Klimaeinflüsse und Masern

The impact of climate on measles incidence in Nigeria

Worldwide, WHO estimated around 110,000 measles deaths that occurred in 2017, mostly among children under the age of five [1]. The majority of these deaths occurred in low-income countries within Africa and Asia. Nigeria is among the priority countries with a significant contribution to global measles mortality. Irrespective of the measles vaccination campaigns going on in the country, measles remain a huge public health burden in Nigeria. It has been observed that the notified number of measles cases varies with the seasons of the year, with an increase in the number of cases in the dry season (December to May) compared to the rainy season (June to November) [2]. This suggests that weather patterns might be playing a significant role in the transmission of the measles virus in Nigeria.

This study aims (1) at modelling the effect of weather on weekly measles incidence in Nigeria, and (2) at building a prediction model for measles incidence.

We will be using the weekly surveillance data, daily meteorological data and monthly vaccination coverage data from Bauchi, Sokoto and Akwa-Ibom states of Nigeria from 2012 to 2018 in our analyses. The  meteorological variables  consist of amount  of rainfall,  temperature  (maximum  and  minimum),  air pressure,  relative  humidity,  wind speed and number of sunshine hours. Since the surveillance data is on a weekly scale, weekly averages will be computed for each meteorological variable. We will use data from two states to build the model and validate the model with data from the third state.

For objective (1), we will use Generalized Attentive Model (GAM) to model the relationship between weekly measles case count and the meteorological variables. For objective (2), we will evaluate the predictive performance of GAM, Long Short-Term Memory Recurrent Neural Network (LSTM) and Support-Vector Machine (SVM) in order to propose the best prediction model for measles case count.

We hope to gain a better understanding of how each meteorological variable is associated with measles incidence and propose a prediction model that could be automatized to assist stakeholders in directing public health resources in Nigeria.

The study is scheduled to be conducted from 2018 to 2020.

Literature

• [1] www.who.int/news-room/fact-sheets/detail/measles
• [2] Ibrahim BS, Usman R, Mohammed Y, et al. Burden of measles in Nigeria: a five-year review of casebased surveillance data, 2012-2016. Pan Afr Med J. 2019;32(Suppl 1):5. Published 2019 Jan 22. doi:10.11604/pamj.supp.2019.32.1.13564

PhD student involved

Silenou Bernard Chawo (PhD programme "Epidemiology")

Partners

• HZI
• Hannover Medical School (MHH), Hannover, Germany
• Nigeria Centre for Disease Control (NCDC), Nigeria