Malaria Temporal Variation and Modelling Using Time-Series in Sussundenga District, Mozambique

Abstract Malaria is one of the leading causes of morbidity and mortality in Mozambique with the 5 th highest prevalence in the world. Sussundenga district in Manica Province has documented high P. falciparum incidence at the local rural health center (RHC). Sussundenga is a rural district located along the Mozambique, Zimbabwe border. P. falciparum transmission in this area is unique as there are differing control policies on each side of the international border. The study objective was to analyze the P. falciparum temporal variation and model its pattern in Sussundenga district, Mozambique. Data from weekly epidemiological bulletins (BES) were collected from 2015 to 2019, which records confirmed P. falciparum cases from health facilities. These data categorize confirmed cases into two age groups: under 5 years and 5-years and older. P. falciparum incidence and temporal variation were calculated. Temporal clusters were identified using dendrograms. A time-series analysis was carried out. For temporal modeling a Box-Jenkins method was used applying an autoregressive integrated moving average (ARIMA). Over the study period, 372,498 cases of P. falciparum were recorded in Sussundenga, 177,957 from under 5 years (47.5 %) and 194,541 (52.2 %) from 5 years and older. There were weekly and yearly variations in incidence overall (p < 0.001). There was a decreasing trend in cases for those under 5 years while there was a slight increase in those 5 and older. For those under 5, week 2 of the year had the highest number of cases (1170 Sd 34 ) while week 35 had the fewest (354 Sd 17.8). For those 5-years and older, cases also peak at week 2 (1295 Sd 245) with week 31 having the fewest cases (341 Sd 193.5). The findings indicate that cases are decreasing in those under 5 years and are increasing slightly in those 5 years and over. The P. falciparum case occurrence presents a weekly temporal pattern peaking during the wet season. Based on the temporal distribution and using ARIMA modelling, more efficient strategies that target this seasonality can be implemented to reduce the overall malaria burden in Sussundenga District, and regionally.