OBJECTIVE: To describe the magnitude and variation of the epilepsy treatment gap worldwide. METHODS: We conducted a systematic review of the peer-reviewed literature published from 1 January 1987 to 1 September 2007 in all languages using PubMed and EMBASE. The purpose was to identify population-based studies of epilepsy prevalence that reported the epilepsy treatment gap, defined as the proportion of people with epilepsy who require but do not receive treatment. Negative binomial regression models were used to assess trends and associations. FINDINGS: The treatment gap was over 75% in low-income countries and over 50% in most lower middle- and upper middle-income countries, while many high-income countries had gaps of less than 10%. However, treatment gaps varied widely both between and within countries. They were significantly higher in rural areas (rate ratio, RR: 2.01; 95% confidence interval, CI: 1.40-2.89) and countries with lower World Bank income classification (RR: 1.55; 95% CI: 1.32-1.82). There was no significant trend in treatment gap over time (RR: 0.92; 95% CI: 0.79-1.07). CONCLUSION: There is dramatic global disparity in the care for epilepsy between high- and low- income countries, and between rural and urban settings. Our understanding of the factors affecting the treatment gap is limited; future investigations should explore other potential explanations of the gap.