Schistosomiasis transmission in Zimbabwe: modelling based on machine learning

Zimbabwe, located in Southern Africa, faces a significant public health challenge due to schistosomiasis. We investigated this issue by predicting the risk of schistosomiasis for the entire population. To this end, we reviewed available data on schistosomiasis in Zimbabwe between 1980 and 2022 from a literature search against the potential impact of 26 environmental and socioeconomic variables obtained from public sources. We studied the population requiring praziquantel with regard to whether or not mass drug administration (MDA) had been regularly applied. Three machine-learning algorithms were tested for their ability to predict the prevalence of schistosomiasis in Zimbabwe based on mean absolute error (MAE), root mean squared error (RMSE) and R2. The findings revealed different roles of the 26 factors with respect to transmission and there were particular variations between Schistosoma haematobium and S. mansoni infections. We found that the top-five correlation factors, such as the earlier period of time, unsettled MDA implementation, constrained economy, and high rainfall during the warmest season were closely associated with higher S. haematobium prevalence, while lower elevation, high rainfall during the warmest season, steeper slope, the earlier period of time and higher minimum temperature in coldest month were more closely related to higher S. mansoni prevalence. The random forest (RF) algorithm was considered as the formal best model construction method, with MAE=0.108 and 0.053; RMSE=0.143 and 0.082; and R2=0.517 and 0.458 for S. haematobium and S. mansoni, respectively. Based on this optimal model, the current schistosomiasis prevalence in Zimbabwe under MDA implementation was 19.8%, with that of S. haematobium at 13.8% and that of S. mansoni at 7.1%, requiring an annual MDA population of 3,003,928. Without MDA, the current schistosomiasis prevalence would be 23.2%, with that of S. haematobium at 17.1% and that of S. mansoni prevalence at 7.4%, requiring an annual MDA population of 3,521,466. The study reveals that MDA alone is insufficient for schistosomiasis elimination, especially S. mansoni. This study predicted a moderate prevalence of schistosomiasis in Zimbabwe. The elimination of schistosomiasis requires comprehensive control measures beyond the existing MDA strategies, including health education, snail control, population surveillance and environmental management.