© 2004 by Oxford University Press
Local Beliefs about Childhood Diarrhoea: Importance for Healthcare and Research

1 Africa Centre for Health and Population Studies, Mtubatuba, KwaZulu/Natal, South Africa 2 Department of Paediatrics and Child Health, Nelson R. Mandela School of Medicine, University of Natal, Durban, South Africa
The aim of this study was to describe perceptions and beliefs about childhood diarrhoea in a rural South African area. A further aim was to describe mismatches with medical concepts that may influence quality of diarrhoea care and validity of population-based diarrhoea research. This was a descriptive study carried out in a rural district of KwaZulu/Natal, South Africa. Using a validated ethnographic method information was compiled from semi-structured interviews, card sorting exercises, case histories, and focus group discussions with 29 caregivers in total. Description of locally perceived diarrhoea types, signs, symptoms, causes and actions taken were recorded.
Eleven local types of diarrhoea were identified, which were classified into three categories: (A) natural causation, (B) supernatural causation, and (C) caused by germs or change in diet. Conceptual mismatches included: (1) not all conditions with frequent or watery stools were perceived as diarrhoea; (2) hygiene was considered irrelevant in the causation of A and B; (3) signs of dehydration were not always attributed to loss of fluid; (4) medical care and oral rehydration therapy were considered useless in B; (5) potentially harmful treatments, such as herbal enemas, are always preferred in B, sometimes in A. These mismatches carry the potential to hamper health promotion, predispose to delayed and poor care and may lead to selection bias, under- and misreporting and poor compliance in research.
This study highlights the discrepancies in concepts between medically trained and local people regarding the nature of childhood diarrhoea, the rationale behind care-seeking behaviour, and what can be expected from proposed interventions. Taking these discrepancies into account is not only essential to programmatic success and continued support but also for epidemiological surveys.
* Correspondence: Dr J. Van den Broeck, Africa Centre for Health and Population Studies, P.O. Box 198 Mtubatuba 3935, South Africa. Tel. +27 35 5507546; Fax + 27 35 5507565. E-mail
jan.Broeck{at}mrc.ac.za
.<br>
Other members of the Child Health Group: M. B. Bennish, R. M. Bland, H. M. Coovadia, A. Coutsoudis, M. L. Newell, and G. Solarsh.<br>