Assessing healthcare quality using routine data: evaluating the performance of the national tuberculosis programme in South Africa

Objective To assess the performance of healthcare facilities by means of indicators based on guidelines for clinical care of TB, which is likely a good measure of overall facility quality. Methods We assessed quality of care in all public health facilities in South Africa using graphical, correlation and locally weighted kernel regression analysis of routine TB test data. Results Facility performance falls short of national standards of care. Only 74% of patients with TB provided a second specimen for testing, 18% received follow-up testing and 14% received drug resistance testing. Only resistance testing rates improved over time, tripling between 2004 and 2011. National awareness campaigns and changes in clinical guidelines had only a transient impact on testing rates. The poorest performing facilities remained at the bottom of the rankings over the period of study. Conclusion The optimal policy strategy requires both broad-based policies and targeted resources to poor performers. This approach to assessing facility quality of care can be adapted to other contexts and also provides a low-cost method for evaluating the effectiveness of proposed interventions. Devising targeted policies based on routine data is a cost-effective way to improve the quality of public health care provided.