A brief insomnia therapy worked during chemotherapy. Light glasses did not.

Insomnia and fatigue are among the most common complaints of women receiving chemotherapy for breast cancer, and they are among the least well served. Cognitive behavioural therapy for insomnia has strong evidence in breast cancer survivorship — a 2020 meta-analysis of 14 randomised trials covering 1,363 patients found pooled effect sizes favouring CBT-I immediately after treatment (Hedges' g = −0.779; 95% CI, −0.949 to −0.609), at short-term follow-up within six months (g = −0.653; 95% CI, −0.808 to −0.498) and at twelve months (g = −0.335; 95% CI, −0.532 to −0.139) [s2]. Almost all of that evidence comes from after treatment ends.

The SleepCARE trial, published in JAMA Network Open on 27 August, asked whether a brief version works during chemotherapy, and whether bright light therapy — cheaper, simpler, requiring no therapist — could do the same job [s1].

The design

SleepCARE was a six-week, 2×2 factorial, superiority, parallel randomised trial run at five metropolitan and regional hospitals in Australia between 22 January 2021 and 21 August 2024 [s1]. Participants were women aged 18 or older receiving chemotherapy for early or metastatic breast cancer [s1].

Two brief interventions were tested alone and in combination: CBT-I, and bright light therapy delivered through light glasses at 1,500 lux. That produced four groups — CBT-I alone, bright light alone, CBT-I plus bright light, and sleep hygiene education — with sleep hygiene education included in every arm [s1]. Each intervention ran six weeks and included a one-to-one consultation session, emails once or twice weekly, and a midpoint call for all groups, plus light glasses for the bright-light arms [s1].

The dual primary outcomes were the Insomnia Severity Index and the PROMIS-Fatigue T score, both patient-reported, assessed at baseline, three weeks, six weeks, and three and six months [s1]. Modified intention-to-treat analyses used latent growth models [s1].

What it found

Of 219 women enrolled (mean age 50.7 years, SD 10.8; 54, or 26.9%, with metastatic cancer), 55 were randomised to CBT-I, 55 to bright light, 52 to both and 57 to sleep hygiene education alone; 208 (95.0%) contributed data at some time point and were analysed [s1].

Insomnia symptoms improved more in the CBT-I groups than in the non-CBT-I groups: an ISI mean difference of −2.19 points (95% CI, −3.33 to −1.05; P = .002) [s1]. Fatigue did not: a PROMIS-Fatigue mean difference of −0.90 (95% CI, −3.08 to 1.28; P = .52) [s1].

Bright light therapy did neither. Compared with the non-bright-light groups it produced no difference in insomnia symptoms (ISI mean difference, −0.88; 95% CI, −2.02 to 0.26; P = .26) or fatigue (−0.71; 95% CI, −2.89 to 1.47; P = .52) [s1]. Comparable results emerged in the high-adherence subgroup and in the subgroups with high initial insomnia and fatigue symptoms [s1].

One exception is worth reporting and worth distrusting in equal measure. In exploratory subgroup analyses restricted to women with metastatic breast cancer, bright light improved both insomnia (ISI mean difference, −2.87; 95% CI, −5.06 to −0.67; P = .01) and fatigue (PROMIS-Fatigue mean difference, −5.16; 95% CI, −9.57 to −0.76; P = .02) [s1]. That subgroup numbered 54 women across four arms [s1]. It is a hypothesis, not a finding.

What it means and does not mean

The insomnia effect is real but small. A 2.19-point shift on the Insomnia Severity Index is narrower than what the survivorship meta-analysis records for full-length CBT-I after treatment ends [s2], which is consistent with a briefer intervention delivered during active chemotherapy — against a comparator that itself included sleep hygiene education in every arm [s1]. Sleep hygiene is not an inert placebo, and using it as the floor makes the CBT-I effect harder to detect, not easier.

The fatigue result is the more instructive one. Insomnia and cancer-related fatigue travel together and are routinely discussed as if treating one treats the other. Here, an intervention that measurably improved insomnia left fatigue where it was [s1]. That decoupling has been visible in the survivorship literature too, and this trial reinforces it under harder conditions.

The authors' conclusion is that brief CBT-I, but not bright light therapy, may reduce insomnia symptoms among women receiving chemotherapy for breast cancer [s1]. The trial was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12620001133921) [s1].

For anyone tracking the growing market in light-based sleep devices, the negative result is the headline. Light glasses at 1,500 lux, delivered with support and monitored for adherence, did not outperform sleep hygiene education in this population [s1]. That is one trial, in one clinical setting, with a chemotherapy schedule that disrupts everything including light exposure. It is not a verdict on bright light therapy in general. It is a clear negative in the specific place it was tested.

Sources

• [s1] "Cognitive Behavior vs Bright Light Therapy for Insomnia in Women Undergoing Chemotherapy for Breast Cancer: A Randomized Clinical Trial," JAMA Network Open, 27 August 2026. https://doi.org/10.1001/jamanetworkopen.2026.30320
• [s2] "Efficacy of cognitive behavioral therapy for insomnia in breast cancer: A meta-analysis," Sleep Medicine Reviews, 7 September 2020. https://doi.org/10.1016/j.smrv.2020.101376