Simulating the effect of evaluation unit size on eligibility to stop mass drug administration for lymphatic filariasis in Haiti

Abstract

Background The Transmission Assessment Survey (TAS) is a decision-making tool to determine when transmission of lymphatic filariasis is presumed to have reached a level low enough that it cannot be sustained even in the absence of mass drug administration. The survey is applied over geographic areas, called evaluation units (EUs); existing World Health Organization guidelines limit EU size to a population of no more than 2 million people. Methodology/Principal findings In 2015, TASs were conducted in 14 small EUs in Haiti. Simulations, using the observed TAS results, were performed to understand the potential programmatic impact had Haiti chosen to form larger EUs. Nine “combination-EUs” were formed by grouping adjacent EUs, and bootstrapping was used to simulate the expected TAS results. When the combination-EUs were comprised of at least one “passing” and one “failing” EU, the majority of these combination-EU would pass the TAS 79% - 100% of the time. Even in the case when both component EUs had failed, the combination-EU was expected to “pass” 11% of the time. Simulations of mini-TAS, a strategy with smaller power and hence smaller sample size than TAS, resulted in more conservative “passing” and “failing” when implemented in original EUs. Conclusions/Significance Our results demonstrate the high potential for misclassification when the average prevalence of lymphatic filariasis in the combined areas differs with regards to the TAS threshold. Of particular concern is the risk of “passing” larger EUs that include focal areas where prevalence is high enough to be potentially self-sustaining. Our results reaffirm the approach that Haiti took in forming smaller EUs. Where baseline or monitoring data show a high or heterogeneous prevalence, programs should leverage alternative strategies like mini-TAS in smaller EUs, or consider gathering additional data through spot check sites to advise EU formation. Author summary Lymphatic filariasis is a disease caused by roundworms that may lead to disability, psychological problems, stigma, and lowered quality of life. One of the key strategies to control and eliminate lymphatic filariasis is mass drug administration (MDA), or repeated treatment of all at-risk people living in affected areas with an annual dose of medicine. To determine whether MDA can be stopped in a particular area, a transmission assessment survey (TAS) is conducted whereby a sample of children are tested for filarial antigen and proportion with a positive result is compared against a target threshold. Existing guidelines for delimiting the geographic areas to conduct TAS permit large evaluation units. In 2015, TASs were conducted in Haiti using more stringent criteria for forming evaluation units, resulting in much smaller geographic areas for evaluation. Using simulations, the authors found that, had Haiti followed the existing guidelines and assessed larger geographic areas, many of the areas might have been misclassified and MDA stopped prematurely in some settings. This research suggests that caution is needed when forming evaluation units for TAS, especially if the prevalence of lymphatic filariasis is not uniform.