Abstracts

Rationale: Epilepsy is one of the most common neurological conditions worldwide. The purpose of this study was to estimate the international prevalence of epilepsy and explore sources of heterogeneity between studies. This study was part of a larger initiative funded by the Public Health Agency of Canada, the National Population Study of Neurological Conditions, to facilitate a better understanding of the burden of 15 neurological conditions in Canada and worldwide. Methods: MEDLINE and EMBASE databases (1985 to Dec 13, 2010) were searched using pre-specified MeSH terms and keywords related to seizures, epilepsy, convulsions, incidence, prevalence and epidemiology. Population-based studies reporting the incidence and/or prevalence of epilepsy were included. Studies focusing solely on febrile convulsions or seizures (not epilepsy) were excluded. Two reviewers independently reviewed all abstracts, full text articles and abstracted data. A third reviewer was consulted to resolve any disagreement. Reported prevalence was abstracted for each study. Median estimates were calculated. A random effects model was also used to pool the prevalence [95% confidence interval] per 1,000 people (after testing for heterogeneity) where feasible. Results: Embase and Medline identified 8061 and 5191 abstracts respectively, with 442 abstracts selected for full text review and 170 articles meeting all eligibility criteria. Globally the prevalence ranged from 1.55 per 1,000 in Hong Kong to 105 per 1,000 in Cameroon. Most studies used some variation of two or more clinical recurrent unprovoked seizures as their definition of epilepsy. An important source of heterogeneity was geographic location, and specifically developed vs. underdeveloped regions. The median estimates of prevalence for each continent were: 7.85 per 1,000 (range: 2-105) in Africa (n=30 studies), 4.88 per 1,000 (range: 1.54-77) in Asia (n=50 studies), 5.25 per 1,000 (range: 2.26-15.4) in Europe (n=50 studies), 7.15 per 1,000 (range: 3.4-28) in North America (n=28 studies) and 9.8 per 1,000 (range: 3.8-32) in South America (n=12 studies). There were no population-based studies examining the prevalence of epilepsy in Australia. The results of the meta-analysis and exploration of additional sources of heterogeneity will be presented. Conclusions: Epilepsy is common, particularly in developing countries, although gaps still exist regarding the global prevalence of epilepsy. Specifically, there was a noticeable lack of evidence from Australia and South America. The higher prevalence found in developing countries is consistent with prior studies, and is believed to be multifactorial, and likely due in part to possible genetic factors, but more significantly to the higher incidence of infections and head trauma in resource poor countries. In continents where much evidence exists, the focus should be shifted towards the investigation of sources of heterogeneity and the identification of modifiable risk factors and approaches to reduce the burden of epilepsy.