Background: There are increasing efforts to monitor progress in maternal and neonatal care, with household surveys the main mode of data collection. Postnatal care (PNC) is considered a priority indicator yet few countries report on it, and the need to improve the construct validity associated with PNC questions is recognized.
Objectives: To determine women's knowledge of what happens to the baby after delivery, women's comprehension of terms and question phrasing related to PNC, and issues with recall periods.
Design: Forty qualitative interviews and four focus group discussions were conducted with mothers, and 10 interviews with health workers in rural Ghana. Data were collected on knowledge and recall of postnatal health checks and language used to describe these health checks.
Results: Mothers required specific probing using appropriate language to report postnatal checks. They only had adequate knowledge of postnatal checks, which were easily observed or required asking them a question. Respondents reported that health workers rarely communicated with mothers about what they were doing, and most women did not know the purpose of the equipment used during health checks, such as why a thermometer was being used. Knowledge of neonatal checks in the first hours after a facility delivery was low if the mother and child were separated, or if the mother was tired or weak. Many women reported that they could remember events clearly, but long recall periods affected reporting for some, especially those who had multiple checks or for those with no problems.
Conclusions: Direct questions about PNC or health checks are likely to underestimate coverage. Validity of inferences can be enhanced by using appropriate verbal probes during surveys on commonly performed checks that are clear and visible to the mother.
Keywords: Ghana; household survey; indicators; measurement; postnatal care; pre-testing; validity.