Table 1.

Studies comparing CBT-I to pharmacological therapies: methods

Study	Design	Patients	Intervention and duration	Comparison	Sleep measurements reported	Comment
Location	Quality	Longest follow-up
CBT-I vs. zopiclone
Sievertsen 2006 [29]	RCT	46 patients, age 55 and up	Individual CBT-I,6 weekly sessions	Zopiclone, 7.5 mg nightly	Sleep diaries, polysomnography	Study also included placebo group. Zopiclone patients had option to continue it after 6 week period.
Norway	5	12 months
						Daytime outcomes reported in [35]
CBT-I vs. zolpidem
Jacobs 2004 [30]	RCT	63 patients, age 25-64	Individual CBT-I, 5 sessions, 6 weeks; plus 1 telephone session	Zolpidem, see comment	Sleep diaries, sleep monitor	Dose 10 mg→5 mg→5 mg q2d over 6 week period.
USA	5	12 months
CBT-I vs. temazepam
Wu 2006 [32]	RCT	77 patients	Individual CBT-I 2 per week, 8 weeks	Temazepam, see comment	Sleep diaries, polysomnography	Dose 7.5 mg→30 mg→15 mg over 8 week period
China	3	8 months
						Study also included placebo and combined therapy groups
Morin 1999 [31]	RCT	78 patients, age 55 and up	Group CBT-I 8 weekly sessions	Temazepam, see comment	Sleep diaries, polysomnography	Dose 7.5 mg→30 mg as needed over 8 week period
						Study also included placebo and combined therapy groups.
Canada	6	24 months
						Adverse effects reported in [36] Patient attitudes reported in [37]
CBT-I vs. triazolam
McCluskey 1991 [33]	RCT	30 patients	Group CBT-I 2 per week, 3 weeks	Triazolam, 0.5 mg, then tapered to 0	Sleep diaries	Triazolam group also had weeklygroup meetings but no CBT-I
USA	4	9 weeks