Spatial and Temporal Changes in Taenia solium Cysticercosis Seroprevalence among Pigs in the Southern Highlands of Tanzania

Taenia solium remains among the major causes of epilepsy in sub-Saharan Africa. Understanding the geospatial distribution of the parasite is important to inform control strategies. This study assessed spatiotemporal changes in the T. solium infection among pigs from a rural area in southwestern Tanzania. Secondary data from a previous repeated cross-sectional study in the Mbeya and Mbozi districts were used, involving sixteen villages, eight from each of the two districts. Serum samples were collected from 482, 460, and 421 pigs in 221, 196, and 139 households, respectively, at seven-month intervals. Ag-ELISA was used to analyse the samples for circulating antigens. Geographic coordinates of the pig corrals adjacent to the households were also recorded using a hand-held GPS device. SatScan v9.7 software was used to perform purely spatial scan analysis, based on the Bernoulli distribution model. Spatial clustering was assumed based on the relative risk of T. solium seropositivity. Statistical significance (set at 0.05) of the cluster was determined by comparing the log-likelihood ratio with the null distribution derived by Monte Carlo simulations. The window with the highest log-likelihood ratio was regarded as a primary cluster. Results showed significant clustering of T. solium cases with the presence of a single primary cluster during each phase. The relative risk of T. solium infection among pigs within the primary cluster areas ranged from 3.0 to 6.7. The radii of the clusters expanded from 1.83 to 27.7 km before shrinking to 1.5 km, and cluster location drifted from north to central and then to the southern part of the study area. These findings suggest that with restricted resources, control measures in the area could be focused on pockets of high transmission. However, changes in the size and location of clusters suggest that to be effective, interventions should be instituted soon after clusters are identified.