Insomnia is common among cancer survivors, affecting an estimated 30%–60% of this population, with roughly 20% meeting diagnostic criteria for insomnia disorder. Cognitive behavioral therapy for insomnia (CBTi) is the recommended first-line treatment, but real-world access is constrained by costs and limited therapist availability. These barriers have driven interest in alternative delivery formats, including digital and telephone-based CBTi, as potential scalable options for oncology populations.
The authors performed a systematic review and a frequentist random-effects network meta-analysis (NMA). They searched PubMed, Embase, CINAHL and the Cochrane Library for randomized controlled trials published up to September 25, 2025. Eligible studies were randomized trials in adult cancer patients that evaluated any CBTi delivery mode and reported subjective sleep outcomes.
The prespecified primary outcome was the Insomnia Severity Index (ISI) score. Secondary outcomes included sleep onset latency (SOL), wake after sleep onset (WASO), total sleep time (TST), and sleep efficiency (SE). The study protocol was prospectively registered on PROSPERO (CRD42023429081).
A total of 22 randomized trials were included, comprising 2040 participants across the network. Details about individual trial characteristics, such as types of cancer beyond the meta-regression finding, specific intervention durations, and exact comparator descriptions, were provided in the original paper but are not fully restated here.
In the NMA comparing different delivery methods to treatment as usual (TAU), digital CBTi (dCBTi) produced the largest reduction in ISI scores, with a mean difference (MD) of -9.70 (95% CI -10.80 to -8.61) versus TAU. Telephone-delivered CBTi also yielded a clinically meaningful reduction in ISI (MD -8.74 versus TAU).
These effect estimates indicate that several nontraditional delivery modes of CBTi can reduce insomnia severity in cancer survivors relative to usual care, with digital formats showing the largest pooled ISI improvement in this analysis.
Compared with TAU, dCBTi was associated with statistically significant improvements across multiple objective sleep-related measures reported subjectively:
In addition, group CBTi significantly reduced SOL (MD -11.60 minutes) and WASO (MD -22.03 minutes) versus TAU. These findings suggest different CBTi delivery modes may differentially affect specific sleep parameters.
Meta-regression analyses explored potential moderators of treatment effects. The authors identified breast cancer diagnosis as a significant moderator of dCBTi outcomes for two secondary measures: SOL (coefficient 25.20, p = 0.015) and SE (coefficient -16.17, p = 0.007). The source reports these moderator coefficients and p-values but does not provide additional moderator details in the abstract.
The overall certainty of evidence across comparisons was rated low to very low. The primary reasons cited were imprecision in some estimates and network incoherence (inconsistency) for selected outcomes. Because of these limitations, the authors advise cautious interpretation of comparative effect sizes.
The abstract notes the need for higher-quality head-to-head trials to confirm comparative effectiveness; specific gaps such as the heterogeneity of cancer types, intervention fidelity, or long-term follow-up were not detailed in the abstract and therefore are not expanded on here.
Within the constraints of the included randomized evidence, digital CBTi (dCBTi) demonstrated the greatest potential for improving insomnia severity and multiple sleep parameters among cancer survivors when compared with treatment as usual. Telephone and group CBTi also showed benefits for selected outcomes.
However, because the certainty of evidence was low to very low and network inconsistency affected some outcomes, these results should be interpreted cautiously. The authors call for high-quality, direct head-to-head randomized trials comparing CBTi delivery modes in oncology populations to better inform clinical decision-making and implementation strategies.