Forecasting paediatric malaria admissions on the Kenya Coast using rainfall

Area of Science:

• Epidemiology
• Climate Science
• Public Health

Background:

• Malaria remains a significant threat to child survival in Africa, transmitted by mosquitoes and influenced by environmental factors.
• Rainfall patterns, particularly along the East African coast, are critical determinants of seasonal malaria transmission.

Purpose of the Study:

• To develop a reliable forecasting model for paediatric malaria admissions at Kilifi District Hospital (KDH).

Main Methods:

• Statistical time-series models were employed to analyze the temporal relationship between monthly paediatric malaria admissions, rainfall, and sea surface temperatures.
• Models considered trend, seasonality, and long-term anomalies to predict malaria admission rates.

Main Results:

• Paediatric malaria admissions at KDH can be forecast using a model based on previous admission proportions.
• Incorporating Indian Ocean Dipole data or recent rainfall significantly improved prediction accuracy.

Conclusions:

• Time-series prediction models utilizing surveillance data can anticipate seasonal malaria burdens in stable transmission zones.
• These models can assist in strategically timing malaria vector control interventions.