WHO guidelines for management of severe malnutrition in rural South African hospitals: effect on case fatality and the influence of operational factors.

BACKGROUND: WHO case-management guidelines for severe malnutrition aim to improve the quality of hospital care and reduce mortality. We aimed to assess whether these guidelines are feasible and effective in under-resourced hospitals. METHODS: All children admitted with a diagnosis of severe malnutrition to two rural hospitals in Eastern Cape Province from April, 2000 to April, 2001, were studied and their case-fatality rates were compared with the rates in a period before guidelines were implemented (March, 1997 to February, 1998). Quality of care was assessed by observation of medical and nursing practices, review of medical records, and interviews with carers and staff. A mortality audit was used to identify cause of death and avoidable contributory factors. FINDINGS: At Mary Theresa Hospital, case-fatality rates fell from 46% before implementation to 21% after implementation. At Sipetu Hospital, the rates fell from 25% preimplementation to 18% during 2000, but then rose to 38% during 2001, when inexperienced doctors who were not trained in the treatment of malnutrition were deployed. This rise coincided with less frequent prescribing of potassium (13% vs 77%, p<0.0001), antibiotics with gram-negative cover (15% vs 46%, p=0.0003), and vitamin A (76% vs 91%, p=0.018). Most deaths were attributed to sepsis. For the two hospitals combined, 50% of deaths in 2000-01 were due to doctor error and 28% to nurse error. Weaknesses within the health system--especially doctor training, and nurse supervision and support--compromised quality of care. INTERPRETATION: Quality of care improved with implementation of the WHO guidelines and case-fatality rates fell. Although major changes in medical and nursing practice were achieved in these under-resourced hospitals, not all tasks were done with adequate care and errors led to unnecessary deaths.