Home > Research > Publications & Outputs > Towards real-time spatio-temporal prediction of...
View graph of relations

Towards real-time spatio-temporal prediction of district-level meningitis incidence in sub-Saharan Africa

Research output: Contribution to journalJournal articlepeer-review

<mark>Journal publication date</mark>06/2014
<mark>Journal</mark>Journal of the Royal Statistical Society: Series C (Applied Statistics)
Issue number3
Number of pages18
Pages (from-to)661-678
Publication StatusPublished
Early online date4/11/13
<mark>Original language</mark>English


Within an area of sub-Saharan Africa termed ‘the meningitis belt’, meningococcal meningitis epidemics are a major public health concern. The epidemic control strategy currently utilised is reactive, such that a vaccination programme is initiated in a district once a pre-defined weekly incidence threshold is exceeded. In this paper we report progress towards the development of an early warning system based on statistical modelling of district-level weekly incidence data. Four modelling approaches are considered and their forecasting performances are compared using
weekly epidemiological data from Niger for the period 1986-2007. We conclude that the models under consideration are advantageous in different situations. The described three-state Markov model in which observed incidence is categorised according to policy-defined thresholds gives the most reliable short term forecasts, whereas the proposed dynamic linear model, using log-transformed weekly incidence as the response variable, gives more reliable predictions of annual epidemics.