EBF decreased over time (measured as the infant’s age in weeks). For week 1, the prevalence of EBF was 67.7% and the cumulative survival was 79.2%. For week 13 (approximately 3 months of age) the prevalence of EBF was 79.3% and the cumulative survival was 41.7%, and for week 26 (approximately 6 months of age) the prevalence of EBF was 29.9% and the cumulative survival was 7.6%. There was a smaller difference in the prevalence of any BF over time. In addition, the prevalence of BF was higher than EBF, fluctuating between 97.9% and 90.2% for the entire 26 weeks of age. In the cross-sectional study, 19.8% of the 777 infants were EBF at 24-hour recall, while only 10.9% were EBF at 7-day recall. Therefore, in the cross-sectional study, the fluctuation in time interval for determining EBF practices influences the prevalence by 8.9%. 

Conclusion: The prevalence and cumulative survival of EBF varies by week of age, and there was a difference between the prevalence of EBF as measured by prevalence at a specific time and cumulative survival analysis at that same week of age. In addition, the prevalence of EBF for 24-hour recall was higher than the prevalence for 7-day recall. Given the global emphasis on EBF research, there is a need for a clear and valid pre-defined time interval for establishing the prevalence of EBF.