New modalities may extend treatment for insomnia

While cognitive behavioral therapy has been proven effective for insomnia, and there are new delivery models to accommodate patients, comparative efficacy of online versus therapist-led treatment is uncertain.


Many professional societies, including ACP, now recommend cognitive behavioral therapy as first-line treatment for insomnia (CBTI). But internists may wonder if all patients are good candidates for this strategy, and what happens when it doesn't work at first.

CBTI has been proven effective in more groups than one might suspect, and there are new delivery models to accommodate patients who are resistant or nonresponding, as well as telehealth options for the pandemic era, said Rachel Manber, PhD, DBSM, a professor of psychiatry and behavioral sciences at Stanford University in California. She spoke about new frontiers of CBTI during SLEEP 2020, which was held virtually in late August.

Patients may consider their insomnia experience as unique and require personalized treatment rather than an automated therapy program Image by microgen
Patients may consider their insomnia experience as unique and require personalized treatment rather than an automated therapy program. Image by microgen

ACP published a clinical practice guideline in the July 19, 2016, Annals of Internal Medicine recommending that for initial treatment of chronic insomnia disorder, “all adult patients receive cognitive behavioral therapy for insomnia,” a strong recommendation with moderate-quality evidence. The National Institutes of Health, the British Association of Psychopharmacology, and most recently the European Sleep Research Society and Australasian Sleep Association have issued similar recommendations.

Research shows that CBTI is beneficial in patients with insomnia with comorbid depression or post-traumatic stress disorder, as well as older patients, said Dr. Manber, who is director of Stanford's sleep health and insomnia program and deputy editor-in-chief of the journal SLEEP. CBTI is also effective in adolescence, an emerging area of research, as well as during pregnancy, postpartum, and perimenopause. While CBTI is most often thought of as a face-to-face treatment, a recent study found that telemedicine delivery was noninferior to in-person therapy, “which is a great relief for all of us who are doing our CBTI work via telemedicine” during the COVID-19 pandemic, Dr. Manber said.

However, comparative efficacy of online versus therapist-led CBTI is uncertain, she told conference attendees. She said certain research found a response rate that was approximately 20% lower for mobile and digital interventions, but this difference was not statistically significant. A study Dr. Manber cited, published in the July 2015 Behaviour Research and Therapy, found no difference between internet-based and in-person group CBTI.

“So this is definitely an area in which we don't know the answer,” Dr. Manber said. “Probably, that will depend on the sample, but it's to be discovered.”

Room for improvement

To further improve CBTI outcomes, Dr. Manber identified three frontiers: enhancing engagement, identifying mechanisms that can serve as targets for improving intervention, and identifying groups at highest risk.

On the subject of engagement, Dr. Manber cited results from evaluation of group CBTI at Stanford that was published in the April 2008 Journal of Psychosomatic Research, which found that a short subjective sleep duration with a cutoff of 4.5 hours predicted dropping out of CBTI.

Other research, such as a study published March 14, 2019, by the Journal of Cognitive Behaviour Therapy, has shown that although CBTI is effective for people with anxiety and depression, elevated anxiety and depression symptoms are additional predictors that CBTI may not work as well.

“The clear issue is that we need to do more work to identify predictors of poor engagement in large samples and potentially for different modalities,” she said.

The third frontier involves identifying who is at risk for poor outcomes. Short objective sleep duration may play a role here, although research on this question has yielded mixed results, Dr. Manber said. “So the jury to date is still out. … We need to do more work to identify other risk factors, as well as investigate this one in larger samples.”

Another predictor that has received a lot of attention recently is cognitive variables, such as rumination, locus of control, and beliefs about sleep. Dr. Manber noted that “Targeting cognitive factors with cognitive therapy may be particularly relevant for those with depression and anxiety.”

Addressing availability

Dr. Manber also discussed ways to increase CBTI's availability in primary care. Delivery modes beyond face-to-face therapy include digital/online, mobile, automated text, video, and chatbots. Delivery can be fully automated, as well as supported or guided by professionals with some training in the area, such as sleep coaches, primary care nurses, or peers, she said.

Digital CBTI can eliminate wait time for the therapy, considering the number of clinicians who provide CBTI is limited. “We reduce the need for missing work, child care, travel, and make it more convenient for people to fold treatment into their life,” she said. “We reduce the stigma and embarrassment about seeking mental health.”

But barriers to patient uptake include the impersonal nature of online CBTI. “Some people crave accountability,” she said. “And there's a greater personal support and some evidence that greater personal support is associated with better outcomes.”

Online or virtual CBTI also requires proficiency and comfort with use of technology; in addition, patients will have different levels of knowledge about sleep, so programs must cater to the average level and may not meet every patient's needs or expectations, Dr. Manber noted. She offered two recent comments from participants as real-life feedback: “One of them is saying the level of the presentation does not fit my existing level of knowledge about sleep. And the other one says, more poignantly, ‘I feel like a fourth grader in a second-grade class. I already know a lot of what was presented to me.’”

Additional limitations from patients' perspectives were summarized in qualitative, structured interviews with 55 patients in a study published in the January-February 2019 Behavioral Sleep Medicine. Themes from that feedback include worry about whether the information provided online is accurate, “just the fact that we are overwhelmed with so much information we don't know what's right anymore.” Also, patients may not be comfortable sharing personal details via the internet or may dislike one-way communication in the therapeutic process. From a patient perspective, “There is the sense that ‘My insomnia experience is unique and I need something very personalized to me,’” Dr. Manber said.

Hybrid models of care may be one way to help tailor CBTI to the individual patient. Hybrid delivery methods include concomitant and stepped-care approaches, “but as we're thinking about hybrids, we need to think about what might be facilitators and barriers of digital CBTI when we want to combine them,” Dr. Manber noted.

One concomitant model reported in the May Psychotherapy and Psychosomatics involved five weekly sessions where patients were sent videos and case examples. Online coaching from community nurses was provided to promote engagement and adherence. On average, nurses spent about five to 20 minutes per session, for a total of 1.5 hours per patient, and the therapy proved to be effective.

Dr. Manber and her colleagues are studying stepped-care approaches that utilize digital and therapist delivery of CBTI at Stanford. In one sequential step-care model, which they are testing in the Apnea and Insomnia Research (AIR) Study, every patient initially gets online CBTI. Those who have adequate response can continue, while those do not may be switched to an in-person therapist.

Triage stepped care, which Dr. Manber described as more sophisticated, is motivated by the observation that once patients have an unsuccessful experience with therapy, they are less likely to want to re-engage in it in a different way. The model uses a simple five-item checklist to identify who is likely to benefit from online therapy and who might need to be assigned to an in-person therapist right away. A patient who does not meet any of the five checklist items receives online treatment, while a patient who does goes directly to an in-person therapist. Patients who start with online therapy but do not respond to treatment within eight weeks can then switch to an in-person therapist, Dr. Manber said.

While CBTI is effective across many populations and via many methods, more work is needed to integrate it into the health care system, Dr. Manber concluded. “I think we need economic analysis to convince health insurance companies to support online and hybrid interventions,” she said. “And we need to know more about how to promote patients' engagement in real-life settings.”