Kåre Mølbak

Doctoral Dissertation
Kåre Mølbak. 2000. The epidemiology of diarrhoeal diseases in early childhood. Dr. Med. Department of Epidemiology Research, Statens Serum Institut. University of Copenhagen.


  The Doctor of Medical Science thesis includes eight previously published papers and a review. Data collection and field work took place at the Bandim Health Project and Laboratório Nacional de Saúde Pública, Guinea-Bissau, 1987-1990. The material has been processed and analysed at Statens Serum Institut, Copenhagen, Denmark. The objectives were to determine morbidity and mortality in relation to diarrhoeal diseases in early childhood, as well as the aetiology and risk factors for diarrhoea. Furthermore, the interaction between nutritional status and diarrhoea was analysed. A particular aim was to assess the significance of persistent diarrhoea, defined as presumed infectious diarrhoea with a duration of at least 2 weeks. 

The under-five mortality in the study area was 243 per 1000, and diarrhoea was the cause in 31% of these deaths. Among the diarrhoeal deaths, 53% was due to persistent diarrhoea. In addition, the morbidity from diarrhoea was high: The average child experienced 9 episodes of diarrhoea per year. Almost all children were breastfed throughout infancy, but from the age of 12 months it was possible to compare the incidence of diarrhoea between weaned and breastfed toddlers. Weaned toddlers had a higher incidence of diarrhoea well into their third year of life, and weaning was associated with a relative mortality of 3.5, independent of age. Other risk factors for diarrhoea included gender (boys had higher incidence), diarrhoea in the previous fortnight, and being nursed by someone else than the mother. Hygienic and environmental factors such as type and ownership of water supply and storage of prepared food for later consumption was significantly associated with diarrhoeal rates. Finally, maternal education, age of family head and ethnic group were determinants for diarrhoea. 

Major pathogens causing diarrhoea comprised rotavirus, enteropathogenic and enterotoxigenic E. coli, Shigella spp. and the parasite Cryptosporidium. Cryptosporidium was found in 15% of cases of persistent diarrhoea, and infants who had cryptosporidiosis had a three times higher mortality than non-infected infants. In an analysis of the nutritional impact of cryptosporidiosis, we found that the infection was followed by a weight deficit, which was not compensated by later catch-up growth. 

A more general assessment of the interaction between diarrhoea burdens ("longitudinal prevalence") and anthopometric status (weight and length/height) showed that diarrhoeal diseases have an negative impact on ponderal and linear growth, and that this impact goes beyond the acute phase. 

An important component of a clinical strategy for the reduction of diarrhoea-related mortality and improvement in nutritional status is better management of children with persistent diarrhoea, which is associated with high diarrhoea burdens and excess mortality. International programmes for diarrhoeal disease controls have previously focused on the promotion of oral rehydration salts for the management of children with acute diarrhoea. This has fostered the belief that diarrhoea control equals oral rehydration treatment. The community studies from Guinea-Bissau shows that the problems are more complex, and that it is necessary to develop strategies for the prevention and differentiated management of children with acute, persistent and dysenteric diarrhoea. 

Published in Danish Medical Bulletin 2000; 47:340-58