Face and content validity of a prospective multidimensional performance instrument for service delivery in district health systems in low-income countries: a Delphi study

Background: Valid performance indicators help to track and improve health services. The aim of this study was to test the face and content validity of a set of performance indicators for service delivery in district health systems of low-income countries. Methods: A Delphi method with three stages was used. A panel of experts voted (yes vs no) on the face value of performance indicators. Agreement on the inclusion of indicators was a score of >75% and >= 50% during stages one and two, respectively. During stage three, indicators with a mean score of >= 3.8 on a five-point scale were included. The panel also rated the content validity of the overall set of indicators. Results: The panel agreed on the face value of 59 out of 238 performance indicators. Agreement on the content validity of the set of indicators reached 100%. Most of the retained indicators were related to the capacity of health facilities, the quality of maternal and child health services and HIV care and treatment. Conclusions: Policymakers in low-income countries could use a set of performance indicators with modest face and high content validity, and mainly aspects of capacity and quality to improve health service delivery in districts.