Objectives: Although cognitive–behaviour therapy for insomnia (CBT-I) has been shown to be effective for primary insomnia, the impact of CBT-I on insomnia comorbid with psychiatric conditions is largely unknown.

Design: A multiple baseline design with 2–3 weeks of baseline, 6 weeks of CBT-I, and a 3-week follow-up was employed.

Methods: Four patients with insomnia comorbid with both anxiety and depressive disorders participated. Sleep onset latency (SOL), wake time after sleep onset (WASO), total sleep time (TST), the Insomnia Severity Index (ISI), the Beck Anxiety Inventory (BAI), and the Beck Depression Inventory (BDI-II) were used as outcome measures.

Results: Improvements in SOL, WASO, and TST were observed for all the four patients. Based on ISI, there were reductions for all the patients (32–92%), three patients responded to treatment, and two patients remitted. Based on BAI, reductions on anxiety were observed for all the patients (40–76%), three patients were treatment responders, and two patients remitted. Based on BDI-II, all the patients experienced reductions in depression (36–53%), three patients responded to treatment, and one patient remitted.

Conclusions: CBT-I was able to reduce insomnia and co-morbid anxiety and depression, thus providing preliminary evidence for using CBT-I on insomnia co-morbid with psychiatric conditions.