H. Bachou (1,3) ([email protected]), J.K. Tumwine1, R.K.N. Mwadime(2), and T. Tylleskar(3)
(1)Department of Paediatrics and Child health, (2)Regional Centre for Quality of Health Care, Makerere University, PO Box 7072, Kampala, Uganda, and (3)Centre for International Health, University of Bergen, N-5021, Norway
Despite poor quality of healthcare having been documented as a major cause of early deaths among hospitalized severely-malnourished children, little has been done to provide information on quality-improvement interventions to reduce early deaths in resource-constrained settings of Africa. The study was set out to map causes of early deaths and implement simple low-cost actions to reduce mortality among children admitted with severe malnutrition.
The study was carried out to assess the risk factors of early deaths among hospitalized severely-malnourished children aged less than 5 years and to determine the effects of simple agreed interventions on mortality among children admitted with severe malnutrition.
This intervention study was conducted in 3 phases--baseline, intervention, and post-intervention--between 2002 and 2005 in the paediatric wards of Mulago Hospital, Uganda. In total, 450 severely-malnourished children, aged less than 5 years (weight-for-height <-3 z-score or oedema)--220 before and 230 after the intervention--were enrolled consecutively and followed till outcome. Demographics, health characteristics, blood and urine specimens were
collected at admission for biochemical, haematological, microbiological, serological and immunological indices, and daily records on patient management were checked. The key intervention was communication to change behaviour of early transfusion and fluid infusion through meetings, workshops, posters, and job-aids posted in wards. Data were analyzed using EPI Info version 6 and SPSS Version 11.5.
There was a reduction in the transfusion and infusion rate (27% to 9% and from 32% to 15% respectively), Deaths associated with transfusion and infusion were reduced (24% to 4% and from 26% to 9% respectively). There were significant differences in the pre- (adjusted odds ration [OR]=3.5, confidence interval [CI] 1.76-6.9, p<0.001and OR=3.1, CI 1.57-5.96, p=0.001 respectively) and the post-interventions (adjusted OR=0.20 CI 0.36-1.16, p=0.072 and
OR=1.64 CI 0.64-4.20, p=0.30 respectively), but the overall case fatality did not change.
Simple behavioural change interventions may reduce early hospital deaths among children admitted with severe malnutrition but do not reduce case fatality.