Cognitive behavioural therapy helps with insomnia

A nurse-supported, self-directed cognitive behavioural therapy for insomnia (CBTi) appears to reduce insomnia severity and improve sleep outcomes among veterans with sleep difficulties, according to a study.

For the study, a total of 178 patients (mean age 55.1 years, 71.9 percent men) with insomnia disorder were recruited from a Veterans Affairs hospital. These patients were randomly assigned to undergo CBTi (intervention) or health education (control).

The nurse-supported, self-directed CBTi involved six weekly nurse phone contacts lasting approximately 20 minutes plus weekly readings from a treatment manual that covered typical CBTi content: sleep restriction, stimulus control, cognitive therapy, relaxation, and sleep hygiene. Health education, on the other hand, also involved six weekly nurse phone contacts lasting approximately 20 minutes plus a treatment manual that focused on a range of health topics but not sleep.

The primary outcome was the Insomnia Severity Index (score range, 0‒28; remission <8; differential improvement of 3 points targeted) score, which was evaluated at 8 weeks after randomization. Sleep outcomes, depression, fatigue, treatment response, and remission were also assessed as secondary outcomes.

At week 8, the intervention group had a significantly greater reduction in Insomnia Severity Index scores compared with the control group (estimated mean 5.7 vs 2.0 points; differential mean improvement, 3.7 points, 95 percent confidence interval [CI], −5.0 to −2.4; p<0.001). This difference remained significant through month 6 (mean, −2.8, 95 percent CI, −4.4 to −1.3; p<0.001).

Additionally, the intervention group had greater improvements in diary sleep onset latency, wake after sleep onset, and sleep efficiency and actigraphy sleep efficiency at week 8. The differential improvements were maintained through month 6.

Depression and fatigue were also significantly reduced at week 8, with the likelihood of treatment response and remission being greater in the intervention group than in the control group.