Intensive Sleep Retraining

A few weeks ago, I attended the annual international sleep conference where new research findings and treatment approaches in sleep medicine are presented by leading researchers and clinicians from around the world. By "attended," I mean that, like every other conference I have been to this year, I viewed online presentations and meetings, some in real-time, some recorded. This is another one of the ongoing impacts of the COVID-19 pandemic and the restrictions on day-to-day life it has imposed. Never before had this conference been held online, but, while I missed having in-person meetings with colleagues I see only once a year, the conference itself was excellent with a good deal of new information being presented.

As you might expect, insomnia continues to be the most common sleep problem in our society, and efforts to find new treatments are ongoing. The treatment of insomnia is a mature field, and people have been trying to find ways of alleviating their sleepless nights throughout human history. In the past 40 years, cognitive, behavioral, and new pharmacological treatments have been developed based on our increasing scientific knowledge of sleep.

Coming up with a new treatment, however, turns out to be quite difficult. A few new medications have appeared in recent years, but they’ve tended to have limitations similar to those of the old standards, such as lack of effectiveness for many patients, the development of tolerance, and even dependence if used frequently over a long period of time.

Developing a new cognitive-behavioral approach is also very challenging, and progress has been slow in recent years. Indeed, many of the presentations at the sleep conference dealt with issues such as making existing treatments more available to the large numbers of people suffering from insomnia and extending these approaches to populations that suffer health care disparities. These populations include minority communities in the United States and across the globe. Shortages of sleep clinicians who are trained in cognitive-behavioral interventions, combined with a long-existing, widespread lack of available and affordable health care coverage, cause significant challenges to the successful treatment of insomnia for large portions of the population.

I have previously addressed the currently most effective treatment approach for insomnia, cognitive behavioral therapy of insomnia (CBT-I), along with its various component methods. These techniques work for about 70 percent to 80 percent of the chronic insomnia patients who use them. They are straightforward and are based on our scientific understanding of sleep. However, they can be difficult to apply in a given individual’s situation as they require several weeks of intensive effort and disciplined use before results are fully realized.

This is challenging for many people. While they typically produce better long-term results than sleeping medication, using CBT-I techniques is harder work and more difficult than taking a pill that works the first night you use it. CBT-I does have the benefit of not having negative side effects (such as loss of effectiveness with regular use, as with most sleeping medications) other than some potentially increased daytime sleepiness when first restricting time in bed as a part of sleep restriction.

One of the messages that came through at the conference was that if these techniques are difficult for some insomnia patients with good insurance, regular work hours, and stable family situations to effectively use, they may be nearly impossible to implement in disadvantaged family settings, where families lack good health insurance, the household environment is chaotic, and people are working highly variable work schedules as they participate in the demanding “gig economy.”

So I was very interested to see a presentation by Hannah Scott, a Ph.D. candidate at Flinders University in Australia, who presented on the use of technology that involves a novel application of a relatively new, but expensive, approach to treating sleep onset insomnia, Intensive Sleep Retraining (ISR). I will return to her presentation below.

ISR is based, as are the standard behavioral techniques of sleep restriction and stimulus control, on building sleep drive through sleep deprivation and deconditioning of the negative associations between the bed environment and the hyperarousal that prevents relaxing into sleep. It does this by utilizing gradually building sleep deprivation along with reassociating the bed with drowsiness and restorative sleep. Instead of looking at your bed at night and thinking, “Oh, man, I’ll probably be up at least half the night again,” you come to see your bed as a comfortable place where you will soon be drifting off to a night of good sleep.

Sleep Essential Reads

Sorry, but Science Says You Should Not Sleep In on Weekends

Beyond Snoring: Unveiling the Hidden Epidemic of Sleep Apnea

How does ISR work? It is primarily targeted at sleep onset problems. In other words, it works best for the kind of insomnia in which patients have difficulty falling asleep when they first get into bed. Patients are instructed to limit time in bed to five hours, using an alarm, the day before ISR. They are then seen in the sleep laboratory, where they are set up for sleep EEG recording, and over a 24-hour period are given repeated opportunities, every 30 minutes, to fall asleep. If not asleep within 20 minutes, the trial is ended.

Each time they fall asleep, which is noted by the use of the EEG recording, they are awakened three minutes after dozing off. After each three-minute sleep episode, they are asked if they felt like they had fallen asleep and, after answering, are provided information from the EEG about whether or not they had actually fallen asleep. Between trials, they sit in a chair to maintain wakefulness until the start of the next trial. This is repeated time and again for 50 trials.

As you can imagine, drowsiness and sleep drive increase progressively as time goes on. With repeated episodes of falling asleep, the patient becomes more aware of the sensations of falling asleep. This is reinforced with feedback from the EEG recording about their onset of sleep. This is very important as many people with insomnia have difficulty distinguishing between fatigue and sleepiness. The distinction is significant because you can be very tired but not sleepy, and the feeling of fatigue is not a good guide for readiness to sleep while drowsiness is.

ISR was introduced in a small pilot study (Lack & Baraniec, 2002) that was later extended to a larger sample by Harris et al. (2007). While it is a treatment technique that works well, it is very difficult to use in practice because it requires having a patient sleep in a laboratory while being monitored and instructed all night by a trained polysomnography technologist. Obviously, these are significant limitations. ISR uses expensive and scarce resources, is cumbersome, and can be difficult to incorporate into a patient’s life, as a full day in the sleep laboratory is required.

Nonetheless, research has indicated that this approach is successful, as patients rapidly experience benefits such as falling asleep about 30 minutes more quickly, being awake about 30 minutes less during the night, and getting about 60 minutes of extra sleep per night after the retraining. Standard CBT-I techniques, such as stimulus control and sleep restriction, may take up to several weeks to have an equivalent effect. A randomized controlled trial of ISR (Harris et al., 2012) showed that it could produce rapid improvement in nighttime sleep and daytime functioning in participants with chronic sleep-onset insomnia.

Furthermore, adding it to stimulus control therapy resulted in superior results over stimulus control alone. Boyle et al. (2018) found that when comparing ISR to standard CBT-I, ISR was about as effective as CBT-I for sleep onset problems, but not as effective for sleep maintenance problems. They noted that ISR could be most effective for patients who have not benefited from standard CBT-I.

Could ISR be extended to home use with a less expensive and intrusive method (see Lack et al., 2017)? Several efforts have been made, including an iPhone app called “Sleep On Cue.” I do not know of research that has looked at the effectiveness of this app, which is designed to provide ISR, but it is inexpensive, is available for the iPhone, and may be worth investigating if you are interested in trying this approach. The developers point out that this is not a clinical device and recommend talking with a medical professional about sleep problems before using it.

A problem with this app is that it requires holding your phone while the app emits a faint sound through earbuds every few minutes. The individual must then give a faint shake to the phone to let the app know they are still awake. After falling asleep and missing responding to the sound, the app will cause the phone to shake more vigorously, which wakes the person up. The app asks if the person thinks they fell asleep so that awareness of drowsiness is enhanced.

The app prompts the person to get out of bed for several minutes and then return and repeat the process. This could be difficult for some people to actually use under real-world conditions. It does provide a protocol that is fairly closely related to the laboratory-based intervention but without the accurate measurement of sleep by EEG recording.

During the sleep conference, Hannah Scott presented data about a new device, the THIM, that could be more accurate than the phone app and make ISR more practical and available in the home environment. THIM is not currently available for purchase but may be in the near future. In her presentation, she described the THIM as a ring-like device that is worn on a finger along with headphones that provide a gentle sound every few seconds.

In her study, it was used along with an EEG recording to measure its accuracy in detecting sleep and providing accurate feedback to the participants. It has a built-in battery and connects by Bluetooth to an app that provides feedback to the user. After placing the ring on a finger, the user attempts to fall asleep. The THIM provides mild vibration from time to time. The wearer taps the ring gently to signal that they are still awake.

When there is no response for two consecutive vibrations, it is assumed that the user is asleep, and a strong alarm vibration is emitted by the device to wake the individual. After a short break, another attempt to fall asleep is made. This process led to rapidly building sleep drive, to high circadian sleep drive in the early morning hours, and to the extinguishment of the conditioned arousal to the bed environment. Results indicated that this was an effective way to deliver ISR in the home environment. Data collection has been halted due to the pandemic, although Ms. Scott hopes that research will be able to resume soon.

Some possible limitations of this mechanism include patients getting tired of responding to the vibrations and giving up on treatment, and the possibility of the unit slipping off the finger during the night without the patient being aware of it and thus losing the effectiveness of the treatment. The possibility exists that new technologies, whether app-based or instrument-based, may soon allow for interventions that were previously possible only in the laboratory. Given the significant suffering caused by insomnia and other sleep problems, especially during this pandemic era, it is my hope that techniques such as ISR will become commonplace—readily available and easily affordable.