A voice-activated cognitive behavioral therapy program led to an 8.4-point reduction in insomnia severity among breast cancer survivors, according to a recent study.

A randomized clinical trial assessed the efficacy of a voice-activated cognitive behavioral therapy for insomnia (CBT-I) among breast cancer survivors. Seventy-six women with insomnia (Insomnia Severity Index [ISI] score >7) were randomized to either a voice-interactive CBT-I intervention delivered via a smart speaker or an educational control for 6 weeks. The intervention group used a personalized program with tailored feedback and daily voice-based interaction, while the control group received access to a static website containing general educational content about CBT-I, sleep, and cancer survivorship. Unlike the intervention, the control content was non-interactive and lacked personalized guidance, with participants able to explore the site freely without tailored support. Participants in the intervention group showed a significant reduction in ISI scores compared to the control group.

Secondary outcomes demonstrated improvements in sleep diary metrics for the intervention group, including sleep quality, wake after sleep onset, sleep onset latency, and sleep efficiency, although no significant change was observed in total sleep time. The intervention group reported a mean System Usability Scale (SUS) score of 72.2, and 85.7% of participants indicated satisfaction and would recommend the program.

The study, published in JAMA Network Open, had a population mean age of 61.2 years, and participants were an average of 9.6 years post breast cancer diagnosis. These findings suggest that a voice-interactive, in-home CBT-I program may improve insomnia symptoms in breast cancer survivors and may be applicable for clinical settings.

Although the study aimed to recruit from a hospital with a significant Black population and lower socioeconomic status, 69.7% of participants were White, and 86.8% had completed college or graduate education, which may limit the generalizability of the findings to other groups. The study only assessed outcomes over a 6-week period, making longer-term effects unknown. The use of self-reported sleep diaries introduces potential bias due to recall inaccuracies; however, prior research supports the reliability and validity of these diaries, and data entry was restricted to prevent backfilling, helping reduce recall bias. 

Analyses of secondary data indicated that participants with more incomplete diaries reported smaller improvements in sleep quality, which may reflect reporting bias and suggest conservative effect size estimates. While the smart speaker program was considered generally user-friendly and appropriate, variability in SUS scores points to areas for enhancement, with technical issues involving study hotspots possibly affecting usability ratings.

Full disclosures can be found in the published study.