Abstract:
Purpose: The paper aims to identify the best-fit trajectory model among selected model candidates
on the epilepsy dataset, emphasising time and demographic factors.
Design/Methods/Approach: The study employed secondary monthly count data from the District
Health Information Management System-2 (DHIMS-2) of the Ghana Health Service. The data was
analysed using zero-inflated Poisson (ZIP) regression, zero-inflated negative binomial (ZINB)
regression, and geometric regression (GR) models. These models were evaluated based on their
goodness-of-fit to the dataset.
Findings: The geometric regression (GR) model emerged as the best-fit model compared to the
ZIP and ZINB models. The results revealed that epilepsy incidence in the Saboba District is sex dependent, with a significant p-value of 0.0369. Furthermore, the study found that epilepsy
prevalence is relatively lower in younger children (0-9 years) and older adults (60-70+ years).
Participants within the economically active age group (18-59 years) were found to be at the highest
risk. Moreover, the year 2015 recorded the highest epilepsy incidence, while 2013 had the lowest.
Research Limitation: The study solely used a dataset from the Saboba District, Ghana. As a
result, its conclusion may not be broadly applicable.
Practical implications: The study provides crucial information for healthcare policymakers to
develop targeted interventions for epilepsy. It emphasizes the need to protect the rights of people
living with Epilepsy (PLWE), integrate risk factors into outreach programs, and improve access to
healthcare services. These findings can help health systems incorporate better treatment and care
strategies for high-risk populations.
Social implications: The study highlights the significant social impact of epilepsy, particularly on
the economically active population. It stresses the importance of addressing the stigma and
discrimination faced by PLWE through public education and awareness campaigns.
Originality /value: This study uses advanced statistical models to offer a vigorous methodological
lens for analysing epilepsy data, thus supporting future research and policy development in mental
health and epilepsy management in Ghana and beyond.
