Behavioral treatment of insomnia in active-duty service members with traumatic brain injury: study protocol for a randomized clinical trial.

Cognitive behavioral therapy for insomnia (CBT-I) has been shown to improve both sleep and depressive symptoms, but predictors of depression outcome following CBT-I have not been well examined. This study investigated how chronotype (i.e., morningness-eveningness trait) and changes in sleep efficiency (SE) were related to changes in depressive symptoms among recipients of CBT-I. Included were 419 adult insomnia outpatients from a sleep disorders clinic (43.20% males, age mean ± standard deviation = 48.14 ± 14.02). All participants completed the Composite Scale of Morningness and attended at least 4 sessions of a 6-session group CBT-I. SE was extracted from sleep diary; depressive symptoms were assessed using the Beck Depression Inventory (BDI) prior to (Baseline), and at the end (End) of intervention. Multilevel structural equation modeling revealed that from Baseline to End, SE increased and BDI decreased significantly. Controlling for age, sex, BDI, and SE at Baseline, stronger evening chronotype and less improvement in SE significantly and uniquely predicted less reduction in BDI from Baseline to End. Chronotype did not predict improvement in SE. In an insomnia outpatient sample, SE and depressive symptoms improved significantly after a CBT-I group intervention. All chronotypes benefited from sleep improvement, but those with greater eveningness and/or less sleep improvement experienced less reduction in depressive symptom severity. This suggests that evening preference and insomnia symptoms may have distinct relationships with mood, raising the possibility that the effect of CBT-I on depressive symptoms could be enhanced by assessing and addressing circadian factors.

Introduction Disordered circadian rest/activity rhythms (RARs) and insomnia frequently co-occur. Aspects of cognitive behavioral therapy for insomnia (CBTI) aim at improving circadian RARs by establishing regular sleep times and consolidating sleep periods. Timing of sleep periods has also been associated with insomnia, wherein evening chronotype has been linked to greater sleep disturbance. However, there is limited research exploring whether CBTI improves insomnia via improved structure and timing of circadian RARs. We examined whether CBTI reduces insomnia severity via improved RAR structure or shifted RAR timing. Methods Our community sample of persons with a history of traumatic brain injury (TBI) and comorbid insomnia (N=38, Mage=38+/-12, 37% female) were randomized to a 4-session treatment group: CBTI or sleep education. Circadian RARs were continually estimated with actigraphy. Variables estimating RAR structure included interdaily stability (IS), intradaily variability (IV), and relative amplitude (RA). RAR timing was estimated as the sleep period midpoint. Effects of CBTI on RAR variables were estimated as the group difference in change in RAR variables from baseline to treatment end. Effect of CBTI on insomnia severity was estimated as the group difference in Insomnia Severity Index (ISI) score change from baseline to 12-weeks post-treatment. Secondary analyses explored whether RAR change moderated effects of treatment group on ISI change. Results Compared to controls, participants randomized to CBTI had improved long-term ISI scores (β= -0.851, SE=0.322, p=0.008). CBTI had no effect on change in sleep midpoint, IV, RA, or IS. However, a significant IV by group interaction suggested that individuals with greater reduction in sleep-wake rhythm fragmentation during CBTI experienced greater reduction in insomnia severity (β= -0.894, SE=0.401, p=0.026). Conclusion Results show that CBTI reduced insomnia severity. While CBTI did not affect IV, RA, IS, or the sleep midpoint, CBTI-induced reductions in insomnia severity were stronger in those who exhibited greater reductions in IV. This suggests that benefits of CBTI may be influenced by changes in RARs in those with TBI. Future analyses will explore factors associated with changes in circadian rhythm and insomnia severity, and examine CBTI effects on circadian rhythm and sleep phase in a larger sample. Support (if any) AASM Foundation Strategic Research Grant Category I.

This study examines the impact of cognitive behavioral therapy for insomnia (CBT-I) and positive airway pressure (PAP) therapy for comorbid insomnia and sleep apnea on nocturnal sleep and daytime functioning. A partial factorial design was used to examine treatment pathways with CBT-I and PAP and the relative benefits of each treatment. One hundred eighteen individuals with comorbid insomnia and sleep apnea were randomized to receive CBT-I followed by PAP, self-monitoring followed by CBT-I concurrent with PAP, or self-monitoring followed by PAP only. Participants were assessed at baseline, PAP titration, and 30 and 90 days after PAP initiation. Outcome measures included sleep diary- and actigraphy-measured sleep, Flinders Fatigue Scale, Epworth Sleepiness Scale, Functional Outcome of Sleep Questionnaire, and cognitive emotional measures. A main effect of time was found on diary-measured sleep parameters (decreased sleep onset latency and wake after sleep onset; increased total sleep time and sleep efficiency) and actigraphy-measured sleep parameters (decreased wake after sleep onset; increased sleep efficiency) and daytime functioning (reduced Epworth Sleepiness Scale, Flinders Fatigue Scale; increased Functional Outcome of Sleep Questionnaire) across all arms (all P < .05). Significant interactions and planned contrast comparisons revealed that CBT-I was superior to PAP and self-monitoring on reducing diary-measured sleep onset latency and wake after sleep onset and increasing sleep efficiency, as well as improving Functional Outcome of Sleep Questionnaire and Flinders Fatigue Scale compared to self-monitoring. Improvements in sleep and daytime functioning were found with PAP alone or concomitant with CBT-I. However, more rapid effects were observed on self-reported sleep and daytime performance when receiving CBT-I regardless of when it was initiated. Therefore, concomitant treatment appears to be a favorable approach to accelerate treatment outcomes. Registry: ClinicalTrials.gov; Name: Multidisciplinary Approach to the Treatment of Insomnia and Comorbid Sleep Apnea (MATRICS); URL: https://clinicaltrials.gov/ct2/show/NCT01785303; Identifier: NCT01785303. Tu AY, Crawford MR, Dawson SC, etal. A randomized controlled trial of cognitive behavioral therapy for insomnia and PAP for obstructive sleep apnea and comorbid insomnia: effects on nocturnal sleep and daytime performance. J Clin Sleep Med. 2022;18(3):789-800.

Identify associations between self-reported history of military and nonmilitary traumatic brain injury (TBI) on hearing loss and hearing difficulty from the Noise Outcomes in Servicemembers Epidemiology (NOISE) study. Cross-sectional. Multi-institutional tertiary referral centers. Four hundred seventy-three Active-Duty Service members (ADSM) and 502 veterans. Self-reported history of no TBI, military TBI only, nonmilitary TBI only, both military and nonmilitary TBI. Pure-tone hearing thresholds, Speech Recognition In Noise Test (SPRINT), Hearing Handicap Inventory for Adults (HHIA), and Speech, Spatial and Qualities of Hearing Scale (SSQ)-12. 25% (120/473) of ADSM and 41% (204/502) of veterans self-reported a TBI. Military TBI was associated with poorer hearing thresholds in all frequency ranges in veterans (adjusted mean difference, 1.8 dB; 95% confidence interval [CI], 0.5-3.0; 3.3, 0.8-5.8; 5.1; 1.7-8.5, respectively), and in the high frequency range in ADSM (mean difference, 3.2 dB; 95% CI, 0.1-6.3). Veterans with military TBI only and nonmilitary TBI only had lower odds of correctly identifying speech in noise than veterans with no TBI (odds ratio [OR], 0.78; 95% CI, 0.72-0.83; 0.90; 0.84-0.98). ADSM with a military TBI (OR, 5.7; 95% CI, 2.6-12.5) and veterans with any TBI history (OR, 2.5; 95% CI, 1.5-4.3; OR, 2.2; 95% CI, 1.3-3.8; OR, 4.5; 95% CI, 2.1-9.8) were more likely to report hearing difficulty on HHIA. SSQ-12 results corroborated HHIA findings. Military TBI was associated with poorer hearing thresholds in veterans and ADSM, and poorer SPRINT scores in veterans. Military TBI was associated with poorer self-perceived hearing ability in ADSM. All types of TBI were associated with poorer self-perceived hearing ability in veterans, although the strength of this association was greatest for military TBI.

Defense and Veterans Brain Injury Center: The First 25 Years

BackgroundPrevious studies have shown the negative impact of sleep disturbances, specifically insomnia symptoms, on glucose metabolism for people with type 2 diabetes (T2D). People with insomnia symptoms are at risk of poor glycemic control and suboptimal diabetes self-care behavior (DSCB). Investigating the impact of a safe and effective intervention for individuals with T2D and insomnia symptoms on diabetes’ health outcomes is needed. Therefore, the aim of this exploratory study is to examine the effects of Cognitive Behavioral Therapy for Insomnia (CBT-I) on glycemic control, DSCB, and fatigue.MethodsTwenty-eight participants with T2D and insomnia symptoms, after passing an eligibility criteria at a medical research center, were randomly assigned to CBT-I (n = 14) or Health Education (HE; n = 14). The CBT-I and HE groups received 6 weekly one-hour sessions. This Randomized Controlled Trial (RCT) used a non-inferiority framework to test the effectiveness of CBT-I. Validated assessments were administered at baseline and post-intervention to assess glycemic control, DSCB, and fatigue. A Wilcoxon signed-rank test was utilized to compare within-group changes from baseline to post-intervention. A Mann-Whitney test was utilized to measure the between-group differences. Linear regression was used to assess the association between the blood glucose level and the number of days in the CBT-I group.ResultsThe recruitment duration was from October 2018 to May 2019. A total of 13 participants completed the interventions in each group and are included in the final analysis. No adverse events, because of being a part of this RCT, were reported. CBT-I participants showed significantly greater improvement in glycemic control, DSCB, and fatigue. There was a significant association between the number of days in the CBT-I intervention with the blood glucose level before bedtime (B = -0.56, p = .009) and after awakening in the morning (B = -0.57, p = .007).ConclusionsThis study demonstrated a clinically meaningful effect of CBT-I on glycemic control in people with T2D and insomnia symptoms. Also, CBT-I positively impacted daytime functioning, including DSCB and fatigue. Future research is needed to investigate the long-term effects of CBT-I on laboratory tests of glycemic control and to understand the underlying mechanisms of any improvements.Trial registrationClinical Trials Registry (NCT03713996). Retrospectively registered on 22 October 2018,

Objectives Chronic insomnia is common in late adulthood. A non-pharmacological approach should take priority in the treatment of insomnia for the elderly. Many studies have shown the efficacy of Cognitive Behavioral Therapy for Insomnia (CBT-I) for elderly diagnosed with insomnia. However the effect of CBT-I on mild insomnia among older adults in community settings has not been ascertained. We conducted a randomized controlled trial to evaluate the effectiveness of a brief CBT-I delivered by nurses, which is feasible in community settings, to improve sleep quality and decrease the dose of hypnotics use for older adults.Methods Participants aged 60 years and over were enrolled in this study. The participants in the intervention group were administrated the brief CBT-I consisting of a group session (60 min) and an individual session (30 min). The primary outcomes were the score differences in the Pittsburgh Sleep Quality Index (PSQI) and Insomnia Severity Index (ISI). The secondary outcomes were the change in the proportion of people diagnosed with insomnia and the dose of hypnotics used. The follow-up period was 3 months.Results The score on PSQI in the intervention group (n=41) significantly decreased compared to the control group (n=38). The effect size (Cohen's d) was 0.56 (95% Confidence interval [CI], 0.07 to 1.05). The score on ISI also decreased significantly and Cohen's d was 0.77 (95%CI, 0.27 to 1.26). According to subgroup analysis, Number Needed to Treat (NNT) for improvement of insomnia was 2.8 (95%CI, 1.5-17.2) and NNT for decreasing of dose of hypnotics use was 2.8 (95%CI, 1.5-45.1).Conclusion The present results have demonstrated that the brief CBT-I significantly improved subjective evaluation of sleep quality and insomnia symptoms among the elderly. In addition, the brief CBT-I decreased the usage of hypnotics. Further studies are needed in terms of the procedure and the effects of brief CBT-I for older adults living in a community.

Cognitive behavioral therapy for insomnia associated with traumatic brain injury: A single-case study

Introduction Recent meta-analyses show that 30-70% of individuals with a history of traumatic brain injury (TBI) suffer from persistent sleep disturbances. Further, more than 80% of all TBI cases are mild (mTBI) and report more insomnia than those with moderate-to-severe TBI. Given the positive track record of Cognitive Behavioral Therapy for Insomnia (CBT-I) in other conditions, CBT-I could also be an effective treatment for insomnia in those with mTBI, despite common comorbidities. Methods In this RCT (NCT03261674) veterans (n=110, 29.1 % female, 51.2±11.5 years old) were randomized to either CBT-I or an active control involving pseudo desensitization. During the treatment, participants had access to a therapist for six 60-minute sessions, once per week, for six consecutive weeks. Treatment was remotely delivered nationwide via telehealth. Independently living veterans with chronic (&amp;gt;3 months since injury) mTBI and insomnia were included in the trial. Comorbid post-traumatic stress disorder (PTSD) was permitted. Central nervous system-active medications were allowed as long as the dose, timing, and formulation were stable (≥ 3 weeks pre-treatment). All sleep measures were based on self-report. The primary outcome measure was the Insomnia Severity Index (ISI) and treatment effects were tested using intent-to-treat analyses based on linear mixed effects models with data derived from multiple imputations. Results CBT-I was associated with significantly greater improvement in ISI (p &amp;lt;.001, Beta=0.59) compared to controls. At post treatment, CBT-I had &amp;gt; 8-point reduction of ISI from 16.4±3.6 to 7.4±5.3; controls only improved moderately from 15.6±4.6 to 10.2±6.2. Pre-post improvements in CBT-I were accompanied with improved (p’s &amp;lt;.001) sleep efficiency (+12.1±12.0 %), mental health (SF-36 mental component score, -4.8±12.3), depression (PHQ-8, -4.2±5.2), and reduced time in bed (-1.0±1.3 h), wake after sleep onset (-32.6±33.6 min), sleep onset latency (-30.4±42.1 min), early morning awakenings (-28.2±42.2 min) while total sleep time remained virtually unchanged (+0.1±1.3 h). Conclusion Despite high rates of co-morbid PTSD and cognitive difficulties associated with mTBI, CBT-I remains a robust treatment for insomnia in this group. Future efforts should focus on increasing accessibility of CBT-I for individuals who have recently sustained a mTBI. Support (if any) VA REHABILITATION RESEARCH &amp; DEVELOPMENT

12009 Background: Women on chemotherapy for breast cancer (BC) report high levels of insomnia and fatigue. This trial aimed to test the main effects of Cognitive Behavioral Therapy for Insomnia (CBT-I) and Bright Light Therapy (BLT) on insomnia and fatigue symptoms. Methods: This multi-center, randomized, controlled, 2 x 2 factorial, superiority, trial enrolled 219 women receiving cytotoxic chemotherapy for any stage BC. Interventions were: (1) neither CBT-I nor BLT (sleep hygiene education; SHE), (2) BLT, (3) CBT-I, and (4) BLT+CBT-I. The 6-week interventions included one telehealth, 1:1 session followed by emails and a mid-treatment call. Assessments occurred at baseline, 3 and 6 weeks. Dual primary outcomes were the insomnia severity index (ISI) and PROMIS Fatigue. Intention-to-treat analyses were latent growth models. Effect sizes are standardized mean differences (SMDs). Results: Mean age was 50.7y and 24% had metastatic cancer. At baseline, average ISI was 13.24 (SD = 5.48; sub-threshold insomnia), and fatigue was 59.57 (SD = 7.91; moderate fatigue). 88% (n = 198) completed the telehealth session. 75% (n = 165) reported post-treatment outcomes. ISI and fatigue decreased in all conditions (see Table). CBT-I improved ISI (mean difference = -2.03; p = .001; SMD = -0.37), but BLT did not (mean difference = -1.09; p = .082; SMD = -0.20). Neither intervention affected fatigue (SMDs -0.06 to -0.07; p &gt; 0.60). There was no BLTxCBT-I interaction for ISI nor fatigue ( p &gt; 0.50). Conclusions: In patients receiving chemotherapy for BC, brief CBT-I can improve insomnia but not fatigue symptoms. BLT did not improve insomnia or fatigue. We found no evidence of an interaction between BLT and CBT-I. During chemotherapy, fatigue may not be responsive to brief sleep and circadian-oriented treatments. Clinical trial information: ACTRN12620001133921 . Between group (main effects) and within group (change). ISI [95% CI] P, SMD Fatigue [95% CI] P, SMD Main Effects BLT -1.09 [-2.31, 0.14] p = .082, SMD = -0.20 -0.49 [-2.87, 1.88] p = .68, SMD = -0.06 CBT-I -2.03 [-3.25, -0.81] p = .001, SMD = -0.37 -0.54 [-2.92, 1.83] p = .65, SMD = -0.07 Change: 0–6 weeks SHE -3.41 [-4.65, -2.17] p &lt; .001, SMD = -0.62 -3.75 [-6.16, -1.34] p = .002, SMD = -0.47 BLT -4.89 [-6.12, -3.66] p &lt; .001, SMD = -0.89 -3.75 [-6.13, -1.37] p = .002, SMD = -0.47 CBT-I -5.83 [-7.12, -4.54] p &lt; .001, SMD = -1.06 -3.80 [-6.30, -1.31] p = .003, SMD = -0.48 CBT-I+BLT -6.53 [-7.88, -5.18] p &lt; .001, SMD = -1.19 -4.79 [-7.44, -2.14] p &lt; .001, SMD = -0.61

Chronic insomnia is a frequently occurring problem, and the number of Danes who experience sleep problems is increasing. Cognitive behavioural therapy for insomnia (CBT-i) as argued in this review is recommended as the first choice of treatment either face-to-face or digitally. Pharmacological treatment should only be used if CBT-i is not sufficient. The effect of CBT-i has been investigated in several meta-analyses and systematic reviews, the effect is found to be good and the recommendation for using CBT-i is strong. Digital CBT-i programs have been developed, and here the efficacy is also found to be good.

Blast-related mild traumatic brain injuries (TBIs), the "signature injury" of post-9/11 conflicts, are associated with clinically relevant, long-term cognitive, psychological, and behavioral dysfunction and disability; however, the underlying neural mechanisms remain unclear. To investigate associations between a history of remote blast-related mild TBI and regional brain volume in a sample of US veterans and active duty service members. Prospective cohort study of US veterans and active duty service members from the Long-Term Impact of Military-Relevant Brain Injury Consortium-Chronic Effects of Neurotrauma Consortium (LIMBIC-CENC), which enrolled more than 1500 participants at 5 sites used in this analysis between 2014 and 2023. Participants were recruited from Veterans Affairs medical centers across the US; 774 veterans and active duty service members of the US military met eligibility criteria for this secondary analysis. Assessment dates were from January 6, 2015, to March 31, 2023; processing and analysis dates were from August 1, 2023, to January 15, 2024. All participants had combat exposure, and 82% had 1 or more lifetime mild TBIs with variable injury mechanisms. Regional brain volume was calculated using tensor-based morphometry on 3-dimensional, T1-weighted magnetic resonance imaging scans; history of TBI, including history of blast-related mild TBI, was assessed by structured clinical interview. Cognitive performance and psychiatric symptoms were assessed with a battery of validated instruments. We hypothesized that regional volume would be smaller in the blast-related mild TBI group and that this would be associated with cognitive performance. A total of 774 veterans (670 [87%] male; mean [SD] age, 40.1 [9.8] years; 260 [34%] with blast-related TBI) were included in the sample. Individuals with a history of blast-related mild TBI had smaller brain volumes than individuals without a history of blast-related mild TBI (which includes uninjured individuals and those with non-blast-related mild TBI) in several clusters, with the largest centered bilaterally in the superior corona radiata and subcortical gray and white matter (cluster peak Cohen d range, -0.23 to -0.38; mean [SD] Cohen d, 0.28 [0.03]). Additionally, causal mediation analysis revealed that these volume differences significantly mediated the association between blast-related mild TBI and performance on measures of working memory and processing speed. In this cohort study of 774 veterans and active duty service members, robust volume differences associated with blast-related TBI were identified. Furthermore, these volume differences significantly mediated the association between blast-related mild TBI and cognitive function, indicating that this pattern of brain differences may have implications for daily functioning.

Effectiveness of cognitive behavioral therapy for pharmacotherapy-resistant chronic insomnia: a multi-center randomized controlled trial in Japan

Objective/Background The primary aim of this study was to examine the effect of Cognitive Behavioral Therapy for Insomnia (CBT-I) on the severity of insomnia in people with Type 2 diabetes (T2D) compared to a health education (HE) control group. The secondary aim was to explore the effect of CBT-I on other sleep outcomes and concomitant symptoms. Participants Twenty-eight participants with T2D were randomly assigned to CBT-I (n = 14) or HE (n = 14). Methods Validated assessments were used at baseline and post intervention to assess sleep outcomes and concomitant symptoms. In addition, actigraph and sleep diaries were used to measure sleep parameters. Independent sample t tests and Mann-Whitney U tests were utilized to measure between-group differences in the mean change scores. Results Participants in the CBT-I group showed higher improvements in the following mean change scores compared to the HE group: insomnia symptoms (d = 1.78; p < .001), sleep quality (d = 1.53; p =.001), sleep self-efficacy (d = 1.67; p < .001). Both actigraph and sleep diary showed improvements in sleep latency and sleep efficiency in the CBT-I group as compared to the HE group. In addition, participants in the CBT-I group showed greater improvement in the mean change scores of depression symptoms (d = 1.49; p = .002) and anxiety symptoms (d = 0.88; p = .04) compared to the HE group. Conclusion This study identified a clinically meaningful effect of CBT-I on sleep outcomes and concomitant symptoms in people with T2D and insomnia symptoms. Further work is needed to investigate the long-term effects of CBT-I in people with T2D and insomnia symptoms.

Insomnia and depression are highly co-morbid conditions that share a complex, bi-directional relationship. By reviewing the literature examining the links between insomnia and depression, it becomes apparent that insomnia symptoms need to be directly targeted alongside depression symptoms when co-morbidities exist for optimal outcomes. The most common treatment in Australia for co-morbid insomnia and depression is currently antidepressant medication. Because a large percentage of individuals given antidepressants fail to remit from their depression and insomnia symptoms, an effective adjunct treatment to antidepressants for co-morbid insomnia and depression is required. Cognitive behavioural therapy for insomnia (CBT-I) is an empirically validated treatment for insomnia, and is efficacious in improving insomnia that is co-morbid with depression. Recent evidence also suggests that CBT-I produces significant improvements in depression severity. The main aim of the thesis was therefore to develop a cost-effective CBT-I intervention that could be easily administered, distributed and implemented on a wide-scale basis, both on its own for insomnia, and primarily as an adjunct therapy for co-morbid insomnia and depression. A therapist manual of CBT-I, based on the latest empirical findings, was specifically developed to increase consistency and ease of treatment administration by therapists wanting to deliver CBT-I treatment. A resource booklet was also developed to give to all individuals undergoing CBT-I treatment, to ensure that they were provided with the latest information about sleep and how to reduce their insomnia severity. A randomised controlled trial was then developed to investigate the efficacy of this CBT-I intervention in individuals with co-morbid insomnia and depression whose symptoms have failed to remit through antidepressant treatment. Forty-one participants (aged 18-64 years) were randomized to receive four sessions of either CBT-I or sleep education (self-help therapy) over an 8-week period (one session every two weeks). Participants had been treated with antidepressants for at least six weeks prior to screening, but were otherwise healthy. The Insomnia Severity Index and the Beck Depression Inventory were assessed at baseline, following each session, and at 3-month follow-up. Secondary outcomes were sleep quality and duration (actigraphy and self-report), anxiety, fatigue, and daytime sleepiness. Results indicated that CBT-I, compared to sleep education, produced significantly reduced depression and insomnia severity, and improved anxiety, fatigue, sleep quality and sleep efficiency. It is the first randomised controlled trial of CBT-I for co-morbid insomnia and depression, with an active control treatment, to show significant reductions in both insomnia and depression severity at post-treatment and follow-up. Large effect sizes were found for both insomnia and depression at post-treatment, and these effect sizes increased further by three month follow-up. At follow-up, ten times more CBT-I participants were in remission from both insomnia and depression than sleep education participants. This demonstrates that CBT-I is an efficacious treatment for co-morbid insomnia and depression, and should be considered an important adjunct therapy in individuals whose symptoms have not remitted through antidepressant treatment. Following the positive results of the randomised controlled trial, the proposed mediators of CBT-I were examined to determine which aspects of the intervention had the largest influence on reductions in insomnia and depression severity at post-treatment and 3-month follow-up. Changes in time-in-bed, bedtime variability, rise-time variability, sleep hygiene practices, dysfunctional beliefs about sleep and stress levels were assessed as potential mediators. Post-treatment results indicated that the significantly reduced depression severity through CBT-I was primarily via reduced stress levels, improved sleep hygiene practices, and other therapeutic effects of CBT-I, whereas reduced insomnia severity was primarily via reduced dysfunctional beliefs about sleep. By follow-up, the significantly reduced depression severity in the CBT-I group was primarily via reduced stress levels and reduced dysfunctional beliefs about sleep, whereas reduced dysfunctional beliefs about sleep and other therapeutic effects of CBT-I explained the significantly reduced insomnia severity. These results indicate that reducing stress levels through CBT-I treatment is required for optimal depression outcomes, whereas reducing dysfunctional beliefs about sleep is essential for optimal insomnia outcomes. Improving sleeping practices behaviourally is important for initial depression improvements, but reducing dysfunctional beliefs about sleep becomes more important for longer-term remission of depression symptoms. Other non-factual therapeutic elements of CBT-I treatment appear to be important, and may lead to a greater willingness to engage in the CBT-I strategies, which can improve depression in the short-term and insomnia in the longer-term. These mediators of CBT-I treatment had not been specifically examined in individuals with co-morbid insomnia and depression before this study. Their identification suggests that relaxation and behavioural sleep interventions should be prioritized initially through CBT-I for co-morbid insomnia and depression, followed by cognitive interventions to target unhelpful beliefs about sleep. By increasing the efficiency of CBT-I, optimal treatment outcomes can be achieved for individuals with co-morbid insomnia and depression, and the risk of future relapse can be minimized.