Methods and Measures to Assess Health Care Provider Behavior and Behavioral Determinants in Reproductive, Maternal, Newborn, and Child Health: A Rapid Review

Advancing Our Understanding of Provider Behavior Change for Improved Health Outcomes

The articles in this supplement highlight the need for strengthening the measure of provider behavior change and provide new evidence and tools for advancing our understanding of provider behavior and effective ways to ensure delivery of high-quality care that supports both clients and providers.

Results From a Multimethod Exploratory Scale Development Process to Measure Authoritarian Provider Attitudes in Democratic Republic of Congo and Togo

BACKGROUND

Health care providers' actions can significantly influence clients' experiences of care, adherence to recommendations, and likelihood of re-engaging with health services. There are currently no validated scales that measure provider attitudes that could affect service delivery in multiple health areas.

METHODS

We developed provider attitude measures in 3 phases. In phase 1 (2019), survey items were developed based on literature reviews, and quantitative items were tested through a health facility survey conducted in the Democratic Republic of the Congo (DRC). Health care providers (N=1,143) completed a 23-question survey focused on 3 subdomains: provider perceptions of clients, provider roles, and gender roles. In phase 2 (2021), cognitive interviews were administered to 17 health care providers in DRC to assess and improve respondents' understanding and interpretation of questionnaire items and response options. In phase 3 (2021), 52 family planning providers were sampled from urban health facilities in Togo to retest and validate the improved measures.

RESULTS

Phase 1 showed the provider attitude items had low scale reliability, and 8 survey items had low variability. In phase 2, results from the cognitive interviews of the 21 items retained from phase 1 found 16 questions were not well understood or had low response variability and thus modified, and 4 survey items were added to test different iterations of specific survey items. In phase 3, exploratory factor analysis resulted in 1 provider attitude scale of 14 items reflecting authoritarian attitudes related to the 3 initial subdomains.

CONCLUSION

This research highlights the importance of iteration and testing during scale development, implementable even across geographic locations. Provider behavior change programming should consider how authoritarian provider attitudes pertaining to professional roles, their clients, and gender norms may interact and influence the quality of health services provided.