Impact of digital antenatal care intervention on paper-based antenatal care recordkeeping: a before-and-after study in primary healthcare facilities in Nepal

Objective To assess the impact of introducing electronic decision support systems (EDSS)-electronic data entry implemented alongside existing paper-based antenatal care (ANC) records-on the completeness and agreement of ANC records. Design Two-phase cross-sectional (before and after) substudy of the mobile health integrated model of hypertension, diabetes and ANC (mIRA project) process evaluation. Setting Four rural districts in Bagmati Province, Nepal, in 19 primary healthcare facilities. Participants ANC records from pregnant women attending facilities before (n=136) and after (n=138) EDSS implementation. Main outcome measures For selected indicators in the ANC card and ANC register, we estimated the percentage completeness (any value recorded) and agreement (whether values matched) before and after EDSS implementation. We also reported the completeness of indicators in the EDSS and calculated the agreement between the ANC card and EDSS. chi(2) or Fisher's exact test, as appropriate, was used to assess differences in completeness before and after implementation. Results Completeness of paper-based ANC records was high before implementation (>90%) for all indicators, except tetanus vaccination (<80%). After EDSS implementation, there was >15% improvement in the completeness of tetanus vaccination date in paper-based ANC records (77.0%-96.4% for ANC cards and 81.9%-98.9% for ANC register). Agreement between the ANC card and ANC register increased slightly for all indicators after implementation, and the tetanus vaccination date showed the largest increase (38.2%-57.2%). Indicator completeness in the EDSS was low, ranging from 38.2% to 88.7%. Conclusion We found slight improvements in the completeness and agreement of paper-based ANC records following EDSS implementation. The lower percentage of completeness in the EDSS suggests that any large-scale implementation should consider how to integrate digital and paper-based records to decrease the data entry burden on ANC providers. However, the study's small sample size limited the ability to examine variation in effects.