This is a meta-analysis of randomized controlled trials (RCTs) comparing cognitive behaviour therapy for insomnia (CBT-I) monotherapy with active control treatment for insomnia in patients with medical or psychiatric comorbidities. Both international (PubMed, EMBASE, PsycINFO, Cochrane Library) and Chinese (WanFang, and CNKI) databases were systematically searched. The random effects model was used. Thirteen RCTs comparing CBT-I (n = 441) and active controls (n = 412) groups were included. CBT-I group showed significant advantage over active controls at post-treatment assessment in terms of Insomnia Severity Index (ISI; SMD = -0.74), sleep onset latency (SMD = -0.36), wake after sleep onset (SMD = -0.21), sleep quality (SMD = 0.56), Pittsburgh sleep quality index total scores (PSQI; SMD = -0.76) and the total score of dysfunctional beliefs and attitudes about sleep scale (DBAS; SMD = -1.09). Subgroup analyses revealed significant improvement in sleep onset latency in patients with psychiatric disorders (SMD = -0.45), while significant reduction of number of wakeup after sleep onset was found in patients with medical conditions (SMD = -0.31). This meta-analysis found that CBT-I monotherapy had greater efficacy than other active control treatment for insomnia in patients with medical or psychiatric comorbidities.