Complexities and subtleties in the measurement and reporting of breastfeeding practices

Background: Monitoring of breastfeeding is vital. However, infant feeding practices are difficult to assess at the population level. Although significant efforts have been made towards the consistent measurement and reporting of breastfeeding, few countries have successfully implemented a system to do so. Many inaccuracies, inconsistencies and issues remain. This paper highlights the main issues relating to the methods and indicators used to monitor breastfeeding, particularly exclusive breastfeeding, at the population level. In doing so, it aims to support progress in this area. Discussion: Indicators are used primarily for comparative purposes and should be broadly consistent with recommended practice; regarding exclusive breastfeeding this is 'to six months'. There are limitations to both main methods used to measure and report on breastfeeding: current status (often 24-hour recall), and longer-term recall. Issues relate to how age is considered within the analysis and interpretation of data, including boundary points or cut offs, as well as how breastfeeding practices are reported against different ages, especially regarding whether to use the preposition 'to' or 'at'. Other issues include the conversion from weeks to months, as well as the 'regular' versus 'first' introduction of something other than breast milk, to signify the deviation from exclusive breastfeeding. Differences in how data are collected, and uncertainties around how data are interpreted, have led to the mixed and often inaccurate reporting of breastfeeding practices, particularly exclusive breastfeeding. Assuming a particular definition of exclusive breastfeeding, such as that of the World Health Organization, the period over which exclusive breastfeeding is measured and how it is determined in the survey are important in relation to indicator phrasing. Often compromises are made in data collected to report against exclusive breastfeeding, despite subsequent reporting of exclusivity. Summary: Indicators to report on breastfeeding must be carefully phrased. The commonly reported indicator exclusive breastfeeding at six months is redundant and should never be reported, while the more appropriate indicator exclusive breastfeeding to six months may not be sufficiently sensitive to change, and cannot be measured by current status methods alone. Importantly, indicators must accurately reflect the data collected to ensure valid comparisons between surveys.