Supportive Supervision as an approach to improve the quality of care for children with acute malnutrition in Arua district, Uganda: Baseline systematic assessment, Cluster Randomised Controlled Trial and Cost-Effectiveness Analysis

INTRODUCTION Moderate and severe acute malnutrition estimates among children in the West Nile region, in Uganda, are higher than the national level (10.4% and 5.6%, respectively versus 3.6 % and 1.3 %). Additionally, the WHO estimates that in 2016, 6.6 million children and young adolescents died from causes attributed to the poor quality of care in such similar settings. Supportive supervision (SS) has been proposed as one of the approaches to improve quality of care. The main objectives of this project were; to determine the baseline status of the quality of care of nutrition services and health outcomes among malnourished children at health facility level; to test the effectiveness of supportive supervision to improve health outcomes and quality of care; and to estimate its cost effectiveness. METHODS Phase one: Six health centers with the highest burden of malnutrition in Arua district, West Nile region, were selected. Information on health outcomes (cured, defaulters, non responders, transferred and died) and quality of case management were extracted from official records. Quality of care was assessed using the national Nutrition Service Delivery Assessment (NSDA) tool, with ten key areas scored as poor, fair, good or excellent. Phase two: The six facilities were randomized to receive either SS or to control. SS was delivered for ten months in two equal five months’ periods; to heath center (HC) staff only (first period), and later extended to community health workers (CHWs) (second period). SS was delivered biweekly for the first three months and later monthly. The package included: monitoring progress, provision of technical support, facilitating good team dynamics and problem solving attitude. The control facilities were assigned to receive the national routine quarterly supervisory visits. Main outcomes included health outcomes, quality of case management, quality of nutrition service delivery and access to care. Phase three: The Incremental Cost Effectiveness Ratios (ICER) for the first and second period were estimated. RESULTS Phase one: A total of 1020 children were assessed at baseline. The cured and defaulter’s rates were 52.9% (95% CI: 49.7 – 56.1) and 38.3% (95%CI: 35.2 – 41.4) respectively. The NSDA revealed 33/60 (55%) areas scored poorly, 25/60 (41%) as fair, 2/60 (3.3%) were good and none were excellent. Main gaps included: lack of trained staff; disorganized patient flow; poor case management; stock out of essential nutrition supplies and weak community linkage. Phase two: 737 children were enrolled, 430 in the intervention and 307 in the control. Significant findings of the intervention versus control included: higher cure rate [83.8% (95%CI: 79.4 – 86.7) versus [44.9% (95%CI: 37.8 – 49.1), p=0.010)], lower defaulting rate [1.4% (95%CI 1.1% to 1.8%) versus 47.2% (95%CI 37.3% to 57.1 %), p=0.001], higher correct complementary treatment (94.0% versus 58.8%, p=0.001) and more NSDA areas scored as either good or excellent [24/30 (80%) versus 14/30 (46.6%), OR = 4.6 (1.3 – 17.4), p=0.007]. Access to care was significantly higher during the second period as compared to the first period [proportion difference = 28.4%, OR = 1.7 (1.3 – 2.3), p = 0.001].Phase three: the ICER of € 9.7 (95%CI:7.4 – 14.9) and € 6.8 (95% CI:4.8 – 9.5) were estimated in the first and second periods respectively. CONCLUSION At baseline, the quality of care provided to children with malnutrition at health center level was greatly substandard. The delivery of SS to HC staff and CHWs significantly improved the cure rate, the quality of case management, the overall quality of care and access to care. SS, especially that delivered to CHWs, was very cost effective.