Identifying malaria epidemic periods in Togo by health district and target group: a generalised additive model approach.

BACKGROUND

Information on seasonal malaria transmission is required for the development of targeted intervention programmes in malaria-endemic countries. This study aimed to determine the epidemic periods of malaria by health district and target group in Togo.

METHODS

Monthly data of confirmed malaria cases from 2013 to 2017 were analysed in this study. Data were routinely collected by the Togo National Malaria Control Programme (NMCP). They were aggregated by health district and target group (children < 5 years old, children ≥ 5 years old and adults, and pregnant women). Estimates of excess malaria cases compared to January were obtained through generalised additive models. The number of epidemic periods, the number of months with an excess of cases, the months with an excess of cases, the relative percentage increase of cases in the first month with an excess of cases, and the maximum relative percentage increase of cases and the corresponding month were described for each health district and target group.

RESULTS

A total of 5,522,650 confirmed malaria cases were reported from 2013 to 2017 in Togo. Children < 5 years old, children ≥ 5 years old and adults, and pregnant women represented 36.6%, 58.5% and 4.9% of the confirmed malaria cases, respectively. A time lag of at least one month was generally observed between the onset of precipitation and the start of the epidemic periods in all three target groups. In the health districts of the Savanes region, the epidemic periods started later in the year and had a greater relative increase in malaria cases compared to January than in the health districts of the other regions. In contrast, in the health districts of the Maritime and Lome-commune regions, the epidemic periods were generally short or undetectable.

CONCLUSIONS

This study suggests that malaria control interventions should be tailored to local transmission, considering the onset and duration of epidemic periods at the district level. These results can be used to adapt the distribution of seasonal malaria chemoprevention in children < 5 years old, and then used for malaria risk stratification in addition to other data.