State–Space Forecasting of Schistosoma haematobium Time-Series in Niono, Mali

Background: Much of the developing world, particularly sub-Saharan Africa, exhibits high levels of morbidity and mortality associated with infectious diseases. The incidence of Schistosoma sp.—which are neglected tropical diseases exposing and infecting more than 500 and 200 million individuals in 77 countries, respectively—is rising because of 1) numerous irrigation and hydro-electric projects, 2) steady shifts from nomadic to sedentary existence, and 3) ineffective control programs. Notwithstanding the colossal scope of these parasitic infections, less than 0.5% of Schistosoma sp. investigations have attempted to predict their spatial and or temporal distributions. Undoubtedly, public health programs in developing countries could benefit from parsimonious forecasting and early warning systems to enhance management of these parasitic diseases. Methodology/Principal Findings: In this longitudinal retrospective (01/1996–06/2004) investigation, the Schistosoma haematobium time-series for the district of Niono, Mali, was fitted with general-purpose exponential smoothing methods to generate contemporaneous on-line forecasts. These methods, which are encapsulated within a state–space framework, accommodate seasonal and inter-annual time-series fluctuations. Mean absolute percentage error values were circa 25% for 1- to 5-month horizon forecasts. Conclusions/Significance: The exponential smoothing state–space framework employed herein produced reasonably accurate forecasts for this time-series, which reflects the incidence of S. haematobium–induced terminal hematuria. It obliquely captured prior non-linear interactions between disease dynamics and exogenous covariates (e.g., climate, irrigation, and public health interventions), thus obviating the need for more complex forecasting methods in the district of Niono, Mali. Therefore, this framework could assist with managing and assessing S. haematobium transmission and intervention impact, respectively, in this district and potentially elsewhere in the Sahel. Author Summary: Adequate forecasting and early warning systems are based upon observations of human behavior, population, disease time-series, climate, environment, and/or a combination thereof, whichever option best compromises among realism, feasibility, robustness, and parsimony. Fully automatic and user-friendly state–space forecasting frameworks, incorporating myriad options (e.g., expert opinion, univariate, multivariate, and spatial-temporal), could considerably enhance disease control and hazard mitigation efforts in regions where vulnerability to neglected tropical diseases is pervasive and statistical expertise is scarce. The operational simplicity, generality, and flexibility of state–space frameworks, encapsulating multiple methods, could conveniently allow for 1) unsupervised model selection without disease-specific methodological tailoring, 2) on-line adaptation to disease time-series fluctuations, and 3) automatic switches between distinct forecasting methods as new time-series perturbations dictate. In this investigation, a univariate state–space framework with the aforementioned properties was successfully applied to the Schistosoma haematobium time-series for the district of Niono, Mali, to automatically generate contemporaneous on-line forecasts and hence, providing a basis for local re-organization and strengthening public health programs in this and potentially other Sahelian districts.