Purpose: To assess the comparative effectiveness, efficacy, and tolerability of lamotrigine (LTG) and sustained‐release carbamazepine (CBZ) in the treatment of newly diagnosed epilepsy in the elderly. Methods: Patients aged 65 years or older, who had experienced at least two unprovoked partial and/or generalized tonic–clonic seizures, were randomized to receive LTG (n = 93) or CBZ (n = 92) according to a multicenter double‐blind, parallel‐group design. Trial duration was 40 weeks and included a 4‐week dose escalation followed by a maintenance phase during which dosages could be adjusted according to response. Initial, maintenance and maximum dosages were 25 mg, 100 mg, and 500 mg per day for LTG, and 100 mg, 400 mg, and 2,000 mg per day for CBZ, respectively. The primary end point was retention in the trial. Results: In the LTG group, 68 patients (73%) completed the 40‐week study period compared with 61 (67%) in the CBZ group, a nonsignificant difference. Time to withdrawal from any cause did not differ between groups (p = 0.34). The number of subjects who completed the 40‐week period and were seizure free in the last 20 weeks was 48 (52%) in the LTG group and 52 (57%) in the CBZ group. Adverse events leading to withdrawal occurred in 13 (14%) subjects in the LTG group and 23 (25%) subjects in the CBZ group. Conclusion: LTG and CBZ showed comparable effectiveness, with a trend for higher seizure‐free rates for CBZ and better tolerability for LTG. Differences in outcome compared with previous trials may be related to different dosing rates and use of a sustained‐release formulation for CBZ.