[The validity of a presumptive diagnosis of cutaneous leishmaniasis performed by community health workers in Colombia].

OBJECTIVE

To validate a method for the presumptive diagnosis of cutaneous leishmaniasis based on the observation of clinical-epidemiological criteria, carried out by community health workers in three endemic municipalities in Santander, a department in northeastern Colombia.

METHODS

This evaluation study of diagnostic technologies was based on a cross-sectional sampling of suspected cases of cutaneous leishmaniasis in the municipalities of Rionegro, El Playón, and Landázuri. After being trained, the community health workers carried out the presumptive diagnoses of cutaneous leishmaniasis between October 2004 and November 2005. At the time of diagnosis, the health workers also collected samples for confirmatory diagnosis through Field's stain method, culturing, and polymerase chain reaction. Four criteria were used to assess the validity of the presumptive diagnoses carried out by the health workers: sensitivity, specificity, positive predictive value, and negative predictive value. Replicability among the health workers was estimated through their observed level of agreement.

RESULTS

According to the laboratory tests, of the 196 patients studied, 33 (16.8%) were negative and 163 (83.2%) were positive. For all the levels of certainty of the clinical diagnosis, the sensitivity was between 52% and 98% (kappa(1, 0) = 39.0%) and the specificity between 9% and 55% (kappa(0, 0) = 14.0%). The area under the receiver operating characteristic curve was 56.5% (95% confidence interval: 45.6% to 67.4%). The proportion of positive agreement and of negative agreement was 86.3% and 43.5%, respectively.

CONCLUSIONS

The sensitivity of the presumptive diagnosis carried out by the health workers surpasses that of the parasitological diagnostic methods generally used in the three endemic areas, but its specificity is much lower. Even though this approach is not useful as a diagnostic test for cutaneous leishmaniasis or as a criterion for deciding to proceed with treatment or not, the method is useful for the active identification of cases in the community.