In addition, to identify demographic and epilepsy-related variables associated with change in academic achievement from baseline to long-term follow-up, multiple linear regressions, using the enter method and pairwise deletion for missing variables, were conducted.