Validation of maternal recall of number of antenatal care visits attended in rural Southern Nepal: a longitudinal cohort study.

Objectives: This study aimed to examine the validity of maternal recall of total number of antenatal care (ANC) visits during pregnancy and factors associated with the accuracy of maternal recall.

Design: This was a longitudinal cohort study conducted from December 2018 through November 2020.

Setting: Five government health posts in the Sarlahi district of Southern Nepal.

Participants: 402 pregnant women between ages 15 and 49 who presented for their first ANC visit at the study health posts.

Main Outcomes: The observed number of ANC visits (gold standard) and the reported number of ANC visits at the postpartum interview (maternal recall).

Results: On average, women in the study who had a live birth attended 4.7 ANC visits. About 65% of them attended four or more ANC visits during pregnancy as recommended by the Nepal government, and 38.3% of maternal report matched the categorical ANC visits as observed by the gold standard. The individual validity was poor to moderate, with the highest area under the receiver operating characteristic curve (AUC) being 0.69 (95% CI: 0.65 to 0.74) in the 1-3 visits group. Population-level bias (as distinct from individual-level bias) was observed in the 1-3 visits and 4 visits groups, where 1-3 visits were under-reported (inflation factor (IF): 0.69) and 4 ANC visits were highly over-reported (IF: 2.12). The binary indicator ANC4+ (1-3 visits vs 4+ visits) showed better population-level validity (AUC: 0.69; IF: 1.17) compared with the categorical indicators (1-3 visits, 4 visits, 5-6 visits and more than 6 visits). Report accuracy was not associated with maternal characteristics but was related to ANC frequency. Women who attended more ANC visits were less likely to correctly report their total number of visits.

Conclusion: Maternal report of number of ANC visits during pregnancy may not be a valid indicator for measuring ANC coverage. Improvements are needed to measure the frequency of ANC visits.