The Effects of Light Therapy on Cognitive Function and Stress in Women With Breast Cancer Before Systemic Treatment.

Cancer‐related cognitive impairment (CRCI), such as impairments in reaction time, processing speed, attention, working memory, verbal memory, and executive function [ 1 ], is a common and potentially long‐lasting side effect of breast cancer (BC) and its treatment [ 2 , 3 ]. Although previous research has primarily focused on the effects of chemotherapy on CRCI [ 4 ], CRCI has also been observed among women with BC prior to any treatment [ 5 , 6 ] and following BC surgery [ 7 , 8 ]. For instance, one study found that post‐operative women with BC were three to four times more likely to experience cognitive impairment compared with a healthy comparison (HC) group [ 9 ]. Additionally, a brain imaging study revealed structural changes in the thalamus of post‐operative women with BC, which were linked to attentional dysfunction (compared with HC) [ 8 ].

Although CRCI may arise from a myriad of complex and interconnected factors, psychological and biological stress are modifiable risk factors that may contribute to CRCI [ 6 , 10 , 11 , 12 ]. The experience of receiving a cancer diagnosis and undergoing treatment can be highly stressful and traumatic [ 13 ]. Consequently, depressive and cancer‐related stress (related to posttraumatic stress disorder [PTSD]) symptoms are common among women with BC [ 14 , 15 ] and have been associated with CRCI [ 6 , 16 , 17 ]. Biologically, acute and prolonged stress from cancer and its treatment can disrupt the stress reactivity systems, specifically the hypothalamic–pituitary–adrenal (HPA) axis and the autonomic nervous system [ 18 , 19 , 20 ], as reflected by their biomarkers, cortisol and α‐amylase [ 21 , 22 ]. Cortisol, a glucocorticoid hormone regulated by the HPA axis [ 21 ] and α‐amylase, a salivary enzyme [ 22 , 23 ], have both been linked to cognitive performance in patients with cancer [ 6 , 24 ]. Disruption in biological stress can affect (cortisol‐receptor‐dense) brain areas important for cognitive function, such as the prefrontal cortex and hippocampus. For instance, higher cortisol levels have been shown to independently predict poorer cognitive performance in post‐operative testicular patients with cancer [ 24 ], and elevated cortisol has been associated with worse cognitive performance in non‐cancerous populations [ 25 , 26 ]. Although the role of cortisol in cognitive function is well established [ 27 ], the role of α‐amylase is less studied. Recent studies have found α‐amylase in the hippocampus, where it is involved in glycogen metabolism, a process critical for energy supply during memory formation [ 28 ]. In our baseline study using the same sample [ 6 ], a flatter diurnal α‐amylase slope was associated with poorer overall cognitive performance. Additionally, elevated α‐amylase levels have been linked to mild cognitive impairment [ 29 ] and decreased cognitive performance [ 30 ] in various populations.

Surgical procedures may further contribute to CRCI by inducing a systemic stress response with increased cortisol production [ 31 ], through anesthesia [ 32 ], or by exacerbating circadian rhythm (CR) disruptions, which are common among women with BC [ 33 , 34 ]. These disruptions illustrate the bidirectional relationship between the stress response and circadian systems. Normally, the circadian system regulates the stress response, activating it before the day begins and reducing it at night, thus optimizing energy and adapting to daily cycles [ 35 ]. Both cortisol and α‐amylase follow specific circadian patterns, with cortisol peaking in the morning and gradually decreasing throughout the day [ 36 ], whereas α‐amylase shows the opposite pattern (with lower levels in the morning that gradually increase and peak in the afternoon) [ 23 ]. CR is generated in the suprachiasmatic nucleus (SCN) [ 37 ] and entrained to the environment via zeitgebers, the strongest being light [ 38 ]. When light enters the eye, it activates the non‐image forming photoreceptor system projecting to the SCN and other brain areas involved in alertness, influencing the sleep/wake cycle and cognitive function (see [ 39 ] for an overview).

Given that light is the strongest zeitgeber of the circadian system [ 38 ], light therapy emerges as a promising low‐burden treatment for CRCI. Light therapy has been shown to influence cognitive function among other participant groups [ 40 , 41 , 42 ]. Evidence suggests that light therapy can synchronize disrupted CR among institutionalized individuals with major neurocognitive disorder [ 43 , 44 ] and prevent CR from desynchronizing during chemotherapy for BC [ 45 ]. Additionally, studies have indicated that circadian‐effective bright light, can prevent fatigue, sleep quality, and quality of life from worsening among women with BC during chemotherapy, compared with non‐circadian stimulating dim red light [ 46 , 47 , 48 ]. Regarding the potential of light therapy to ameliorate stress, light therapy has been found to regulate cortisol levels among healthy individuals [ 49 , 50 ] and decrease PTSD symptoms among veterans, as well as augment cognitive behavioral therapy for individuals with PTSD and panic disorders [ 51 , 52 ]. Moreover, one study found that environmental bright light reduced the severity of depressive symptoms among hospitalized patients with multiple myeloma compared with circadian‐ineffective dim light (DL) [ 53 ].

The current study is the first (that we know of) to examine whether circadian‐effective blue light (BL) can mitigate the impact of CRCI following BC surgery using a circadian‐ineffective DL comparison condition. The primary aim of the present double‐blind randomized controlled trial was to test the efficacy of a 4‐week light therapy intervention on CRCI (i.e., overall cognitive performance assessed via global composite score) among post‐operative women with BC. We hypothesized that circadian‐effective BL would mitigate the negative impact of BC (i.e., the stress associated with the diagnosis, the surgical procedure, and associated factors) on cognitive function, predicting better overall cognitive performance, reaction time, processing speed, working memory, verbal memory, and fewer cognitive complaints in women receiving BL [ 54 , 55 ], compared with those receiving the circadian‐ineffective DL. Additionally, we expected that the BL would mitigate the negative effects of BC (i.e., the stress associated with the diagnosis, surgery, and associated factors) on biological and psychological stress markers, with the BL group exhibiting reduced biological (via steeper diurnal α‐amylase and cortisol slopes) and psychological stress (i.e., depressive symptoms, overall cancer‐related stress and its symptoms: hyperarousal, avoidance, and intrusion) compared with the DL group [ 49 , 50 , 51 , 53 ].

Materials and Methods

Participants

Eligible participants were newly diagnosed women with BC (Stages I–III) scheduled for surgery. All newly diagnosed women with BC in Iceland were invited to participate. Participants were excluded if they were scheduled to undergo a BC treatment other than surgery as their first treatment (i.e., neo‐adjuvant), if they had any contraindications for light therapy or a neurological condition that could influence neuropsychological functioning (e.g., autism diagnosis, epilepsy, and traumatic brain injury), age younger than 18, if they had other cancers or were undergoing treatment for another cancer, current pregnancy, pre‐existing anemia, shift work, eye diseases limiting the ability to process light, sleep disorders (e.g., narcolepsy), history of bipolar disorder/mania, travel to another time zone during the study, lack of access to a phone or a computer, and inability to understand or read Icelandic. All participants provided informed consent, which included information about the study examining the effects of different light intensities on the psychological side effects of BC and its treatment. See Figure 1 for the participant flow through the study and Table 1 for a more detailed sociodemographic overview of the participants. The study was approved by the chief medical officer at the National University Hospital, the National Bioethics Committee (VSN‐18‐199) of Iceland and the Icelandic Data Protection Authority.

Randomization and Blinding

The participants were randomly allocated to receive either BL or DL using a block randomization design. The research team members who administered the neuropsychological assessments were blinded to the group allocation of the participants.

Light Therapy

Light therapy was delivered via AYO glasses which safely emit light from LEDs 15 mm from the eye [ 56 ]. The circadian‐effective light glasses used by the experimental group (BL) emit narrowband light (peak λ = 470 nm) blue light with an irradiance of 250 μW/cm 2 (120 lx). The circadian system is maximally sensitive to short‐wavelength light peaking at 460 nm; therefore, the light level needed to have an effect on the circadian clock is much lower than white light sources, typically found in the built environment [ 56 , 57 ]. In contrast, the circadian‐ineffective sham glasses (DL) emit blue light with an irradiance of 2.5 μW/cm 2 (1.16 lx), which is below activation of the circadian clock. Participants were asked to wear the glasses for 30 min per day, 7 days per week for 4 weeks after surgery.

Neuropsychological Assessments

The neuropsychological tests were chosen based on recommendations from the International Cancer and Cognition Task Force [ 58 ] and available translations. Sustained attention and reaction time were measured with the 5‐min Psychomotor Vigilance Task [ 59 ], processing speed with the computerized Trail Making Test‐A (TMT‐A) [ 60 , 61 ], working memory with the Wechsler Adult Intelligence Scale‐Fourth Edition Digit Span subtest [ 62 ] and verbal memory with the Rey Auditory Verbal Learning Fluency Test [ 63 ].

Psychological Assessments

Cognitive complaints were assessed via Patient‐Reported Outcomes Measurement Information System (PROMIS)—Cognitive Function 8a [ 64 ]. T‐scores for cognitive complaints were calculated using Icelandic norms ( from the COVIDMENT project ) where higher scores indicated better self‐reported cognitive function. Depressive symptoms were measured with the Center for Epidemiological Studies Depression Scale (CES‐D) [ 65 ] and overall cancer‐related stress with the Impact of Events Scale‐Revised (IES‐R) [ 66 ] total score. As light therapy could potentially influence the cancer‐related stress symptoms differently, we conducted three separate path analyses for each of the three IES‐R subscales: intrusive thoughts, hyperarousal, and avoidance. Treatment credibility was measured at baseline with the first item of the Credibility/Expectancy Questionnaire (CEQ) [ 67 ] (i.e., “ How logical does the light therapy offered to you seem ”). The item “ How do you rate this treatment overall ?” from the Functional Assessment of Chronic Illness Therapy‐Treatment Satisfaction‐General (FACIT‐TS‐G) was utilized to assess treatment satisfaction [ 68 ]. Anxiety symptoms (included in Table 1 and when comparing completers with non‐completers in Supporting Information S1 ) were assessed with the Generalized Anxiety Disorder‐7 (GAD‐7) [ 69 ]. Unless otherwise specified, higher scores indicated more severe symptoms or traits. The internal consistency (Cronbach's α ) of the psychological assessments ranged from 0.88 (CES‐D) to 0.96 (PROMIS‐Cognitive Function).

Cortisol and α‐Amylase

See our previous baseline study [ 6 ] for a description of the saliva sampling and processing. The women with BC were instructed to refrigerate the saliva samples and bring them to their next scheduled hospital appointment, if applicable. A research team member retrieved the saliva samples either from the hospital, the home of the participants or from the post office if participants lived outside the greater capital area.

Covariates

Sociodemographic and Clinical Variables

Participants provided information regarding their sociodemographic background, including gender, age, education, relationship status, body mass index (BMI), physical activity, and menopause status. Age and education, commonly accounted for when testing cognitive function [ 70 ], were examined for their association with overall cognitive performance, reaction time, processing speed, working memory, and verbal memory. Additionally, we adjusted for age in the biological models as cortisol and α‐amylase responses can differ with age [ 71 , 72 ].

Clinical Variables

Clinical data regarding cancer stage, type of BC surgery, subsequent BC treatment, human epidermal growth factor receptor 2 (HER2), estrogen, and progesterone status were retrieved from the medical records of the participants. As BC stage (i.e., Stages I, II or III) and subsequent BC treatments (no further BC treatment (0), endocrine therapy (1) or chemotherapy/radiotherapy (2)) could potentially influence both cognitive function [ 73 ], as well as biological and psychological stress [ 74 ], we included these variables as covariates to adjust for their effects.

General Procedure

All assessments were conducted remotely due to COVID‐19 social restrictions and to limit participant burden. The self‐report assessment was conducted online via Research Electronic Data Capture (REDCap) [ 75 ] and the neuropsychological tests were conducted over the phone or on a computer (see [ 6 ] for more details). To minimize potential differences in chronotype, participants were allowed to choose their testing times (between 8:00 a.m. and 10:00 p.m.) for their participation in the neuropsychological assessment. Participants were instructed to select a time when they felt (or performed) their best, ideally between 10:00 a.m. and 04:00 p.m. The participants completed all baseline measurements before any BC treatment. The day after their BC surgery, participants were instructed to start light therapy while still in the hospital. The light therapy session using AYO glasses was administered through a light therapy smartphone application so the date and duration of the light sessions could be monitored. Participants were instructed to initiate the session as soon as possible after waking up, as light therapy is most effective shortly after awakening [ 76 ]. To ensure compliance, the smartphone application restricted the light therapy sessions to between 5:00 a.m. and 12:00 p.m. The light therapy session lasted for half an hour and participants were instructed to adhere to this treatment protocol daily for four consecutive weeks. After completing the 4‐week light therapy, participants were instructed to partake in the post‐light therapy assessments as soon as possible within 1 week.

Statistical Data Analysis

Descriptive statistics, independent samples t ‐tests and Fisher's tests were used to compare the BL and DL groups on sociodemographic, clinical and psychological variables. See our previous publication on the same sample [ 6 ] for how the initial global composite score, diurnal cortisol and α‐amylase slopes were calculated. For the sake of consistency, all neuropsychological z ‐scores were coded such that higher scores would reflect better performance. The post‐light therapy z ‐scores were calculated as follows [ 54 ]: Z − Score Post − LT = Raw score Post − LT − μ Baseline / σ Baseline where μ represents the mean at baseline and σ the baseline standard deviation. A postlight therapy global composite score was calculated with the mean z ‐scores of all neuropsychological test outcomes for participants with complete data. As the computerized neuropsychological tests were automated and undertaken remotely without supervision, time values exceeding 1.5 * interquartile range (IQR) from the mean (28 from PVT and 7 from the TMT‐A) were classified as outliers and excluded from the analyses. We corrected for multiple comparisons using the false discovery rate (FDR) with the Benjamini–Hochberg procedure.

In order to test the direct influence of treatment condition on post‐treatment function, while concurrently controlling for covariates, as well as handling missing data using maximum likelihood estimation (MLE) [ 77 , 78 ], we utilized the lavaan R package to perform path analyses [ 79 ]. The Yeo–Johnson transformation was utilized on skewed data using the caret package in R [ 80 ], as it is particularly effective for variables consisting of both positive and negative values (i.e., diurnal slopes and z ‐scores of the neuropsychological tests). As the responses to cognitive complaints and perceived treatment credibility were still skewed post transformation (most participants rated their perceived cognitive function and treatment credibility as high), they were categorized into binary variables (see Supporting Information S1 for details).

All statistical analyses were performed in R and conducted on an intention‐to‐treat (ITT) basis. Sensitivity analyses were conducted to test the robustness of the findings across different conditions, that is among women with BC with a minimum of 50% treatment adherence ( N = 79) and among those who participated during the winter ( N = 81), when daylight hours were under average globally, during October to March. Additionally, we conducted a complete case analysis (without MLE, to evaluate the impact of MLE handling the missing data) and lastly, compared the characteristics of completers to non‐completers (the only analysis not corrected for multiple testing).

First, analyses were conducted on autoregressive models containing only autoregressive paths from pre‐ to post‐light therapy for each outcome, and fit indices were examined. Second, the full model containing treatment condition and covariates was analyzed. To improve model fit, modification indices were examined to identify potential paths or relationships. Covariates were included based on theoretical rationale and their relationship with key outcomes. Third, fit indices between the autoregressive and full models were compared using likelihood ratio tests to determine the best‐fitting model. Model fit was assessed with standard indices for path analyses: a non‐significant χ 2 , the Comparative Fit Index (CFI > 0.90), Tucker–Lewis Index (TLI < 0.95), root mean square error approximation (RMSEA < 0.05) and standardized root mean residual (SRMR < 0.08) [ 81 ].

The path analysis models evaluated cognitive performance—overall cognitive performance, reaction time, processing speed, working memory, and verbal memory—each in separate models. These models (including cognitive complaints) included the direct effects of treatment group (BL compared with DL) and baseline cognitive function on post‐treatment cognitive function, while controlling for age, education, BC stage, subsequent BC treatment and perceived treatment credibility. As the completers and non‐completers in overall cognitive performance and overall cancer‐related stress significantly differed in cognitive complaints, depressive and anxiety symptoms (see Table S5 ), these were additionally controlled for (i.e., cognitive complaints, depressive, and anxiety symptoms in the overall cognitive performance path analysis, and cognitive complaints in the overall cancer‐related stress path analysis). Additionally, the models allowed both age and BC stage to correlate with subsequent BC treatment as BC treatment is often dependent on the BC stage and age (age may influence BC treatment decisions through factors, such as physiological tolerance, comorbidities, and risk tolerance). Age was also allowed to correlate with education (e.g., considering younger women may be more educated than older women). The biological and psychological stress models followed the same structure, except education was not included, and age was only included in the biological stress models.

Results

Participant Characteristics

A total of 117 women with BC (age range: 25–81) participated before any BC treatment and again after a 4‐week light therapy following BC surgery. Participants were randomly allocated to receive either a circadian‐effective BL ( N = 60) or circadian‐ineffective DL ( N = 57). Missing data ranged from 6% for cognitive complaints to 52% for salivary cortisol. Table 1 summarizes the main participant characteristics and shows that groups did not differ on any sociodemographic, clinical, biological, or psychological variables, except for perceived treatment credibility, p = 0.010 (see Table 1 ). To adjust for this group difference in perceived treatment credibility, the variable was included in the path analyses.

Model Comparisons

The autoregressive models, which consist solely of autoregressive paths (how pre‐light therapy levels predict post‐light therapy levels), consistently showed perfect fit indices, indicating an excellent fit to the data. The full models, incorporating the treatment condition (BL vs. DL) and theoretically relevant covariates, also demonstrated excellent fit indices across most outcomes. However, in the overall cognitive performance model, the TLI = 0.90 and RMSEA = 0.06. Similarly, in the intrusive thoughts model, all fit indices were excellent, except for RMSEA = 0.05, indicating very good fits rather than excellent ones.

For a comparison between the autoregressive models and the full models, see Table S1 in Supporting Information S1 . Likelihood ratio tests indicated no significant improvements in model fit (all p > 0.05) when the treatment condition and theoretically relevant covariates (i.e., age, education, BC stage, subsequent BC treatment, and perceived treatment credibility) were added, likely due to the already excellent fit of the autoregressive paths (as well as increased complexity). Efforts to simplify the model by removing variables (i.e., education, subsequent BC treatment, and BC stage) and their correlations, did not lead to a significantly better fitting model. Although the fit indices for the full models did not show significant improvements over the autoregressive models, we still interpreted the full models for two reasons. Firstly, because they included the treatment effects and secondly, as incorporating covariates is essential to adjust for confounding factors and ensure accurate estimates of the treatment effects on cognitive function and stress.

Path Analyses

To explore the efficacy of light therapy on cognitive function and stress among post‐operative women with BC, path analyses were conducted. These analyses tested the hypotheses that cognitive performance (i.e., overall cognitive performance, reaction time, processing speed, and working memory) would be better in the circadian‐effective BL compared with the circadian‐ineffective DL, that cognitive complaints would be reduced, and that psychological (i.e., depressive symptoms, overall cancer‐related stress, and its symptoms: hyperarousal, avoidance, and intrusion) and biological stress (via steeper diurnal α‐amylase and cortisol slopes) would also be lower.

Table 2 shows the differences between the BL and the DL groups in cognitive performance, cognitive complaints, as well as in psychological and biological stress after undergoing the 4‐week light therapy.

Post‐light therapy, there were no significant group differences in overall cognitive performance, processing speed, working memory, or verbal memory (FDR p > 0.05). However, there were non‐significant trends in the BL group for faster reaction time ( β = 0.24, SE = 0.19, z = 2.02, FDR p = 0.08), suggesting moderate effects ( d = 0.49), when compared with the DL group. Furthermore, participants in the BL group reported significantly fewer cognitive complaints ( β = 0.22, SE = 0.07, z = 2.79, FDR p = 0.011), reflecting a small effect ( d = 0.23). In terms of psychological stress markers, the BL group exhibited a non‐significant trend for fewer intrusive thoughts ( β = −0.16, SE = 0.05, z = −2.08, FDR p = 0.07) with small‐to‐moderate effects ( d = −0.36). No significant group differences were observed in overall cancer‐related stress, other cancer‐related stress symptoms (i.e., hyperarousal and avoidance), or depressive symptoms. Lastly, there were no significant group differences in biological stress markers (i.e., diurnal α‐amylase and cortisol slopes) between the groups (FDR p > 0.05). Path analysis results from sensitivity analyses, as well as the characteristics of the completers compared with non‐completers are provided in Supporting Information S1 .

Discussion

This is the first study to examine the efficacy of light therapy for CRCI among post‐operative women with BC in a double‐blinded randomized controlled trial. The hypothesis that circadian‐effective BL could mitigate the negative impact of BC surgery on cognitive performance compared with circadian‐ineffective DL was partially supported. Although no group differences were observed in overall cognitive performance, processing speed, working memory, or verbal memory, the BL group did demonstrate a non‐significant trend for faster reaction times, compared with the DL group. This aligns with prior studies suggesting that circadian‐effective (i.e., bright) light is associated with faster reaction times in the general population [ 55 , 82 , 83 ]. Additionally, the BL group also reported significantly fewer cognitive complaints (i.e., rated their own cognitive function better) compared with the DL group. Potential reasons why our study showed group differences (in reaction time and cognitive complaints), whereas the study of Wu et al. [ 54 ] found none could include (a) our larger sample, and (b) our use of DL instead of dim red light. As the authors themselves argued, long wavelength red light, such as employed in their study, may have enhanced the impact of subsequent short wavelength light (e.g., daylight) on cognitive function [ 54 , 84 ] (although this hypothesis needs further investigation).

The hypothesis that BL could mitigate the negative effects of BC surgery by reducing biological stress (via steeper diurnal cortisol and α‐amylase slopes) and psychological stress (i.e., depressive symptoms and cancer‐related stress) compared with DL was partially supported. The BL group demonstrated a non‐significant trend for fewer intrusive thoughts (a symptom of cancer‐related stress) compared with the DL group (including the results from the sensitivity analyses provided in Tables S2 and S4 ). This finding warrants further research as intrusive thoughts may be a risk factor for poorer patient outcomes and worse quality of life in women with BC [ 85 ]. The fact that previous evidence has shown that circadian‐effective (i.e., bright) light can decrease PTSD symptoms among other participant populations [ 51 , 52 , 86 ] underscores the need to study this further. However, in the current study, no significant group differences were found in overall cancer‐related stress or other cancer‐related stress symptoms, that is, avoidance or hyperarousal.

In line with previous results from another cancer patient group (Non‐Hodgkin Lymphoma survivors), we found no group differences in depressive symptoms [ 87 ]. This contrasts with Valdimarsdottir et al. [ 53 ] who reported that environmental bright light (in comparison with DL) alleviated depressive symptoms in hospitalized patients with multiple myeloma. These differences could largely be attributed to the more controlled hospital environment of their study, which may also have facilitated better treatment adherence due to the use of environmental light. The controlled hospital setting might have minimized confounding variables, such as exposure to natural daylight, which could have varied widely for our participants.

Regarding biological stress, the finding of comparable cortisol steepness in the BL and DL groups is consistent with previous results among patients with cancer [ 88 ]. Although a sensitivity analysis among women with BC with at least 50% treatment adherence showed that the BL group demonstrated flatter diurnal α‐amylase slopes relative to the DL group, this was neither observed in the main analysis or other sensitivity analyses, and therefore requires further investigation. Although previous research did not find effects of light on α‐amylase either [ 89 , 90 , 91 ], these are not directly comparable to ours as they examined acute effects of light on α‐amylase, unlike our 4‐week light therapy intervention. Long‐term light exposure, like in our study, may help shift or strengthen the CR robustness underlying the α‐amylase secretion [ 23 ].

The strengths of the current study include its broad scope of the efficacy of light therapy on cognitive function and stress among women with BC, consisting of both self‐reported and objective cognitive function, as well as both biological and psychological stress markers. The study design is another merit, as it involves a double‐blinded randomized controlled trial with a population‐based sample of women with BC (where all newly diagnosed women with BC in Iceland were invited to participate), featuring a novel comparison condition, and utilizing an ITT analysis.

Limitations include that only 79 participants (67.5%, out of N = 117) adhered to a minimum of 50% of the light therapy treatment protocol, which might have influenced the findings. The study took place during the COVID‐19 pandemic, which likely impacted treatment adherence by introducing significant psychological and logistical challenges for these newly diagnosed women with BC. Recruiting participants prior to BC surgery is particularly difficult due to the urgency of starting treatment and the heightened anxiety associated with both the diagnosis and treatment [ 92 ]. Due to this small sample size (especially in the saliva sampling), the findings should be interpreted with caution. However, despite challenges with missing data and treatment adherence, it is noteworthy that the main findings generally paralleled the sensitivity analyses, reinforcing the reliability of our findings despite these limitations. Additionally, although differences in chronotype, which may influence cognitive function [ 93 ], were not addressed, participants were allowed to select their cognitive assessment times to minimize potential chronotype‐related variations. Another limitation of the current study was that, due to small sample size, the mechanisms underlying the effects of BL on cognitive function and intrusive thoughts could not be examined. BL may influence cognitive function and intrusive thoughts through several potential mechanisms. For example, BL could synchronize CR, modulate stress systems, and/or improve sleep quality [ 94 ]. However, as changes in these domains would likely also affect depressive symptoms, an alternative explanation is that BL enhanced attentional control through increased alertness [ 95 ], potentially reducing intrusive thoughts (via interference inhibition) [ 96 ]. This latter mechanism may explain why BL was associated with faster reaction times, reduced cognitive complaints, and fewer intrusive thoughts, without affecting depressive symptoms.

Future studies with larger sample sizes are needed to confirm our findings and, if replicated, to explore the underlying mechanisms. These future studies should assess chronotype and be adequately powered to comprehensively explore the impact of light therapy on cognitive function, as well as biological and psychological stress markers. Additionally, extending this investigation by incorporating inflammatory markers would be beneficial. Although the current study examined CRCI post‐surgery, it is important to recognize that more severe cognitive impairments might occur later in the cancer treatment trajectory, such as during chemotherapy. Therefore, future studies (incorporating a DL comparison group) could investigate whether the effects of light therapy are more pronounced during these later stages, where the potential for observable improvement may be greater. Moreover, to enhance the practicality and applicability of light therapy further, future research should also focus on understanding the obstacles to treatment adherence and developing motivational incentives to boost adherence.

In conclusion, the findings in this study suggest that light therapy holds promise for mitigating the effects of BC (including the stress associated with the diagnosis, surgery and associated factors) on cognitive function (i.e., cognitive complaints and reaction time), as well as on intrusive thoughts. These findings emphasize the importance of early intervention in the cancer treatment trajectory, as that might result in better long‐term patient outcomes. The observation that simply wearing light therapy glasses daily for half an hour for 4 weeks may lead to, albeit small to modest, improvements in mood and cognition among women with BC is promising. This potential benefit, coupled with the non‐invasiveness, cost‐effectiveness, and minimal side effects of light therapy, warrants further investigation into its clinical and practical implications among patients with cancer.

Author Contributions

Snaefridur Gudmundsdottir Aspelund: conceptualization, investigation, funding acquisition, writing – original draft, writing – review and editing, visualization, validation, methodology, software, formal analysis, data curation, project administration. Thorhildur Halldorsdottir: conceptualization, investigation, funding acquisition, methodology, validation, visualization, formal analysis, software, project administration, writing – review and editing, data curation. Gudjon Agustsson: investigation, data curation. Hannah Ros Sigurdardottir Tobin: data curation, formal analysis, methodology, validation, visualization, investigation, writing – review and editing, software, project administration, funding acquisition. Lisa M. Wu: conceptualization, methodology, validation, visualization, writing – review and editing, formal analysis. Ali Amidi: conceptualization, methodology, validation, visualization, writing – review and editing. Kamilla R. Johannsdottir: conceptualization, methodology, validation, writing – review and editing. Susan K. Lutgendorf: methodology, validation, writing – review and editing, conceptualization. Rachel Telles: methodology, conceptualization, validation. Huldis Franksdottir Daly: conceptualization, investigation, methodology, data curation. Kristin Sigurdardottir: investigation, conceptualization, methodology. Mariana G. Figueiro: methodology, validation, writing – review and editing, conceptualization. William H. Redd: conceptualization, methodology, validation. Heiddis B. Valdimarsdottir: conceptualization, investigation, funding acquisition, methodology, validation, visualization, writing – review and editing, formal analysis, project administration, data curation, supervision, resources. Birna Baldursdottir: data curation, supervision, resources, conceptualization, investigation, funding acquisition, methodology, validation, visualization, writing – review and editing, formal analysis, project administration.

Funding

This work was funded by the Icelandic Research Fund (grant number 184999‐051), the Icelandic Cancer Society's Research Fund, and the Reykjavik University Research Fund.

Ethics Statement

The study conformed to the ethical standards of the Declaration of Helsinki and was approved by the chief medical officer at the National University Hospital, the National Bioethics Committee (VSN‐18‐199) of Iceland and the Icelandic Data Protection Authority.

Consent

All participants provided informed consent.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Fit indices for autoregressive and full models across all outcomes.

Table S2: Path analysis results for the treatment groups, the experimental group who received circadian‐effective BL compared to the comparison group who received the circadian‐ineffective DL, including only participants with a minimum of 50% treatment adherence .

Table S3: Path analysis results for the treatment groups, the experimental group who received circadian‐effective BL compared to the comparison group who received the circadian‐ineffective DL, on participants who participated during winter .

Table S4: Path analysis results for the treatment groups, the experimental group who received circadian‐effective BL compared to the comparison group who received the circadian‐ineffective DL, on complete data without maximum likelihood estimation to handle any missingness .

Table S5: Comparison of characteristics between completers (participants who met the minimum 50% light therapy adherence threshold, or provided data for the Global composite score, saliva sampling, computerized cognitive performance, and overall cancer‐related stress measures) and non‐completers (participants who did not).

Acknowledgments

We would like to sincerely thank all our participants and our collaborators at the University Hospital of Iceland. Furthermore, we would like to thank Dr. Harpa Lind Hjördísar‐Jónsdóttir et al. for providing us with the PROMIS‐Cognitive function norms for Iceland, as well as Kristine Wichuk for assisting with making and maintaining the CONSORT flowchart.

Data Availability Statement

Due to compliance with the Icelandic data protection laws and the approval conditions set by the National Bioethics Committee of Iceland for this study, the data cannot be publicly shared. However, deidentified data may be made available to researchers who submit a methodologically sound proposal. Notably, access to the data is contingent upon approval by the National Bioethics Committee of Iceland. Researchers interested in accessing the data should direct their proposals to heiddisb@ru.is .

References

1 J. C. Pendergrass , S. D. Targum , and J. E. Harrison , “ Cognitive Impairment Associated With Cancer ,” Innovations in Clinical Neuroscience 15 , no. 1–2 ( 2018 ): 36 – 44 . 29497579 PMC5819720

2 V. Koppelmans , M. Breteler , W. Boogerd , C. Seynaeve , C. Gundy , and S. Schagen , “ Neuropsychological Performance in Survivors of Breast Cancer More Than 20 Years After Adjuvant Chemotherapy ,” Journal of Clinical Oncology 30 , no. 10 ( 2012 ): 1080 – 1086 . 22370315 10.1200/JCO.2011.37.0189

3 M. Lange , F. Joly , J. Vardy , et al., “ Cancer‐Related Cognitive Impairment: An Update on State of the Art, Detection, and Management Strategies in Cancer Survivors ,” Annals of Oncology 30 , no. 12 ( 2019 ): 1925 – 1940 . 31617564 10.1093/annonc/mdz410 PMC8109411

4 G. Winocur , I. Johnston , and H. Castel , “ Chemotherapy and Cognition: International Cognition and Cancer Task Force Recommendations for Harmonising Preclinical Research ,” Cancer Treatment Reviews 69 ( 2018 ): 72 – 83 . 29909223 10.1016/j.ctrv.2018.05.017

5 M. Lange , I. Hardy‐Léger , I. Licaj , et al., “ Cognitive Impairment in Breast Cancer Patients Before Surgery: Results From a CANTO Cohort Subgroup ,” Cancer Epidemiology, Biomarkers & Prevention 29 ( 2020 ): 1759 – 1766 . 10.1158/1055-9965.EPI-20-0346 32611581

6 S. G. Aspelund , T. Halldorsdottir , G. Agustsson , et al., “ Biological and Psychological Predictors of Cognitive Function in Breast Cancer Patients Before Surgery ,” Supportive Care in Cancer 32 , no. 1 ( 2024 ): 5 , https://link.springer.com/epdf/10.1007/s00520‐023‐08282‐5 . 10.1007/s00520-023-08282-5 38185720

7 T. A. Ahles , A. J. Saykin , B. C. McDonald , et al., “ Cognitive Function in Breast Cancer Patients Prior to Adjuvant Treatment ,” Breast Cancer Research and Treatment 110 , no. 1 ( 2008 ): 143 – 152 . 17674194 10.1007/s10549-007-9686-5 PMC3114441

8 C. Sato , A. Sekiguchi , M. Kawai , et al., “ Postoperative Structural Brain Changes and Cognitive Dysfunction in Patients With Breast Cancer ,” PLoS One 10 , no. 11 ( 2015 ): e0140655 , https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4633203/ . 26536672 10.1371/journal.pone.0140655 PMC4633203

9 J. Debess , J. Ø. Riis , L. Pedersen , and M. Ewertz , “ Cognitive Function and Quality of Life After Surgery for Early Breast Cancer in North Jutland, Denmark ,” Acta Oncologica 48 , no. 4 ( 2009 ): 532 – 540 . 19096948 10.1080/02841860802600755

10 C. Andreotti , J. C. Root , T. A. Ahles , B. S. McEwen , and B. E. Compas , “ Cancer, Coping, and Cognition: A Model for the Role of Stress Reactivity in Cancer‐Related Cognitive Decline ,” Psycho‐Oncology 24 , no. 6 ( 2015 ): 617 – 623 . 25286084 10.1002/pon.3683 PMC4387099

11 T. A. Ahles and J. C. Root , “ Cognitive Effects of Cancer and Cancer Treatments ,” Annual Review of Clinical Psychology 14 , no. 1 ( 2018 ): 425 – 451 . 10.1146/annurev-clinpsy-050817-084903 PMC9118140 29345974

12 A. Papanastasiou , T. Seliniotaki , E. Rizos , et al., “ Role of Stress, Age and Adjuvant Therapy in the Cognitive Function of Patients With Breast Cancer ,” Oncology Letters 18 , no. 1 ( 2019 ): 507 – 517 . 31289522 10.3892/ol.2019.10361 PMC6540331

13 E. Dedert , E. Lush , A. Chagpar , et al., “ Stress, Coping, and Circadian Disruption Among Women Awaiting Breast Cancer Surgery ,” Annals of Behavioral Medicine 44 , no. 1 ( 2012 ): 10 – 20 . 22450856 10.1007/s12160-012-9352-y

14 W. Linden , A. Vodermaier , R. Mackenzie , and D. Greig , “ Anxiety and Depression After Cancer Diagnosis: Prevalence Rates by Cancer Type, Gender, and Age ,” Journal of Affective Disorders 141 , no. 2–3 ( 2012 ): 343 – 351 . 22727334 10.1016/j.jad.2012.03.025

15 L. Dinapoli , G. Colloca , B. Di Capua , and V. Valentini , “ Psychological Aspects to Consider in Breast Cancer Diagnosis and Treatment ,” Current Oncology Reports 23 , no. 3 ( 2021 ): 38 . 33709235 10.1007/s11912-021-01049-3 PMC7952347

16 Y. Yang and C. C. Hendrix , “ Cancer‐Related Cognitive Impairment in Breast Cancer Patients: Influences of Psychological Variables ,” Asia‐Pacific Journal of Oncology Nursing 5 , no. 3 ( 2018 ): 296 – 306 . 29963592 10.4103/apjon.apjon_16_18 PMC5996591

17 K. Hermelink , V. Voigt , J. Kaste , et al., “ Elucidating Pretreatment Cognitive Impairment in Breast Cancer Patients: The Impact of Cancer‐Related Post‐Traumatic Stress ,” Journal of the National Cancer Institute 107 , no. 7 ( 2015 ): 913443 , https://academic.oup.com/jnci/article/107/7/djv099/913443 . 10.1093/jnci/djv099 25882713

18 G. P. Chrousos and P. W. Gold , “ The Concepts of Stress and Stress System Disorders. Overview of Physical and Behavioral Homeostasis ,” Journal of the American Medical Association 267 , no. 9 ( 1992 ): 1244 – 1252 . 1538563

19 C. A. Stratakis and G. P. Chrousos , “ Neuroendocrinology and Pathophysiology of the Stress System ,” Annals of the New York Academy of Sciences 771 , no. 29 ( 1995 ): 1 – 18 . 8597390 10.1111/j.1749-6632.1995.tb44666.x

20 C. Wan , M. È. Couture‐Lalande , T. A. Narain , S. Lebel , and C. Bielajew , “ Salivary Alpha‐Amylase Reactivity in Breast Cancer Survivors ,” International Journal of Environmental Research and Public Health 13 , no. 4 ( 2016 ): 353 . 27023572 10.3390/ijerph13040353 PMC4847015

21 C. Kirschbaum and D. H. Hellhammer , “ Salivary Cortisol in Psychobiological Research: An Overview ,” Neuropsychobiology 22 , no. 3 ( 1989 ): 150 – 169 . 2485862 10.1159/000118611

22 N. Ali and U. M. Nater , “ Salivary Alpha‐Amylase as a Biomarker of Stress in Behavioral Medicine ,” International Journal of Behavioral Medicine 27 , no. 3 ( 2020 ): 337 – 342 . 31900867 10.1007/s12529-019-09843-x PMC7250801

23 U. M. Nater , N. Rohleder , W. Schlotz , U. Ehlert , and C. Kirschbaum , “ Determinants of the Diurnal Course of Salivary Alpha‐Amylase ,” Psychoneuroendocrinology 32 , no. 4 ( 2007 ): 392 – 401 . 17418498 10.1016/j.psyneuen.2007.02.007

24 A. Amidi , L. M. Wu , M. Agerbæk , et al., “ Cognitive Impairment and Potential Biological and Psychological Correlates of Neuropsychological Performance in Recently Orchiectomized Testicular Cancer Patients ,” Psycho‐Oncology 24 , no. 9 ( 2015 ): 1174 – 1180 . 25851607 10.1002/pon.3804

25 H. C. Abercrombie , J. Giese‐Davis , S. Sephton , E. S. Epel , J. M. Turner‐Cobb , and D. Spiegel , “ Flattened Cortisol Rhythms in Metastatic Breast Cancer Patients ,” Psychoneuroendocrinology 29 , no. 8 ( 2004 ): 1082 – 1092 . 15219660 10.1016/j.psyneuen.2003.11.003

26 B. K. Lee , T. A. Glass , M. J. McAtee , et al., “ Associations of Salivary Cortisol With Cognitive Function in the Baltimore Memory Study ,” Archives of General Psychiatry 64 , no. 7 ( 2007 ): 810 – 818 . 17606815 10.1001/archpsyc.64.7.810

27 D. Gaysina , M. P. Gardner , M. Richards , and Y. Ben‐Shlomo , “ Cortisol and Cognitive Function in Midlife: The Role of Childhood Cognition and Educational Attainment ,” Psychoneuroendocrinology 47 , no. 100 ( 2014 ): 189 – 198 . 25001968 10.1016/j.psyneuen.2014.05.018 PMC4103483

28 E. Byman , I. Martinsson , H. Haukedal , G. Gouras , K. K. Freude , and M. Wennström , “ Neuronal α‐Amylase Is Important for Neuronal Activity and Glycogenolysis and Reduces in Presence of Amyloid Beta Pathology ,” Aging Cell 20 , no. 8 ( 2021 ): e13433 . 34261192 10.1111/acel.13433 PMC8373367

29 N. Yamane , A. Ikeda , K. Tomooka , et al., “ Salivary Alpha‐Amylase Activity and Mild Cognitive Impairment Among Japanese Older Adults: The Toon Health Study ,” Journal of Prevention of Alzheimer's Disease 9 , no. 4 ( 2022 ): 752 – 757 . 10.14283/jpad.2022.51 36281680

30 F. B. Almstrup , M. Nordentoft , M. A. Jensen , et al., “ Associations Between Saliva Alpha‐Amylase, Heart Rate Variability, Saliva Cortisol and Cognitive Performance in Individuals at Ultra High‐Risk for Psychosis ,” Schizophrenia Research 255 ( 2023 ): 165 – 172 . 37001391 10.1016/j.schres.2023.03.041

31 A. Prete , Q. Yan , K. Al‐Tarrah , et al., “ The Cortisol Stress Response Induced by Surgery: A Systematic Review and Meta‐Analysis ,” Clinical Endocrinology 89 , no. 5 ( 2018 ): 554 – 567 . 30047158 10.1111/cen.13820

32 P. K. Mandal , V. Fodale , S. E. Mason , A. Noel‐Storr , and C. W. Ritchie , “ The Impact of General and Regional Anesthesia on the Incidence of Post‐Operative Cognitive Dysfunction and Post‐Operative Delirium: A Systematic Review With Meta‐Analysis ,” Journal of Alzheimer's Disease 22 , no. s3 ( 2010 ): S67 – S79 . 10.3233/JAD-2010-101086 20858956

33 V. Blakeman , J. L. Williams , Q. J. Meng , and C. H. Streuli , “ Circadian Clocks and Breast Cancer ,” Breast Cancer Research 18 , no. 1 ( 2016 ): 89 . 27590298 10.1186/s13058-016-0743-z PMC5010688

34 L. Liu , M. Rissling , A. Neikrug , et al., “ Fatigue and Circadian Activity Rhythms in Breast Cancer Patients Before and After Chemotherapy: A Controlled Study ,” Fatigue: Biomedicine, Health & Behavior 1 , no. 1–2 ( 2013 ): 12 – 26 . 10.1080/21641846.2012.741782 PMC3568994 23412418

35 K. L. Gamble , R. Berry , S. J. Frank , and M. E. Young , “ Circadian Clock Control of Endocrine Factors ,” Nature Reviews. Endocrinology 10 , no. 8 ( 2014 ): 466 – 475 . 10.1038/nrendo.2014.78 PMC4304769 24863387

36 S. Edwards , P. Evans , F. Hucklebridge , and A. Clow , “ Association Between Time of Awakening and Diurnal Cortisol Secretory Activity ,” Psychoneuroendocrinology 26 , no. 6 ( 2001 ): 613 – 622 . 11403981 10.1016/s0306-4530(01)00015-4

37 R. Y. Moore and V. B. Eichler , “ Loss of a Circadian Adrenal Corticosterone Rhythm Following Suprachiasmatic Lesions in the Rat ,” Brain Research 42 , no. 1 ( 1972 ): 201 – 206 . 5047187 10.1016/0006-8993(72)90054-6

38 D. B. Boivin , J. F. Duffy , R. E. Kronauer , and C. A. Czeisler , “ Sensitivity of the Human Circadian Pacemaker to Moderately Bright Light ,” Journal of Biological Rhythms 9 , no. 3–4 ( 1994 ): 315 – 331 . 7772798 10.1177/074873049400900311

39 A. S. Fisk , S. K. E. Tam , L. A. Brown , V. V. Vyazovskiy , D. M. Bannerman , and S. N. Peirson , “ Light and Cognition: Roles for Circadian Rhythms, Sleep, and Arousal ,” Frontiers in Neurology 9 ( 2018 ): 56 , https://www.frontiersin.org/articles/10.3389/fneur.2018.00056/full . 29479335 10.3389/fneur.2018.00056 PMC5811463

40 S. J. Kim , S. H. Lee , I. B. Suh , J. W. Jang , J. H. Jhoo , and J. H. Lee , “ Positive Effect of Timed Blue‐Enriched White Light on Sleep and Cognition in Patients With Mild and Moderate Alzheimer's Disease ,” Scientific Reports 11 , no. 1 ( 2021 ): 10174 . 33986349 10.1038/s41598-021-89521-9 PMC8119443

41 W. D. S. Killgore , J. R. Vanuk , B. R. Shane , M. Weber , and S. Bajaj , “ A Randomized, Double‐Blind, Placebo‐Controlled Trial of Blue Wavelength Light Exposure on Sleep and Recovery of Brain Structure, Function, and Cognition Following Mild Traumatic Brain Injury ,” Neurobiology of Disease 134 ( 2020 ): 104679 . 31751607 10.1016/j.nbd.2019.104679

42 G. Bersani , D. Marconi , L. Limpido , E. Tarolla , and E. Caroti , “ Pilot Study of Light Therapy and Neurocognitive Performance of Attention and Memory in Healthy Subjects ,” Psychological Reports 102 , no. 1 ( 2008 ): 299 – 304 . 18481690 10.2466/pr0.102.1.299-304

43 S. Ancoli‐Israel , J. L. Martin , D. F. Kripke , M. Marler , and M. R. Klauber , “ Effect of Light Treatment on Sleep and Circadian Rhythms in Demented Nursing Home Patients ,” Journal of the American Geriatrics Society 50 , no. 2 ( 2002 ): 282 – 289 . 12028210 10.1046/j.1532-5415.2002.50060.x PMC2764401

44 S. Ancoli‐Israel , P. Gehrman , J. L. Martin , et al., “ Increased Light Exposure Consolidates Sleep and Strengthens Circadian Rhythms in Severe Alzheimer's Disease Patients ,” Behavioral Sleep Medicine 1 , no. 1 ( 2003 ): 22 – 36 . 15600135 10.1207/S15402010BSM0101_4

45 A. B. Neikrug , M. Rissling , V. Trofimenko , et al., “ Bright Light Therapy Protects Women From Circadian Rhythm Desynchronization During Chemotherapy for Breast Cancer ,” Behavioral Sleep Medicine 10 , no. 3 ( 2012 ): 202 – 216 . 22742438 10.1080/15402002.2011.634940

46 N. Jeste , L. Liu , M. Rissling , et al., “ Prevention of Quality‐Of‐Life Deterioration With Light Therapy Is Associated With Changes in Fatigue in Women With Breast Cancer Undergoing Chemotherapy ,” Quality of Life Research 22 , no. 6 ( 2013 ): 1239 – 1244 . 22865153 10.1007/s11136-012-0243-2 PMC3780416

47 M. Rissling , L. Liu , S. D. Youngstedt , et al., “ Preventing Sleep Disruption With Bright Light Therapy During Chemotherapy for Breast Cancer: A Phase II Randomized Controlled Trial ,” Frontiers in Neuroscience 16 ( 2022 ): 815672 , https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.815872/full . 10.3389/fnins.2022.815872 PMC8959343 35356054

48 S. Ancoli‐Israel , M. Rissling , A. Neikrug , et al., “ Light Treatment Prevents Fatigue in Women Undergoing Chemotherapy for Breast Cancer ,” Supportive Care in Cancer 20 , no. 6 ( 2012 ): 1211 – 1219 . 21660669 10.1007/s00520-011-1203-z PMC3192914

49 C. M. Jung , S. B. S. Khalsa , F. A. J. L. Scheer , et al., “ Acute Effects of Bright Light Exposure on Cortisol Levels ,” Journal of Biological Rhythms 25 , no. 3 ( 2010 ): 208 – 216 . 20484692 10.1177/0748730410368413 PMC3686562

50 F. A. J. L. Scheer and R. M. Buijs , “ Light Affects Morning Salivary Cortisol in Humans ,” Journal of Clinical Endocrinology and Metabolism 84 , no. 9 ( 1999 ): 3395 – 3398 . 10487717 10.1210/jcem.84.9.6102

51 A. Kawamura , T. Yoshiike , A. Yoshimura , et al., “ Bright Light Exposure Augments Cognitive Behavioral Therapy for Panic and Posttraumatic Stress Disorders: A Pilot Randomized Control Trial ,” Sleep and Biological Rhythms 18 ( 2019 ): 101 – 107 , 10.1007/s41105-019-00248-7 .

52 S. D. Youngstedt , C. E. Kline , A. M. Reynolds , et al., “ Bright Light Treatment of Combat‐Related PTSD: A Randomized Controlled Trial ,” Military Medicine 187 , no. 3–4 ( 2022 ): e435 – e444 . 33511988 10.1093/milmed/usab014

53 H. B. Valdimarsdottir , M. G. Figueiro , W. Holden , et al., “ Programmed Environmental Illumination During Autologous Stem Cell Transplantation Hospitalization for the Treatment of Multiple Myeloma Reduces Severity of Depression: A Preliminary Randomized Controlled Trial ,” Cancer Medicine 7 , no. 9 ( 2018 ): 4345 – 4353 . 30099857 10.1002/cam4.1690 PMC6144249

54 L. M. Wu , H. B. Valdimarsdottir , A. Amidi , et al., “ Examining the Efficacy of Bright Light Therapy on Cognitive Function in Hematopoietic Stem Cell Transplant Survivors ,” Journal of Biological Rhythms 37 , no. 5 ( 2022 ): 471 – 483 . 35904252 10.1177/07487304221107833 PMC9645769

55 S. L. Chellappa , R. Steiner , P. Blattner , P. Oelhafen , T. Götz , and C. Cajochen , “ Non‐Visual Effects of Light on Melatonin, Alertness and Cognitive Performance: Can Blue‐Enriched Light Keep Us Alert? ,” PLoS One 6 , no. 1 ( 2011 ): e16429 . 21298068 10.1371/journal.pone.0016429 PMC3027693

56 J. M. Suls , C. Alfano , and C. Yap , “ Personalized (N‐Of‐1) Trials for Patient‐Centered Treatments of Multimorbidity ,” Harvard Data Science Review 4 , no. SI3 ( 2022 ): 5 , 10.1162/99608f92.d99e6ff5 . PMC10673634 38009131

57 G. Tosini , I. Ferguson , and K. Tsubota , “ Effects of Blue Light on the Circadian System and Eye Physiology ,” Molecular Vision 22 , no. 24 ( 2016 ): 61 – 72 . 26900325 PMC4734149

58 J. S. Wefel , J. Vardy , T. Ahles , and S. B. Schagen , “ International Cognition and Cancer Task Force Recommendations to Harmonise Studies of Cognitive Function in Patients With Cancer ,” Lancet Oncology 12 , no. 7 ( 2011 ): 703 – 708 . 21354373 10.1016/S1470-2045(10)70294-1

59 D. F. Dinges and J. W. Powell , “ Microcomputer Analyses of Performance on a Portable, Simple Visual RT Task During Sustained Operations ,” Behavior Research Methods, Instruments, & Computers 17 , no. 6 ( 1985 ): 652 – 655 .

60 R. M. Reitan , “ Validity of the Trail Making Test as an Indicator of Organic Brain Damage ,” Perceptual and Motor Skills 8 , no. 3 ( 1958 ): 271 – 276 .

61 D. L. Woods , J. M. Wyma , T. J. Herron , and E. W. Yund , “ The Effects of Aging, Malingering, and Traumatic Brain Injury on Computerized Trail‐Making Test Performance ,” PLoS One 10 , no. 6 ( 2015 ): e0124345 . 26060999 10.1371/journal.pone.0124345 PMC4465490

62 D. Wechsler , Wechsler Adult Intelligence Scale–Fourth Edition (WAIS–IV) , vol. 22 ( NCS Pearson , 2008 ), 498 .

63 A. Rey , L'examen Clinique en Psychologie ( Presses Universitaires de France , 1964 ).

64 J. S. Lai , L. I. Wagner , P. B. Jacobsen , and D. Cella , “ Self‐Reported Cognitive Concerns and Abilities: Two Sides of One Coin? ,” Psycho‐Oncology 23 , no. 10 ( 2014 ): 1133 – 1141 . 24700645 10.1002/pon.3522 PMC4185008

65 L. S. Radloff , “ The CES‐D Scale: A Self‐Report Depression Scale for Research in the General Population ,” Applied Psychological Measurement 1 , no. 3 ( 1977 ): 385 – 401 .

66 D. S. Weiss and C. R. Marmar , “ The Impact of Event Scale—Revised ,” in Assessing Psychological Trauma and PTSD ( Guilford Press , 1997 ), 399 – 411 .

67 G. J. Devilly and T. D. Borkovec , “ Psychometric Properties of the Credibility/Expectancy Questionnaire ,” Journal of Behavior Therapy and Experimental Psychiatry 31 , no. 2 ( 2000 ): 73 – 86 . 11132119 10.1016/s0005-7916(00)00012-4

68 J. D. Peipert , J. L. Beaumont , R. Bode , D. Cella , S. F. Garcia , and E. A. Hahn , “ Development and Validation of the Functional Assessment of Chronic Illness Therapy Treatment Satisfaction (FACIT TS) Measures ,” Quality of Life Research 23 , no. 3 ( 2014 ): 815 – 824 . 24062239 10.1007/s11136-013-0520-8

69 R. L. Spitzer , K. Kroenke , J. B. W. Williams , and B. Löwe , “ A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD‐7 ,” Archives of Internal Medicine 166 , no. 10 ( 2006 ): 1092 – 1097 . 16717171 10.1001/archinte.166.10.1092

70 A. Amidi , L. M. Wu , A. D. Pedersen , et al., “ Cognitive Impairment in Testicular Cancer Survivors 2 to 7 Years After Treatment ,” Supportive Care in Cancer 23 , no. 10 ( 2015 ): 2973 – 2979 . 25716340 10.1007/s00520-015-2663-3

71 J. Strahler , A. Mueller , F. Rosenloecher , C. Kirschbaum , and N. Rohleder , “ Salivary α‐Amylase Stress Reactivity Across Different Age Groups ,” Psychophysiology 47 , no. 3 ( 2010 ): 587 – 595 . 20070573 10.1111/j.1469-8986.2009.00957.x

72 C. A. Larsson , B. Gullberg , L. Råstam , and U. Lindblad , “ Salivary Cortisol Differs With Age and Sex and Shows Inverse Associations With WHR in Swedish Women: A Cross‐Sectional Study ,” BMC Endocrine Disorders 9 , no. 1 ( 2009 ): 16 . 19545400 10.1186/1472-6823-9-16 PMC2711063

73 S. J. Hardy , K. R. Krull , J. S. Wefel , and M. Janelsins , “ Cognitive Changes in Cancer Survivors ,” American Society of Clinical Oncology Educational Book 38 ( 2018 ): 795 – 806 . 30231372 10.1200/EDBK_201179

74 L. C. Brown , A. R. Murphy , C. S. Lalonde , P. D. Subhedar , A. H. Miller , and J. S. Stevens , “ Posttraumatic Stress Disorder and Breast Cancer: Risk Factors and the Role of Inflammation and Endocrine Function ,” Cancer 126 , no. 14 ( 2020 ): 3181 – 3191 . 32374431 10.1002/cncr.32934 PMC9069707

75 P. A. Harris , R. Taylor , R. Thielke , J. Payne , N. Gonzalez , and J. G. Conde , “ Research Electronic Data Capture (REDCap)—A Metadata‐Driven Methodology and Workflow Process for Providing Translational Research Informatics Support ,” Journal of Biomedical Informatics 42 , no. 2 ( 2009 ): 377 – 381 . 18929686 10.1016/j.jbi.2008.08.010 PMC2700030

76 J. J. Gooley , “ Treatment of Circadian Rhythm Sleep Disorders With Light ,” Annals of the Academy of Medicine, Singapore 37 , no. 8 ( 2008 ): 669 – 676 . 18797560

77 R. Larsen , “ Missing Data Imputation Versus Full Information Maximum Likelihood With Second‐Level Dependencies ,” Structural Equation Modeling: A Multidisciplinary Journal 18 , no. 4 ( 2011 ): 649 – 662 .

78 B. Muthén , D. Kaplan , and M. Hollis , “ On Structural Equation Modeling With Data That Are Not Missing Completely at Random ,” Psychometrika 52 , no. 3 ( 1987 ): 431 – 462 .

79 Y. Rosseel , T. D. Jorgensen , N. Rockwood , et al., “lavaan: Latent Variable Analysis [Internet],” ( 2023 ), https://cran.r‐project.org/web/packages/lavaan/index.html .

80 M. Kuhn , “ Building Predictive Models in R Using the Caret Package ,” Journal of Statistical Software 28 , no. 5 ( 2008 ): i05 , http://www.jstatsoft.org/v28/i05/ .

81 L. t. Hu and P. M. Bentler , “ Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives ,” Structural Equation Modeling: A Multidisciplinary Journal 6 , no. 1 ( 1999 ): 1 – 55 .

82 S. W. Lockley , E. E. Evans , F. A. J. L. Scheer , G. C. Brainard , C. A. Czeisler , and D. Aeschbach , “ Short‐Wavelength Sensitivity for the Direct Effects of Light on Alertness, Vigilance, and the Waking Electroencephalogram in Humans ,” Sleep 29 , no. 2 ( 2006 ): 161 – 168 . 16494083

83 S. A. Rahman , E. E. Flynn‐Evans , D. Aeschbach , G. C. Brainard , C. A. Czeisler , and S. W. Lockley , “ Diurnal Spectral Sensitivity of the Acute Alerting Effects of Light ,” Sleep 37 , no. 2 ( 2014 ): 271 – 281 . 24501435 10.5665/sleep.3396 PMC3900613

84 S. L. Chellappa , J. Q. M. Ly , C. Meyer , et al., “ Photic Memory for Executive Brain Responses ,” Proceedings of the National Academy of Sciences 111 , no. 16 ( 2014 ): 6087 – 6091 . 10.1073/pnas.1320005111 PMC4000819 24616488

85 A. Dupont , J. E. Bower , A. L. Stanton , and P. A. Ganz , “ Cancer‐Related Intrusive Thoughts Predict Behavioral Symptoms Following Breast Cancer Treatment ,” Health Psychology 33 , no. 2 ( 2014 ): 155 – 163 . 23379385 10.1037/a0031131 PMC3796187

86 A. K. Zalta , K. Bravo , Z. Valdespino‐Hayden , M. H. Pollack , and H. J. Burgess , “ A Placebo‐Controlled Pilot Study of a Wearable Morning Bright Light Treatment for Probable PTSD ,” Depression and Anxiety 36 , no. 7 ( 2019 ): 617 – 624 . 30995350 10.1002/da.22897 PMC6721597

87 D. E. J. Starreveld , L. A. Daniels , J. M. Kieffer , et al., “ Light Therapy for Cancer‐Related Fatigue in (Non‐)Hodgkin Lymphoma Survivors: Results of a Randomized Controlled Trial ,” Cancers 13 , no. 19 ( 2021 ): 4948 . 34638428 10.3390/cancers13194948 PMC8508131

88 J. A. Johnson , U. Subnis , L. E. Carlson , et al., “ Effects of a Light Therapy Intervention on Diurnal Salivary Cortisol in Fatigued Cancer Survivors: A Secondary Analysis of a Randomized Controlled Trial ,” Journal of Psychosomatic Research 139 ( 2020 ): 110266 . 33070045 10.1016/j.jpsychores.2020.110266

89 M. G. Figueiro and M. S. Rea , “ The Effects of Red and Blue Lights on Circadian Variations in Cortisol, Alpha Amylase, and Melatonin ,” International Journal of Endocrinology 2010 ( 2010 ): e829351 . 10.1155/2010/829351 PMC2905913 20652045

90 L. Sahin , B. M. Wood , B. Plitnick , and M. G. Figueiro , “ Daytime Light Exposure: Effects on Biomarkers, Measures of Alertness, and Performance ,” Behavioural Brain Research 274 , no. 1 ( 2014 ): 176 – 185 . 25131505 10.1016/j.bbr.2014.08.017

91 I. A. Ivanova , K. V. Danilenko , and L. I. Aftanas , “ Investigation of an Immediate Effect of Bright Light on Oxygen Consumption, Heart Rate, Cortisol, and α‐Amylase in Seasonal Affective Disorder Subjects and Healthy Controls ,” Neuropsychobiology 74 , no. 4 ( 2016 ): 219 – 225 . 28637032 10.1159/000477248

92 V. Jenkins , R. Thwaites , M. Cercignani , et al., “ A Feasibility Study Exploring the Role of Pre‐Operative Assessment When Examining the Mechanism of ‘Chemo‐Brain’ in Breast Cancer Patients ,” Springerplus 5 , no. 1 ( 2016 ): 390 . 27047716 10.1186/s40064-016-2030-y PMC4816933

93 K. Evansová , K. Červená , O. Novák , et al., “ The Effect of Chronotype and Time of Assessment on Cognitive Performance ,” Biological Rhythm Research 53 , no. 4 ( 2022 ): 608 – 627 .

94 S. Ancoli‐Israel , L. Liu , L. Natarajan , et al., “ Reductions in Sleep Quality and Circadian Activity Rhythmicity Predict Longitudinal Changes in Objective and Subjective Cognitive Functioning in Women Treated for Breast Cancer ,” Supportive Care in Cancer 30 , no. 4 ( 2022 ): 3187 – 3200 . 34957532 10.1007/s00520-021-06743-3 PMC8857013

95 M. Canazei , M. Dick , W. Pohl , et al., “ Impact of Repeated Morning Bright White Light Exposures on Attention in a Simulated Office Environment ,” Scientific Reports 13 , no. 1 ( 2023 ): 8730 . 37253767 10.1038/s41598-023-35689-1 PMC10229615

96 S. Kertzman , A. Avital , A. Weizman , and M. Segal , “ Intrusive Trauma Recollections Is Associated With Impairment of Interference Inhibition and Psychomotor Speed in PTSD ,” Comprehensive Psychiatry 55 , no. 7 ( 2014 ): 1587 – 1594 . 25023383 10.1016/j.comppsych.2014.05.004