OBJECTIVE: Epilepsy is a common, chronic neurological disorder that disproportionately affects individuals living in low- and middle-income countries (LMICs), where the treatment gap remains high and adherence to medication remains low. Community health workers (CHWs) have been shown to be effective at improving adherence to chronic medications, yet no study assessing the costs of CHWs in epilepsy management has been reported. METHODS: Using a Markov model with age- and sex-varying transition probabilities, we determined whether deploying CHWs to improve epilepsy treatment adherence in rural South Africa would be cost-effective. Data were derived using published studies from rural South Africa. Official statistics and international disability weights provided cost and health state values, respectively, and health gains were measured using quality adjusted life years (QALYs). RESULTS: The intervention was estimated at International Dollars ($) 123 250 per annum per sub-district community and cost $1494 and $1857 per QALY gained for males and females, respectively. Assuming a costlier intervention and lower effectiveness, cost per QALY was still less than South Africa's Gross Domestic Product per capita of $13 215, the cost-effectiveness threshold applied. SIGNIFICANCE: CHWs would be cost-effective and the intervention dominated even when costs and effects of the intervention were unfavorably varied. Health system re-engineering currently underway in South Africa identifies CHWs as vital links in primary health care, thereby ensuring sustainability of the intervention. Further research on understanding local health state utility values and cost-effectiveness thresholds could further inform the current model, and undertaking the proposed intervention would provide better estimates of its efficacy on reducing the epilepsy treatment gap in rural South Africa.