Context: Therapeutic drug monitoring (TDM) allows personalizing the dose of valproic acid in patients with epilepsy to optimize drug therapy, minimize adverse effects and detect interactions. Aims: To determine valproic acid concentrations in serum samples from people with epilepsy and to analyze its clinical implications. Methods: Cloned donor enzyme immunoassay; descriptive, cross-sectional, non-randomized, convenience recruitment study of 57 voluntary patients with epilepsy (n = 39 male, 68.42%; n = 18 female, 31.58%) aged between 19 and 62 years. After three months of treatment with valproic acid, a single blood sample was collected from each volunteer at a minimal concentration. Results: Serum drug concentrations 51.30-100.10 mg/L (SD 5.94) and level/dose 2.17-5.31 (SD 1.14) were observed. Association was shown between the dose ratio/dose of valproic acid (R2 = 0.8693; p<0.05) and the Mann-Whitney U test (p<0.05). Valproic acid monotherapy and association with carbamazepine and phenytoin are not different between treatment groups (Mann-Whitney U test: p = 0.391 > α = 0.05). Conclusions: Serum valproic acid concentrations are within the therapeutic range, and there is a significant inverse linear correlation between dose ratio/dose, which must be considered to personalize the dose and optimize the pharmacotherapeutic result.