Modelling the scorpion stings using surveillance data in El Bayadh Province, Algeria

Abstract

ObjectiveTo examine some epidemiological features of scorpion envenomations, analyse and interpret the recorded data, find any relationship between the incidence of scorpion stings and climatic factors, and finally develop a statistical model to estimate the variability among future cases in El Bayadh Province, Algeria. MethodsTo assess the effects of climate variability on the scorpion envenomations, we applied the count data regression models to the monthly recorded scorpion stings in El Bayadh Province from 2001 to 2012. ResultsThe epidemiological analysis of data revealed that scorpion stings occured mainly in rural areas, round the clock, all year long with the highest seasonal incidence in summer, and the lowest in winter, all ages with male predominance. The ends of upper and lower limbs were the most affected parts of the human body. The majority of cases (95.7%) were classified as mild envenomations and systemic toxicity was observed in 4.3% of cases. The use of count data regression models showed that the negative binomial regression was appropriate to forecast cases and the fitted data agreed considerably with the actual data. Moreover, the model had predicted the monthly scorpion sting cases for the year of 2013, with satisfactory accuracy. ConclusionsThis study shows an optimism for forecasting scorpion stings by modelling and calibration with surveillance data and climate information. This knowledge could help to contain any unusual situation and assist health decision-makers to strengthen the prevention and control measures and to be in a state of readiness.