Prevalence and Risk Factors Associated with Epilepsy in Six Health Districts of Mali: A Community-Based Cross-Sectional and Nested Case-Control Study

Abstract

Introduction: In resource-limited countries, epilepsy prevalence is underestimated and little is known about its risk factors. Objectives: This study aimed to determine the prevalence and risk factors for epilepsy in six health districts (HDs) in Mali. Methods: A community-based cross-sectional and nested case-control study was conducted in 180 villages with the highest number of suspicious epilepsy cases (SECs) in the six study HDs. The SECs were observed as part of a Phase 1 screening conducted by community health workers. For the nested case-control study, one case was matched with at least one control based on residence and age. A case of epilepsy was a person diagnosed with convulsive epilepsy after clinical assessment by a neurologist. A control was a person diagnosed as normal after neurological assessment by a neurologist. Data were collected on sociodemographic characteristics, familial and medical history of epilepsy, consanguinity, place of delivery, preterm birth, length/type of delivery, and history of meningitis and cerebral malaria. A univariate and multivariate binomial logistic regression model was used to analyse factors associated with epilepsy. Results: A total of 1,506 cases of epilepsy and 2,199 controls were enrolled in six HDs. The mean prevalence of epilepsy was 2‰, with the highest in Kenieba (3‰), a previously meso-endemic-onchocerciasis HD, and the lowest in Kadiolo (1.5‰), a hypo-endemic-onchocerciasis HD. Age (adjusted odds ratio [aOR] = 1.02 [95% confidence intervals [CI] 1.02–1.03]), history of cerebral malaria (aOR = 11.41 [95% CI 8.86–14.85]), history of meningitis (aOR = 1.95 [95% CI 1.16–3.29]), living in the HD of Tominian (aOR = 1.69 [95% CI 1.29–2.22]), delayed delivery (aOR = 3.21 [95% CI 2.07–5.07]), and dystocia (aOR = 3.37 [95% CI 2.03–5.73]) were all significantly associated with epilepsy. Conclusion: The prevalence of epilepsy (3‰) in a previously meso-endemic-onchocerciasis HD was much lower than the prevalence (13.35‰) documented in onchocerciasis endemic areas in 2,000. This decrease epilepsy prevalence in the previously meso-endemic region was induced by onchocerciasis, and the reduction was due to an effective community direct treatment with ivermectin programme. Cerebral malaria and obstetrical complications were the main risk factors for epilepsy and interventions improving malaria prevention/treatment and optimizing prenatal and obstetrical care need to be implemented to reduce incidence.