Tools from ecology: useful for evaluating infection risk models?

Abstract

Despite the increasing number of models to predict infection risk for a range of diseases, the assessment of their spatial limits, predictive performance and practical application are not widely undertaken. Using the example of Schistosoma haematobium in Africa, this article illustrates how ecozonation and receiver-operator characteristic analysis can help to assess the usefulness of available models objectively.

Fig. 1

Spatial distribution of urinary schistosomiasis in Cameroon (a) and Tanzania (b). The infection prevalence was assessed by microscopy of urine in Cameroon, and the data was available for 19 524 children from 333 schools. In Tanzania, the infection prevalence was estimated from carefully validated questionnaire surveys (schoolchildren were asked whether they have urinary schistosomiasis or blood in their urine, termed locally as kichocho). The data for Tanzania are available for 166 099 children from 1960 schools. Although the prevalence of kichocho in schools underestimates the parasitological prevalence of infection, the parasite prevalence can be calibrated for each locality to define the extrapolated risk of having infection prevalence (>50%), and thus comparable to the parasitological data from Cameroon (see Ref. [17] for calibration and data sources).

Fig. 2

Prediction models for Schistosoma haematobium transmission in Cameroon (a) and Tanzania (b). The map shows the probability risk (0-1) of a particular area having an infection prevalence that exceeds the >50% threshold. Reproduced with permission, from Refs [17,18].