JMIR Publications

ABSTRACT

Background:

While the use of routine health information system data derived from medical records in healthcare facilities is a widely recognized strategy to improve the quality of care during and around childbirth, evidence showed that low quality data has been limited its use. There is an overall consensus that non-financial incentives can enhance healthcare providers' commitment and achieve desired healthcare quality. However, there is a shortage of evidence on the effect of non-financial incentives in improving the data quality of institutional birth services in Ethiopia.

Objective:

This study aimed to examine the effect of performance-based non-financial incentives on the data completeness and consistency of individual medical records for women who attended institutional birth in Northwest Ethiopia.

Methods:

The study used a nonequivalent control group pre-post quasi-experimental design in Northwest Ethiopia's "Wegera" and "Tach-armachiho" districts. The "Wegera" district was assigned to receive the intervention, while the "Tach-armachiho" district was designated as the comparison group. The study's intervention included a multi-component non-financial incentive, including smartphones, flash disks, power banks, certificates of recognition, and scholarship opportunities offered at healthcare facilities, departments, and individual healthcare providers levels. A sample size of 1969 combination of women’s individual medical records was used in this study. We included individual medical records of women who gave birth within a six-month period preceding the intervention (April to September 2020) and following the intervention (February to July 2021). The completeness and consistency of data elements were assessed by reviewing three types of individual medical records: the integrated card, the Integrated individual folder, and the delivery register. To determine completeness, we examined the presence of data elements within the three distinct medical records, and to evaluate consistency, we verified the agreement of data elements across the medical records. A two-sample test of proportion and difference-in-differences analysis was employed to compare the outcomes between the study groups.

Results:

The data completeness of individual medical records in the intervention district was nearly four times higher compared to the comparator district, with an Average Treatment Effect on the treated,3.86 (95% CI [2.21, 5.51], P=.021). Regarding consistency, the individual medical records’ data in the intervention district were approximately twelve times more likely to be consistent compared to the comparator district, with an Average Treatment Effect on the Treated of 11.6 (95% CI [4.18, 19.02], P=0.032).

Conclusions:

Our study demonstrates a significant improvement in data quality within individual medical records of institutional births through performance-based non-financial incentives. Stakeholders aiming to improve maternal and newborn healthcare quality can consider the value of offering non-financial incentives, particularly in healthcare facilities of resource-limited setting.