Discussion

Cognitive behavioral therapy for insomnia (CBT-I) is an evidence-based insomnia treatment that leads to sustained benefit by addressing perpetuating behaviors and beliefs (51). Behavioral strategies include consolidating sleep and strengthening the association between the bedroom and sleep. Cognitive strategies involve sleep education so as to demystify any myths or beliefs that might perpetuate poor sleep. Relaxation and mindfulness techniques are also implemented to decrease cognitive and physiological hyperarousal that is frequently part of the insomnia symptom profile. Advantages to using CBT-I compared to medications include no known side effects, lower cost, greater efficiency, and durability (42). The American Academy of Sleep Medicine and the American College of Physicians recommend CBT-I as a first-line treatment for insomnia over medication, even in the context of psychological and medical comorbidity (44; 11).

Despite such clinical indications and professional endorsement, the uptake and widespread dissemination and implementation of CBT-I in medical settings is limited, and the gap between this gold standard treatment and clinical practice persists. Patient and physician awareness, lack of qualified CBT-I clinicians, and poor geographical distribution of providers trained in CBT-I continue to be major impediments to widespread usage (54). In addition, integration of mental health services in medical settings face operational and economic barriers that currently limit widespread application of this evidence-based approach. These barriers to access are accentuated for minorities and the underserved who often suffer more severe insomnia consequences. The implications of this lack of access are an individual and public health concern. The cost of untreated insomnia includes, but is not limited to, increased absenteeism, morbidity, disability, and healthcare expenses (07; 41). These barriers to treatment are well documented and have motivated a growing body of research examining alternative adaptations for disseminating CBT-I more widely. Innovations such as briefer behavioral treatments, group therapy, stepped care approaches, expanding training opportunities, and non-face-to-face options, including phone, bibliotherapy, and self-help options, have demonstrated effectiveness in improving sleep and are not reliant on traditional (and limited) in-person treatment (23; 38). Telehealth options also address some of these barriers, and the U.S. Department of Health and Human Services stresses the importance of “health internet technology” as a means of improving access to healthcare. Sleep telemedicine offers another opportunity for increasing access to this gold standard treatment.

Sleep telemedicine

Sleep telemedicine is a growing field of research and clinical interest and has been the subject of several reviews exploring its effectiveness and viability as an alternative option for treating insomnia (48; 52; 49; 50). Defined in a 2015 AASM position paper (updated in 2020 to reflect lessons learned from the COVID-19 pandemic) as “use of sleep-related medical information exchanged from one site to another via electronic communications to improve a patient’s health,” sleep telemedicine incorporates different modalities, including telephone, video, internet, or smartphone applications.

Table 1. Types of TeleHealth and Telemedicine

Method	Synchronous versus Asynchronous	Description
Telephone	Synchronous	Audio only interaction
Live video conferencing	Synchronous	Video and audio real-time interaction
Asynchronous video	Asynchronous	Video and audio stored and then sent for later reply
Software apps, professional-guided	Asynchronous	Patient enters data and obtains advice and guidance from professional
Software apps, self-guided	Asynchronous	Patient enters data and obtains advice and self-guidance from the app algorithm

The COVID-19 pandemic dramatically accelerated the adoption and use of telemedicine in an effort to protect patients, providers, and staff. Quickly, video and telephone-based encounters were implemented, and the research from this initial period shows high levels of acceptability from both patients and providers and effectiveness of treatment even when patients switched from in-person to teletherapy, mid-treatment (39). Although this accelerated use occurred in the context of a worldwide pandemic, the benefits of sleep telemedicine have been broadly recognized and will likely remain an alternative option for sleep medicine care (31; 43). As part of a larger randomized non-inferiority trial, patients and providers shared their experiences about telemedicine CBT-I and noted feeling satisfied overall with CBT-I treatment in this modality compared to in-person treatment and appreciated the benefits to access (21). Although sleep telemedicine evolves in the clinical and research setting, focus has also expanded to stress the importance and necessity of adjusting teaching and training to reflect this new standard of care. In a survey of sleep medicine program directors, it was recognized that there was a need to develop a standard telemedicine curriculum to enhance fellows’ training (16), and sleep medicine fellows from the University of Minnesota proposed a model of virtual sleep education to address this gap in training (47).

The AASM makes the distinction between two categories of sleep telemedicine: asynchronous and synchronous. Asynchronous delivery of CBT-I (digital CBT-I, dCBT-I or internet CBT-I, iCBT-I) falls into three distinct modalities of delivery: supportive dCBT-I, therapist-guided dCBT-I, and fully automated dCBT-I (34; 35; 57).

Table 2. Digital CBT-I Programs

dCBT-I programs	Description
CBT-I Coach	Department of Veterans Affairs app
Sleepio	Covered by insurance or employer
Go! To Sleep	Asynchronous 6-week program designed by Cleveland Clinic sleep specialists
Sleep Reset	CBT-I with asynchronous coaching
Stellar Sleep	CBT-I app developed by sleep specialists from Brigham and Women's Hospital and Harvard Medical School

Synchronous telemedicine approaches make use of telehealth for the delivery of sleep medicine via “audio and video two-way, real time, interactive communication.” In this article, we review evidence for and applications of both asynchronous and synchronous telemedicine adaptations of CBT-I. Similar to the stepped-care approaches recommended for in-person delivery of CBT-I (13; 36), delivery of telehealth-delivered CBT-I can range from fully automated to various degrees of therapist assistance (25; 57). Meta-analyses discuss benefits of multiple modalities of both asynchronous and synchronous telehealth-delivered CBT-I as a viable alternative to face-to-face treatment of insomnia (52; 49). Not only do these reviews highlight the effectiveness of this alternative delivery of CBT-I in treating insomnia, but they also stress the promising advances in expanding accessibility and availability of CBT-I worldwide and the associated important public health implications.

Asynchronous. Asynchronous telehealth approaches to CBT-I include communication between patient and provider that does not occur in real-time, if at all. As with synchronous-guided internet-delivered CBT-I, there is a growing body of research supporting the benefits of asynchronous approaches as an alternative to in-person CBT-I treatment that addresses well-documented concerns of access. Using a computer and other electronic devices (cellphone, tablet), digital platforms allow patients to access interactive websites or apps with engaging and informative interfaces that are fully automated and scalable (42). Popular examples of digital CBT-I apps that are empirically supported include Sleepio, SHUTi (Somryst), and CBT-I Coach (12).

In the literature, digital CBT-I is described in various ways, including digital, internet-based, and online. For the purposes of this article, and consistent with recent reviews, the terms “digital” or “dCBT-I” are used (35).

In 2004, Strom and colleagues published the first data of a randomized control trial demonstrating the benefits of digital CBT-I in a self-help format. To date, multiple studies have added to this work and demonstrate that benefits resulting from CBT-I administered via sleep telemedicine modalities are comparable to those delivered during in-person office visits in both adult and pediatric populations (48; 52). These studies have demonstrated improvements in insomnia, sleep quality, pre-sleep cognitive arousal, and dysfunctional beliefs about sleep (58).

Hoping to increase access to insomnia treatment in the primary care setting, Bernstein and colleagues conducted a randomized control trial examining a 6-week, digital, interactive, CBT-based program (Go! To Sleep) designed to foster good sleep practices, sleep restriction, and stimulus control therapies (02). Compared to their control, those who participated in the dCBT-I program demonstrated clinically meaningful improvements in insomnia severity as well as perceived stress that were sustained at 4-month follow-up. Similarly strong support has been demonstrated for SHUT-I (Sleep Healthy Using the Internet), a dCBT-I program. Cancer survivors who completed SHUT-I reported both statistically significant and clinically meaningful improvements in insomnia severity and sleep efficiency compared to a waitlist control (45). Based on this research demonstrating the robust effects of fully automated digital CBT-I on nighttime symptoms, Espie and colleagues expanded their focus to study the effects on daytime functioning and found that engagement in dCBT-I (Sleepio program) was associated with improvement in global health, mental well-being, and sleep-related impairment to quality of life when compared to participation in sleep-hygiene education and that these effects were mediated by insomnia improvements associated with dCBT-I (14). dCBT-I has also been shown to be effective during pregnancy. Kalmbach and colleagues randomized participants to Sleepio or a digital sleep education control (28). (The one modification for the Sleepio arm was that participants were not restricted to less than 6 hours of sleep.) In this study, evidence supported that dCBT-I improves sleep quality and sleep duration during pregnancy and after childbirth. Treatment effects related to depression and cognitive arousal were not observed.

Even for participants experiencing subthreshold insomnia, a randomized control trial studying female participants demonstrated that fully automated dCBT-I (6 weekly CBT-I sessions delivered by an animated “virtual therapist” via the online platform Sleepio) was associated with greater improvements in insomnia compared to a control group as well as improvements in anxiety, paranoia, and perceived stress (08). Fully automated dCBT-I has also been found to be effective for patients with chronic health conditions, including chronic migraines and breast cancer. A 2020 proof of concept study provided evidence that dCBT-I (again, Sleepio) is feasible and acceptable in patients with chronic migraines and insomnia. The treatment was also associated with clinically meaningful responses indicated by reduction in insomnia severity to nonclinical levels and improvement in migraines (05). Similarly, breast cancer survivors who participated in fully automated dCBT-I demonstrated reductions in insomnia severity (that were also clinically significant) and overall improvement in sleep quality (59). It should be noted that in addition to the studies detailed here, this research area continues to grow and expand as indicated by the reporting of new study protocols examining the effectiveness of dCBT-I in different populations, including those with alcohol use disorder (04) and depression (10).

Only a few studies have examined unguided dCBT-I compared to in-person CBT-I. Taylor and colleagues randomized active duty military personnel participants to one of three treatment arms (unguided dCBT-I, in-person CBT-I, and waitlist control) and found that both in-person and unguided dCBT-I were effective interventions and associated with improvements in sleep efficiency, sleep latency, insomnia severity, dysfunctional sleep beliefs, and number of awakenings compared to the control group (53). However, the authors also noted that in-person effect sizes were consistently better than the unguided dCBT-I group, suggesting the importance of triaging care depending on patient needs. Kallestad and colleagues conducted a non-inferiority trial examining fully automated dCBT-I with individual face-to-face CBT-I (26). Although both groups showed significant and clinically meaningful reductions in insomnia severity, participants in the dCBT-I group did not fare as well in reducing sleep medication and long-term dysfunctional beliefs. Because of this, the authors note that with respect to inferiority or non-inferiority, the result is inconclusive, and that future research should aim to focus on the selection of people who would most likely benefit from dCBT-I.

Concern has also been voiced about racial disparities in the real-world uptake and engagement with dCBT-I. Kalmbach and colleagues found racial disparities in treatment engagement and patient-reported outcomes among pregnant women treated with dCBT-I (29). In a round-table discussion on telemedicine and digital delivery of CBT-I, insomnia experts discussed the under-representation of underserved and marginalized populations in the development and testing of this treatment approach as well as the barrier of cost given that digital therapies are often not covered by insurance (27; 37). They agreed that future research should seek to implement a lens of equity when developing and testing these automated dCBT-I platforms. This will be especially critical with the integration of prescription digital therapeutics for insomnia treatment. Somryst is the first (and only, at this time) FDA-approved prescription digital therapeutic for the treatment of insomnia. Originally, SHUT-I, Somryst also has a robust body of research demonstrating that it is an effective treatment modality of dCBT-I for use in asynchronous care. This work is encouraging, and further research may indicate which patients are good candidates for these therapies.

Consistent with the stepped care model of insomnia treatment, asynchronous treatments can also incorporate a therapist/coach who provides feedback, explanations, and encouragement. Among healthy patients, a growing body of literature supports the use of this blended therapy experience through guided CBT-I. dCBT-I offers several advantages, including efficiency of therapist time as well as increased access for patients with daytime responsibilities (work, caretaking) or those with geographical or logistical barriers to coming to clinic in person. van Straten and colleagues demonstrated that guided dCBT-I is feasible, acceptable, and associated with improvements in sleep quality and sleep efficiency, as well as depression, anxiety, and quality of life, when compared to a waitlist control group (56). This same research group compared dCBT-I with and without support and found that motivational support provided via email was associated with sustained benefits for sleep latency, insomnia symptoms, and depressive symptoms (32). Blom and colleagues randomized participants to face-to-face group CBT-I therapy or therapist-guided dCBT-I (03). The therapist-guided digital therapy included active support via written feedback after the participant completed each module. The researchers found that both treatment groups demonstrated significant improvements in insomnia severity (measured by the ISI) from pre- to post-assessments and that these improvements were maintained over a 6-month period. Participants in both treatment groups also showed improvements in secondary outcome measures: sleep latency, efficiency, and sleep quality as well as depression. Moreover, subsequent research has demonstrated that therapist-guided dCBT-I is feasible, well-accepted, and effective for patients with breast cancer (09). As asynchronous treatments can range from fully automated to patient directed, dCBT-I appears to be a viable treatment option that is wide-ranging and acceptable for patients who might not otherwise have access to this treatment. Triaging specific needs and determining which patients will benefit from more or less assistance and guidance is an important area of future research.

Synchronous. Synchronous telemedicine approaches to CBT-I require the patient and therapist to be interacting at the same time. This method offers the benefit of communicating directly with a therapist without the burden of having to travel to a clinic. Similar to the state of the science on asynchronous approaches, the research literature focusing on synchronous telehealth approaches to CBT-I is rapidly expanding. In an interesting effort to determine which CBT-I modality to offer in their clinic, Holmqvist and colleagues compared the impact of a six-session CBT-I intervention for rural dwelling adults offered either through a digital platform (asynchronous) or telehealth (synchronous) (24). Participants who were randomized to the telehealth arm were included as members of an in-person treatment group. Their findings indicated that both interventions were effective in producing meaningful change, including improvements in insomnia severity, fatigue, and work and social impairment. Over the course of the study, participants in both groups also reduced their use of sedating medications. Telehealth delivery of group CBT-I has also been examined as part of the VA’s National Telemental Health Program and CBT-I dissemination efforts. Gehrman and colleagues found that telehealth delivery of group CBT-I is not only feasible in a VAMC setting but also scalable as evidenced by its use in other locations without diminishing effectiveness or acceptability (19). In a 2020 study, this same research group found that for veterans with posttraumatic stress disorder, delivery of CBT-I by clinical video telehealth was non-inferior to in- person treatment (17).

Telehealth CBT-I for individual treatment also has empirical support. Rural dwelling, middle-aged to older adults who participated in integrated telehealth-delivered CBT-I and CBT-D (cognitive behavioral therapy for depression) reported significant reductions in insomnia severity compared to those who received care as usual. Furthermore, these reductions in insomnia severity were clinically significant, with a majority of participants not meeting diagnostic criteria for insomnia at the end of the study. In an earlier stage of this study, this same research group established that rural-dwelling participants were engaged in the intervention and found telehealth-delivered CBT-I to be an acceptable modality of treatment (33). Similarly, McCarthy and colleagues conducted a pilot study that demonstrated that telemedicine-delivered CBT-I (six sessions) is feasible and acceptable for breast cancer survivors living in rural areas (40). Participation in treatment was also associated with improvements in sleep efficiency, sleep latency, insomnia severity, and increase in quality of life and daytime functioning. Researchers found similar results in a pilot study that assessed the feasibility and treatment effect of tele-CBT-I in participants diagnosed with multiple sclerosis (55). In addition to demonstrating that the treatment is feasible (retention and adherence rates were 100%), participants experienced improvements in sleep quality, insomnia severity, fatigue, sleep latency, sleep self-efficacy, depression, and quality of life.

In a novel application of synchronous telehealth treatment, Gieselmann and Pietrowsky conducted a waitlist control study examining the effects of a brief three-session treatment for insomnia, delivered in-person or through synchronous text-based chats (20). Their findings suggest that the chat-based insomnia treatment was as effective as face-to-face treatment for sleep quality and outperformed face-to-face treatment on subjective total sleep time, depression, anxiety, and pre-sleep cognitive arousal. The researchers suggest that the potential benefits of chat-based treatment include focus, sense of responsibility, as well as autonomy.

Non-inferiority trials. Gehrman and colleagues demonstrated that a synchronous telehealth approach produces clinically significant improvements in insomnia severity that are non-inferior to in-person treatment (18). Specifically, they found that CBT-I delivered via telehealth is not clinically inferior to in-person treatment and is superior to a waitlist control. Participants in both treatment groups reported improved insomnia severity compared to the control group, which was maintained at the 3-month follow-up. There were also significant improvements in mood/anxiety and daytime functioning for the two treatment groups compared to the control group. Arnedt and colleagues observed similar post-treatment effects in their study that randomly assigned participants to either six sessions face-to-face or telehealth CBT-I (01). They found sustained improvements in insomnia, fatigue, depression and anxiety symptoms, sleep-related cognitions, and quality of life for both groups. Of note, therapeutic alliance ratings were also similar regardless of whether treatment was in person or via telehealth. Similar effectiveness of synchronous telehealth has also been observed in real-world settings during the COVID-19 pandemic. Martin and colleagues examined how veterans fared in the transition from face-to-face CBT-I to telehealth delivery of CBT-I and found that CBT-I remained highly effective, regardless of treatment modality (39). Taken together, these findings lay the foundation for the use of synchronous telehealth-delivered CBT-I as a promising alternative to in-person treatment that has the potential to produce similar improvements in sleep health and patient-centered care.

Conclusion

Cognitive behavioral therapy for insomnia (CBT-I) is considered a first-line treatment for chronic insomnia. Sleep telemedicine approaches (asynchronous and synchronous) can help to close the gap between clinical research and clinical care and offers an important and promising alternative to expanding care and optimizing sleep health. The studies referenced here indicate that both asynchronous and synchronous telehealth deliveries of CBT-I are effective and have the potential to address long-standing issues of CBT-I dissemination and provide access to care for a greater number of individuals without sacrificing clinical effectiveness. This is an important and growing area of research. Given the many different options for telehealth delivery of CBT-I that all demonstrate promising clinical effectiveness, future research will ideally seek to determine how to best personalize treatment to meet patient needs with equity, efficiency, and effectiveness.