However, epidemiological data have found that nearly 30% of patients with epilepsy achieve spontaneous remission without treatment, and AEDs can produce different degrees of adverse effects. The timing of drug administration is mainly considered in terms of reducing recurrence and improving prognosis. Berg et al. conducted a meta-analysis of 16 studies and found that the recurrence rate of first unexplained seizure-like epilepsy ranged from 23% to 71%, with a mean recurrence rate of 42% (95% CI 39%-44%), and that 60%-70% of patients had a recurrence within 6 months after the initial seizure. In contrast, patients with ≥2 seizures are generally considered to have a higher risk of recurrence than patients with a first seizure. The Italian “First Seizure Trial Group” found in a randomized, multicenter study that the risk of recurrence was 2.8 times higher in the delayed treatment group than in the immediate treatment group (95% CI 1.9 ~ 4.2). A meta-analysis of six papers by Wiebe et al [7] showed that treatment immediately after the first seizure reduced the recurrence rate by a mean of 34% (95% CI 15% – 52%), suggesting that treatment with AEDs immediately after the first seizure reduces the risk of early recurrence. The FIRST study showed that at 3-year follow-up, the “up to 1 or 2 years seizure-free” rates were 87% and 68% in the immediate treatment group and 83% and 60% in the delayed treatment group, respectively, with no statistically significant difference between the two groups. The “up to 2 years seizure-free” rate was 85% in the immediate treatment group and 86% in the delayed treatment group (P = 0,07); the “up to 5 years seizure-free” rate was 64% in both groups (P = NS). There was no statistically significant difference between 76% of patients in the immediate treatment group and 77% of patients in the delayed treatment group who were seizure-free in years 3-5; and there was no difference in quality of life, depression and anxiety, accidents, persistent epilepsy and serious complications between the two groups. The above large sample, multicenter randomized study suggests that immediate treatment of patients with single seizures or early infrequent seizures may reduce the recurrence rate in the short term (1 ~ 2 years), but does not affect their long-term prognosis. 3. Factors affecting seizure recurrence and prognosis (1) Factors affecting seizure recurrence In the early stages of epilepsy, it is instructive to clarify the factors affecting seizure recurrence to determine the timing of medication use. However, the factors affecting recurrence of epilepsy are complex and varied, and have not yet been fully established. The more definite factors include symptomatic epilepsy, EEG abnormalities especially focal epileptic wave abnormalities; possible factors are mainly persistent status epilepticus or cluster seizures, partial seizures, positive family history, late age of onset, seizures during sleep, and history of febrile convulsions. (2) Factors affecting the prognosis of epilepsy Factors affecting the prognosis include weekly seizures early in drug treatment (within 1 year) and weekly seizures before treatment, while seizure type, age at first seizure, and gender were not significantly correlated with prognosis. The prognosis was poor in patients with a high number of pre-treatment seizures. 4. Selection of the timing of initial treatment in patients with epilepsy Kim et al. summarized the risk score criteria for recurrence after non-induced seizures based on the MESS study, and the timing of treatment of AEDs was decided based on the results of these scores.