OBJECTIVE Juvenile myoclonic epilepsy (JME) is a widely accepted epilepsy syndrome. Factors that are potentially predictive for seizure outcome in JME are not entirely clear yet. In the current study, we tried to identify factors that are potentially predictive for seizure outcome in patients with JME. METHODS In this case-control study all patients with a clinical diagnosis of JME were recruited at the outpatient epilepsy clinic at Shiraz University of Medical Sciences from 2008 to 2012. All patients were followed for at least 18 months. We divided the patients into two groups: patients who were seizure free in the last 12 months of their follow-up period and those who had any seizures (i.e., generalized tonic-clonic (GTC), myoclonic or absences). RESULTS 116 patients were studied. Sixty-eight patients were seizure free in the last 12 months of their follow-up and 48 patients had at least one seizure of any type. Shorter follow-up period, drugs other than valproate in the drug regimen, and poor drug adherence were significantly more frequent among those who were not seizure free. CONCLUSIONS A correct diagnosis of JME does not guarantee an optimal response to therapy. Poorer drug adherence might result in more seizures; this should be highlighted for patients in every single visit. Longer follow-up gives the treating physician more time to adjust the drug(s) and formulate a plan that is appropriate for patients with difficult to treat seizures. We recommend valproate if seizure control is the main objective.