Maternal recall of exclusive and any breastfeeding duration during the first 6 months- an examination of retrospective accuracy at 12 months within a large prospective breastfeeding survey in Germany
摘要
Effective promotion and support of breastfeeding requires reliable data on breastfeeding duration, particularly of exclusive breastfeeding (EBF). Maternal retrospective recall of breastfeeding duration is often used to assess breastfeeding rates and trends in a certain population, but recall accuracy is limited by potential recall bias or imprecision of memory. Objective of this study was to examine mothers’ ability to accurately recall their breastfeeding duration at 12 months postpartum after individual recall periods of 6, 8 or 10 months.
MethodsThis assessment has been embedded into the web-based nationwide prospective survey on breastfeeding and infant nutrition SuSe-II 2017–19 in Germany. A set of 2–3 simple recall questions on duration of EBF (as defined by World Health Organization) and any breastfeeding (ABF) has been compared to prospective maternal reports of breastfeeding at 2, 4 and 6 months as reference.
ResultsData pairs of recalled duration and reference duration were provided by 572 mothers for EBF and 163 mothers for ABF. On average, duration of EBF and ABF differs statistically significantly between methods, but deviations are small (2–3 weeks), and tend to increase with longer recall period. For both, EBF and ABF, at a tolerance level of ±2 weeks, the recall overestimates breastfeeding duration in about 50% of recall-reference pairs, while underestimation is scarce (about 10%) and about 40% of data pairs are accurate. In scenarios with a deviation tolerance of 1, 2, or 4 weeks, around 20%, 40% or 70% of recalls would be classified as accurate.
ConclusionsAsking about duration of EBF and ABF at 12 months postpartum with a simple set of questions in a large sample of mothers was found to be suitable for assessing ABF and EBF duration within the first 6 months. Specific scenarios of tolerance levels suggest a deviation between recall and reference breastfeeding duration of ±2 weeks as feasible compromise between high accuracy and high proportion of acceptable recalls.
Trial registrationGerman Clinical Trials Register (DRKS) (ID: DRKS00014601).