Less is more: Developing an approach for assessing clustering at the lower administrative boundaries that increases the yield of active screening for leprosy in Bihar, India.

Abstract

BackgroundIn India, leprosy clusters at hamlet level but detailed information is lacking. We aim to identify high-incidence hamlets to be targeted for active screening and post-exposure prophylaxis.MethodologyWe paid home visits to a cohort of leprosy patients registered between April 1st, 2020, and March 31st, 2022. Patients were interviewed and household members were screened for leprosy. We used an open-source app(ODK) to collect data on patients' mobility, screening results of household members, and geographic coordinates of their households. Clustering was analysed with Kulldorff's spatial scan statistic(SaTScan). Outlines of hamlets and population estimates were obtained through an open-source high-resolution population density map(https://data.humdata.org), using kernel density estimation in QGIS, an open-source software.ResultsWe enrolled 169 patients and screened 1,044 household contacts in Bisfi and Benipatti blocks of Bihar. Median number of years of residing in the village was 17, interquartile range(IQR)12-30. There were 11 new leprosy cases among 658 household contacts examined(167 per 10,000), of which seven had paucibacillary leprosy, one was a child under 14 years, and none had visible disabilities. We identified 739 hamlets with a total population of 802,788(median 163, IQR 65-774). There were five high incidence clusters including 12% of the population and 46%(78/169) of the leprosy cases. One highly significant cluster with a relative risk (RR) of 4.7(pConclusionsWe found a high yield of active household contact screening. Our tools for identifying high-incidence hamlets appear effective. Focusing labour-intensive interventions such as door-to-door screening on such hamlets could increase efficiency.