Use of the capture-recapture method for determining the prevalence of neurological parasitic diseases.

Fleury et al. [1] reported that the prevalence of human neurocysticercosis (NC) caused by Taenia solium, pork tapeworm, in Tepetzezintla, Mexico, was 6.5–9.1%. NC is caused by the encysted larval stage (cysticercus) occurring in the brain. The prevalence of NC has not been very well established and is likely higher than reported estimates. Studies determining the prevalence of NC are difficult to perform, with a likely large undercount. To obtain a better estimate of the prevalence of this neurological disease, as well as others, especially those that are rare, the capturerecapture method (CRM) can be employed [2, 3]. CRM is a common procedure used by population biologists [4] and more recently by epidemiologists [2, 3], including for neurological epidemiology [5]. This method can use lists that have been previously generated or generate a list that can be compared to a previous list. This has been referred to as the two-source method [3]. There are certain requirements of the lists in employing the CRM [6]. Comparison lists allow for matching of similar items, such as people appearing on both lists, with a statistical analysis to determine the estimate of cases along with a confidence interval [3]. CRM can be used to estimate both incidence and prevalence in epidemiological studies [3]. For example, by obtaining cases from hospital records and the health department, two independent lists may exist for use by the CRM. If only a hospital list exists, the same method of evaluation for NC can be conducted in the geographical area served by the hospital to generate the second list. This is a low-cost and effective method for performing counts. It is suggested that researchers in neurological epidemiology use this method when conducting incidence and prevalence studies [7]. In many ways, the CRM may actually be the ‘gold standard’ for estimating the number of cases as well as providing an estimate of the undercount of the lists [8].