Letter to the Editor
Article first published online: 26 OCT 2005
Tropical Medicine & International Health
Volume 10, Issue 11, page 1207, November 2005
How to Cite
Van den Ende, J. and Mukaminega, M. (2005), Letter to the Editor. Tropical Medicine & International Health, 10: 1207. doi: 10.1111/j.1365-3156.2005.01508.x
- Issue published online: 26 OCT 2005
- Article first published online: 26 OCT 2005
To the editor,
In the excellent paper on the attributable fraction of malaria in Mali by Dicko et al. (2005) it is not clear if the malaria ‘case definition’ is also a suggestion to treat. The ‘optimal’ cut-off value, where the sensitivity curve crosses the specificity curve cannot be used for the decision to prescribe treatment. This cut-off value is rather an epidemiological threshold than a clinical one. For identification of the treatment threshold, decision theory should be applied. The individual and global weight of false positives and false negatives should be taken into account (Zweig & Campbell 1993; Xiao-Hua et al. 2002). These are based on prevalence of the disease, mortality, morbidity and cost and side effects of current treatment. For some conditions, possible effects on transmission directly or through the reservoir function of patients and carriers, and stigma might enter the scene as well.
In the setting where the study has been done, the high prevalence, the high mortality and morbidity of the disease, the so far cheap treatment with few and non-life threatening side effects all point towards a predominance of the individual and global weight of the false negatives, warranting a much lower treatment threshold.
- 2005) Season, fever prevalence and pyrogenic threshold for malaria disease definition in an endemic area of Mali. TM & IH 10, 550–556. , , et al. (
- 2002) Statistical Methods in Diagnostic Medicine. John Wiley & Sons, New York, p. 437. , & (
- 1993) Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clinical chemistry 39, 561–577. & (