Insomnia and obstructive sleep apnea (OSA) are the two most prevalent sleep disorders and frequently co-occur (AASM, 2014; Sweetman et al., 2017a). Approximately 30–40% of people with insomnia have co-morbid OSA (Luyster et al., 2010; Sweetman et al., 2021), however most people with OSA remain undiagnosed (Lichstein et al., 1999). People with Co-Morbid Insomnia and Sleep Apnea (COMISA) generally have worse sleep (Bianchi et al., 2013), daytime function (Krakow et al., 2001), mental health (Lang et al., 2017), physical health (Lechat et al., 2022a), productivity (Sivertsen et al., 2013), and quality of life (Björnsdóttir et al., 2015) compared to people with neither sleep disorder, and often compared to people with either insomnia-alone or OSA-alone (Sweetman et al., 2019a; Ong et al., 2020a). Three recent population-based studies have reported that people with COMISA experience a 50–70% increased risk of mortality over 10–20 years of follow-up, compared to people with neither condition (Lechat et al., 2021, 2022b; Sweetman et al., 2022). Given the high prevalence and adverse health consequences of COMISA, it is important to develop and implement effective evidence-based management approaches for this condition (Sweetman et al., 2017a).

Cognitive Behavioral Therapy for insomnia (CBTi) is the recommended “first line” treatment for insomnia (Qaseem et al., 2016; Ree et al., 2017; ASA, 2024), and is effective in patients with COMISA (Sweetman et al., 2023a). CBTi is a multi-component therapy that has historically been delivered by trained therapists/psychologists over 4–8 weekly sessions. CBTi aims to identify and treat the underlying precipitating triggers and perpetuating factors that maintain insomnia, and leads to sustained improvements in sleep (Morin et al., 1994; van der Zweerde et al., 2019a), daytime function and mental health (Ye et al., 2015). A recent systematic review and meta-analysis reported that CBTi is associated with large improvements in insomnia severity in the presence of treated and untreated co-morbid OSA (Sweetman et al., 2023a). Furthermore, CBTi may improve the severity of un-treated OSA (Sweetman et al., 2020a), and two of four recent randomized controlled trials indicate that CBTi may improve subsequent acceptance and use of CPAP therapy in patients with COMISA (Bjorvatn et al., 2018; Sweetman et al., 2019b; Ong et al., 2020b; Alessi et al., 2021).

Despite a wealth of evidence supporting the effectiveness of CBTi, access to this “first line” treatment is extremely limited. In Australia, ~90% of primary care patients with insomnia are managed with sleeping pills while only 1% are referred to psychologists for CBTi (Miller et al., 2017; Haycock et al., 2022). In a recent analysis of the Veterans Health Administration, the largest healthcare system in the US, veterans initially presenting to various clinics with insomnia disorder were 11-times more likely to receive a medication prescription first than to be referred to CBTi (Pfeiffer et al., 2023). In this analysis including over five million veterans, CBTi was provided to only 0.2% of the sample. Rates of access to CBTi across Europe mirror those elsewhere in the world with only 50–3,000 patients per country receiving psychological treatment for their insomnia per year (Baglioni et al., 2020).

Clinician-delivered CBTi programs have been translated to self-guided digital programs which have the potential to increase CBTi availability and access (Thorndike et al., 2008; Espie et al., 2012). However, self-guided digital CBTi programs have generally been developed for people with insomnia-alone, and there is very limited evidence on the effectiveness, safety, or suitability of digital CBTi in people with COMISA (Eldridge-Smith et al., 2022). In particular, people with COMISA may commence treatment with higher levels of daytime sleepiness, be less responsive to self-guided sleep restriction therapy in the presence of apnea events that promote repeated awakenings, or experience increased sleepiness during the acute phase of sleep restriction (Sweetman et al., 2020b; Turner et al., 2023). Because of these concerns, some self-guided digital programs tailored for people with insomnia-alone may not be appropriate for individuals with OSA. Indeed, clinical trials of digital CBTi often exclude people with untreated OSA (Christensen et al., 2016; Ritterband et al., 2017; Nazem et al., 2023) and suspected OSA (Espie et al., 2019; van der Zweerde et al., 2019b; Vedaa et al., 2020). This may fuel the belief that insomnia is “secondary” to OSA, or that OSA must be diagnosed and sufficiently treated before commencing insomnia management (Sweetman et al., 2017a, 2021). However, our systematic review demonstrates that CBTi can be safely and effectively delivered by clinicians to patients with untreated OSA (Sweetman et al., 2023a). These results are particularly welcoming when considering the long delay to diagnosis and treatment of OSA in many countries.

Overnight polysomnography is the “gold standard” measure of OSA presence and severity, however, is costly, time consuming, and often incurs long waiting lists. In the absence of overnight sleep studies, self-report measures have been developed and validated to screen for a “high-risk” of OSA (Chai-Coetzer et al., 2011; Chung et al., 2016). To expedite access to digital CBTi in patients with COMISA, it may be possible to identify people with insomnia and a “high risk” of co-morbid OSA according to self-report insomnia and OSA-screening questionnaires. CBTi could then be administered concurrent to the diagnosis and treatment of possible OSA. However, there are safety concerns when providing sleep restriction therapy to people with suspected COMISA, as this treatment may increase an already higher level of daytime sleepiness (Sweetman et al., 2020b). A digital CBTi program that can tailor therapy to mitigate excessive sleepiness during CBTi would thus be invaluable to patients with suspected COMISA, who otherwise experience very limited access to CBTi partly due to sleepiness-related and safety concerns.

The aim of this study was therefore to develop and test a self-guided digital CBTi program that includes treatment algorithms designed for people with insomnia-alone and COMISA (Sweetman et al., 2023c). We aimed to compare the effectiveness of an interactive digital CBTi program, Bedtime Window, on changes in symptoms of insomnia, daytime function and mental health in people with insomnia-alone vs. people with comorbid insomnia and a “high-risk” of OSA (suspected COMISA group), and monitor weekly changes in sleepiness during treatment.

We report here secondary data analysis from an online clinical trial of digital CBTi in a community-based sample of people with insomnia symptoms (Sweetman et al., 2023c). The trial was approved by the Southern Adelaide Clinical Human Research Ethics Committee and registered on the Australian and New Zealand Clinical Trials Registry (ACTRN12621001395820). Secondary data analysis was conducted to investigate the effect of insomnia-alone vs. insomnia and high-risk OSA on treatment-response to digital CBTi from baseline to 8- and 16-week follow-up, and between-group changes in weekly self-reported sleepiness during CBTi.

The main finding of this study is that this interactive digital CBTi program was associated with large and sustained improvements in nocturnal and daytime symptoms of insomnia and associated mental health symptoms in participants with insomnia-alone, and in those with a high-risk of undiagnosed OSA. The lack of any significant group by time interactions provides preliminary evidence to suggest that those with high-risk of OSA benefited from this digital CBTi program at least as much as those with insomnia-alone. An important secondary finding is that symptoms of daytime sleepiness did not increase significantly during any week of the digital CBTi program in either group, suggesting that the program may be an effective, safe, and scalable treatment for insomnia in people with suspected COMISA. Importantly, further research in patients with a doctor diagnosis of OSA according to gold-standard polysomnography is required to confirm these promising initial results, before scaling access in different settings in which patients with COMISA are managed.

A non-significant trend was observed that indicated greater overall improvements in symptoms of insomnia, sleepiness and depression in the suspected COMISA group compared to the insomnia-alone group. Caution should be applied in interpreting these non-significant interaction effects, which were of limited clinical significance, and may have also been inflated by the multiple repeated-measures outcomes that were analyzed. The between-group difference in rates of insomnia remission at 16-weeks should similarly be interpreted with caution, given that these “responder” outcomes were based on complete-case analyses, and a small overall sample was recruited. Of more importance, we observed no evidence that participants with a high risk of co-morbid OSA experienced a reduced treatment-response to this digital CBTi program compared to those with insomnia-alone. Indeed, the most striking findings were the main effects of time that indicated moderate-to-large and sustained improvements in all outcomes for both groups combined.

In a recent systematic review and meta-analysis of the effect of CBTi in people with COMISA (Sweetman et al., 2023a), we observed a large effect of CBTi on improving ISI scores in people with insomnia and co-morbid treated and un-treated OSA. In the present study, those in the suspected COMISA group also experienced a large and sustained 11.2-point average ISI improvement by 16-weeks. These data suggest that CBTi is an efficacious treatment for insomnia in the presence of high-risk co-morbid OSA, when delivered by trained clinicians, or via an interactive self-guided digital program tailored for COMISA. Before scaling access to self-guided digital CBTi in people with COMISA, these results require replication in larger samples of patients with baseline sleep studies to confirm the presence and severity of co-morbid sleep apnea. One additional pilot study is currently being conducted in people with co-morbid insomnia and a doctor-diagnosis of OSA (Sweetman et al., 2022), and one study is being conducted in people with insomnia-alone and COMISA in primary care (Sweetman et al., 2022). Additional research is also required to determine the effectiveness and acceptability of digital CBTi in patients with confirmed COMISA recruited from different settings, including sleep clinics in which patients may present with more severe OSA, symptomatic OSA (i.e., with higher levels of sleepiness at baseline), and different factors that motivate treatment-seeking and engagement (Sweetman et al., 2019b; Alessi et al., 2021).

Although self-guided digital CBTi programs have been studied in samples with insomnia (Soh et al., 2020), and co-morbid insomnia and mental health symptoms (Ye et al., 2015), there has been understandable reservation about using self-guided digital CBTi in patients with COMISA (Eldridge-Smith et al., 2022). Sleep restriction therapy is one of the most effective behavioral components of CBTi, which aims to temporarily reduce time in bed, consolidate sleep periods and reduce time awake throughout the night (Spielman and Saskin, 1987). Similarly, stimulus control therapy which aims to re-associate the bed with a state of sleep and rest rather than conditioned arousal, requires the patient to “get out of bed if not asleep within about 15 minutes”, and consequently may delay sleep onset and reduce sleep duration in the early treatment phase (Bootzin, 1972). Although these behavioral treatments often lead to rapid improvement in insomnia, they are associated with an acute increase in daytime sleepiness in patients with insomnia alone (Kyle et al., 2014) and COMISA (Sweetman et al., 2020b). This increase in sleepiness is reduced after sleep improves, and time in bed is gradually extended from week to week until a comfortable equilibrium between sleep duration, time in bed, and sleepiness is achieved (Sweetman et al., 2020b). However, patients with COMISA may commence treatment with pre-existing levels of sleepiness due to un-treated OSA (Sweetman et al., 2017b), and experience an increased vulnerability to the effects of sleep restriction on daytime sleepiness and neurocognitive function (Vakulin et al., 2009; Turner et al., 2023). Hence, there has been reservation about the use of digital CBTi programs designed for patients with insomnia-alone, in patients with COMISA. In the current study, participants with insomnia-alone and suspected COMISA did not report any overall increase in daytime sleepiness during any week of the digital CBTi program. This is likely due to specific algorithms within the program that continuously monitor for levels of alertness and sleepiness, and adapt personalized therapy recommendations to mitigate risk of alertness-failure. In fact, an overall effect of time on reduced levels of daytime sleepiness was observed by 16-week follow-up. Given reports that CBTi is associated with acute impairment in neurocognitive function (Turner et al., 2023), future research is required to confirm the safety of this digital CBTi program in patients with different levels of OSA severity, and in the presence of excessive daytime sleepiness, before implementing broader access.

The recommended “first line” treatment for moderate and severe OSA is Continuous Positive Airway Pressure (CPAP) therapy, and lifestyle/weight management recommendations where indicated (Epstein et al., 2009; ASA, 2024). Although CPAP therapy improves stability of the upper airway during sleep and many of the associated consequences of OSA, it is limited by suboptimal acceptance and long-term use for the duration of the sleep period (Weaver and Grunstein, 2008). Co-morbid insomnia symptoms are associated with approximately a 30% reduction in initial acceptance of CPAP therapy, and a 2-h reduction in average nightly CPAP use compared to patients with OSA alone (Sweetman et al., 2023b). Two of four recent randomized controlled trials have reported that CBTi improves subsequent acceptance and use of CPAP therapy in patients with COMISA (Bjorvatn et al., 2018; Sweetman et al., 2019b; Ong et al., 2020b; Alessi et al., 2021). An ongoing study aims to investigate the effect of digital CBTi on insomnia symptoms and CPAP use among people with COMISA after commencement of CPAP recommendations and management (Eldridge-Smith et al., 2022). Future studies should also investigate the possibility of using digital CBTi to treat insomnia prior commencing CPAP, to improve initial experiences with CPAP therapy (i.e., the first week of use), and improve long-term CPAP use in patients with confirmed COMISA.

This study found that Bedtime Window, a self-guided interactive digital CBTi program tailored for insomnia-alone and COMISA, was associated with large and sustained improvements in sleep, depression, anxiety, sleepiness, fatigue, and maladaptive beliefs about sleep in people with insomnia-alone and those with suspected COMISA. Given the high prevalence and adverse health consequences of COMISA, and extremely limited access to CBTi, these preliminary results highlight the potential to investigate the effectiveness, safety and acceptability of this digital CBTi program in people with a confirmed diagnosis of OSA in different settings, before scaling access to people with COMISA throughout the health system.

The datasets presented in this article are not readily available because of Human Research Ethics Committee requirements. Requests to access the datasets should be directed to AS, alexander.sweetman@flinders.edu.au.

The studies involving humans were approved by the Southern Adelaide Clinical Human Research Ethics Committee (2021/HRE00287). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

AS: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Visualization, Writing – original draft, Writing – review & editing. CRe: Investigation, Methodology, Project administration, Resources, Writing – review & editing. LL: Writing – review & editing. AV: Writing – review & editing. CC-C: Writing – review & editing. DW: Writing – review & editing. MC: Writing – review & editing. CRi: Data curation, Investigation, Methodology, Project administration, Resources, Writing – review & editing.