A simple method for defining malaria seasonality

Background: There is currently no standard way of defining malaria seasonality, resulting in a wide range of definitions reported in the literature. Malaria cases show seasonal peaks in most endemic settings, and the choice and timing for optimal malaria control may vary by seasonality. A simple approach is presented to describe the seasonality of malaria, to aid localized policymaking and targeting of interventions. Methods: A series of systematic literature reviews were undertaken to identify studies reporting on monthly data for full calendar years on clinical malaria, hospital admission with malaria and entomological inoculation rates (EIR). Sites were defined as having 'marked seasonality' if 75% or more of all episodes occurred in six or less months of the year. A 'concentrated period of malaria' was defined as the six consecutive months with the highest cumulative proportion of cases. A sensitivity analysis was performed based on a variety of cut-offs. Results: Monthly data for full calendar years on clinical malaria, all hospital admissions with malaria, and entomological inoculation rates were available for 13, 18, and 11 sites respectively. Most sites showed year-round transmission with seasonal peaks for both clinical malaria and hospital admissions with malaria, with a few sites fitting the definition of 'marked seasonality'. For these sites, consistent results were observed when more than one outcome or more than one calendar year was available from the same site. The use of monthly EIR data was found to be of limited value when looking at seasonal variations of malaria transmission, particularly at low and medium intensity levels. Conclusion: The proposed definition discriminated well between studies with 'marked seasonality' and those with less seasonality. However, a poor fit was observed in sites with two seasonal peaks. Further work is needed to explore the applicability of this definition on a wide-scale, using routine health information system data where possible, to aid appropriate targeting of interventions.