Ditch the Sleeping Pills and Opt for Cognitive Behavioral Therapy for Insomnia

Michelle Jonelis
9 min readMar 11, 2024

Did you know that sleeping pills are no longer the go-to remedy for those struggling with sleep? A growing body of evidence suggests that Cognitive Behavioral Therapy for Insomnia (CBT-I) is a much better choice. Major medical organizations now recommend CBT-I as the first line treatment for insomnia, and for good reason. It stands toe-to-toe with sleeping medications in the short term, and outshines them in the long run. Plus, the risk of harm is significantly lower with CBT-I compared to sleeping pills.

Despite these findings, sleeping medications remain the top choice for treating insomnia. Shockingly, a mere 5% of insomnia patients were referred for CBT-I by their primary care physicians, while a whopping 52% were handed prescriptions for sleep aids according to a 2021 study published in the Journal of Clinical Sleep Medicine.

Why is CBT-I recommended instead of sleeping pills? Let me walk you through the data:

Sleeping Pills Eventually Stop Working

Sleeping medications (prescription, over-the-counter and herbal products such as cannabis) work by changing the levels of the neurotransmitters in the brain involved in sleep and wakefulness, such as GABA, acetylcholine, norepinephrine, serotonin, dopamine, histamine, and orexin/hypocretin. When we first start taking these medications our brains suddenly have more sedating chemicals or fewer alerting chemicals circulating and we feel relaxed, allowing us to drift more easily into sleep. Over time, however, our brain realizes it will be getting a sedating substance every night and stops producing as much of its own sedating neurotransmitters. Once this happens, the sleeping medication will be less effective at producing sedation, a phenomenon known as tolerance.

At this point, if we do not take the expected pill one night, our brain will be unable to produce enough of its own sedating neurotransmitters and we will feel even more anxious and awake than we did at baseline, a phenomenon known as dependence. We will mistakenly be led to the conclusion that we “need” the medication, when in reality what we are experiencing is simply withdrawal effects. Even with “as needed” or intermittent dosing of sleeping medications our neurochemistry can change in a manner that makes it increasingly hard to sleep without the medication.

The amount of time that someone needs to be on a sleeping medication before developing tolerance and dependence depends on that person’s age, the type of medication, the dose of medication, and genetic make-up. Some people develop tolerance and dependence after just a few nights, in other people it can take many months. After a long enough period of time, however, tolerance and dependence develop in almost everyone.

Because of these changes in our neurochemistry which lead to tolerance, dependence and withdrawal, research has shown that people who chronically use sleeping medications do not actually sleep any better than people who do not use these types of medications.

Sleeping Pills Erode Our Confidence in Our Ability To Sleep

In addition to physiological dependence and tolerance, there is another phenomenon called psychological dependence which can be just as problematic for people using sleeping pills. Psychological dependence means your brain has come to believe that the only way you can sleep is by taking a pill. If you do not have access to that pill (even if you are not physiologically dependent on it) you will worry that you will be unable to sleep, which will cause increased sympathetic nervous system activation at night which actually does interfere with your sleep. Psychological dependence can even happen months to years after you stop taking a sleeping medication when your sleep again becomes disrupted and you feel that resuming sleeping pills is your only option for improving your sleep.

Another way to put this is that taking a pill for sleep robs you of your ability to trust your own brain and body. It robs you of the chance to practice alternative strategies that might help you relax at night, such as staying up a bit later, writing out your worries, doing deep breathing and even just accepting that wakefulness can be a normal part of life.

Sleeping Pills are Less Effective Than Most of Us Think

Sleeping medications are much less effective than most people taking them (or prescribing them) realize. For instance, a 2012 study found that while Z-drugs (such as zolpidem) reduced objective sleep latency (time to fall asleep on an in-lab sleep study) by 22 minutes on average (compared to placebo), total sleep time for the night only improved by 14 minutes on average, which was not even significantly different from placebo. It turns out that getting someone to fall asleep 22 minutes faster WITHOUT a medication is not very hard to do. You can simply have them stay up later and/or reduce the amount of blue light they are exposed to before bed.

Sleeping Pills Do Not Produce Sleep

All sleeping medications currently on the market either produce sedation, which is different than sleep, and/or distort normal sleep architecture. Natural sleep is produced by a combination of different neurotransmitters involved in sleep and wakefulness, with the brain levels of each fluctuating over the course of the night to produce different types (or “stages”) of sleep. You need all of these neurotransmitters, and in the correct amounts, at the correct times, in order to produce natural sleep. Sleeping medications typically only act on a single neurotransmitter in the brain and therefore can never produce the same type of sleep that our bodies naturally produce. Most sleeping medications are actually felt to reduce the overall depth of sleep over the course of the night, even if there are fewer awakenings.

One way that commonly prescribed sleeping medications work is by producing amnesia, meaning that you will not accurately store memories of events which occur when the medication is circulating in your system. This means that the medication may not improve your sleep at all but will trick you into thinking that you slept well, simply because you do not remember being awake. (Flight attendants are well aware of these effects and sometimes refer passengers on sleeping medications as “Ambien zombies”.)

Sleeping Pills Do Not Improve Health, Cognition or Daytime Function

Surprisingly, despite the fact that poor sleep is associated with worsened mood and overall health, giving sleeping medications to people with difficulty sleeping does not actually improve their mood or health. News headlines caution us that poor sleep worsens our memory, but taking sedative hypnotics has never been shown to improve cognition or memory.

In fact, sleeping medications do not even improve objective next-day function in research studies funded by the drug makers themselves. For instance, next day driving is at least mildly impaired following administration of nearly all sedative hypnotics currently available.

All Medications Have the Potential for Side Effects

Just like all medications and supplements, sleeping pills have the potential for side effects ranging from those attributed to increased sedation and amnesia (falls, confusion, motor vehicle crashes, doing things you would not normally do without memory of having completed them) to allergic reactions, to more bizarre yet very serious issues such as a prolonged and painful erection (priapism, which can occur with trazodone).

CBT-I is Effective with Minimal Risk

Research shows that CBT-I is effective in reducing the severity of insomnia in about 70–80% of people. CBT-I remains effective in people with co-existing medical and/or psychiatric conditions. CBT-I reduces wakefulness during the night and improves daytime fatigue. Cost-effectiveness analyses of sleeping medications versus CBT-I favors CBT-I. Importantly, the benefits of CBT-I vastly outweigh the potential for harms (which are minimal).

Why Are Healthcare Workers Not Recommending CBT-I?

The main justification used for not recommending CBT-I to patients with insomnia is that it is not accessible. In the past, CBT-I had to be done by a psychologist trained in Behavioral Sleep Medicine and, unfortunately, these remarkable individuals are few and far between. In recent years, however, app and web-based CBT-I programs have proliferated, many of which are excellent. App and web-based CBT-I programs have been shown to have good efficacy in research trials, even in individuals with medical and psychiatric comorbidities. One trial even found a benefit from digital CBT-I in patients with Mild Cognitive Impairment. Digital CBT-I is also relatively low cost to the patient compared to months or years of copays for sleeping pills. (A CBT-I app developed by the VA is even available for free: CBT-I coach). There are some groups of people who cannot access digital CBT-I such as people who do have access to or are unable to use a computer or smartphone, people who are not proficient in English, people with cognitive or memory impairment or very low health literacy. Nevertheless, most patients who would be able to engage with digital CBT-I are not even offered this as an option. Those of us who work in the behavioral sleep medicine field are also actively working to train more providers who can competently provide CBT-I and CBT-I variants, which will hopefully will further increase access in the future.

Another justification for using sedative hypnotics to treat insomnia is that CBT-I is not always effective. It is true that about 20–30% of patients do not improve in research trials of CBT-I. There is no evidence to suggest, however, that sleeping medications work better than in the general insomnia population for this 20–30% of individuals. The same issues with sleeping medications: physiological and psychological dependence, lack of demonstrated benefit to health and cognitive function, risk of side effects, still applies to this population. What should we do then for the 20–30% of folks who do not respond to CBT-I? We are still not sure but research is starting to show that at least some of them may respond to other types of non-CBT-I behavioral therapies such as Acceptance and Commitment Therapy for Insomnia.

What If You Currently Take Sleeping Pills?

First of all, know that you are in good company: recent data suggests that at least 50% of Americans took a prescription, over the counter or herbal medication for sleep during the Covid-19 pandemic. The media has been telling all of us that sleep is important for our health and that if we don’t sleep well there will be consequences. The fact that you take a sleeping pill means that you care about your sleep and that you have been paying attention to this message. Unfortunately, you now understand that the best way to fix your sleep is not with a pill but instead with behavioral therapy for insomnia. If you have become dependent on a sleeping medication and cannot sleep without it, do not despair! Just as our brains adapt to getting a sleeping medication each night, we can get them to adapt to not getting a sleeping medication at night. The trick is just to reduce the dose of the medication little by little, very slowly, so your brain can again start to produce normal levels of sedating neurotransmitters. If you have tried to stop a sleeping medication in the past and failed, it was most likely because you tapered off of it too quickly. Before or while doing a taper, you should work with a Behavioral Sleep Medicine provider or complete an app or web-based insomnia program to learn alternative coping strategies for sleep loss. You can most likely get off any medication, at any age, with a slow enough taper. Some of these types of tapers may take a year or longer, but if you are patient you can eventually get off. (For example, see these guidelines from Canada). Please do not taper off any medication without consulting the person who prescribes you the medication.

In summary, sleeping medications all have risks associated with them, lack demonstrated long-term efficacy and nearly always lead to dependence and tolerance. Additionally there is no evidence that taking sleeping pills improves health, mood or cognition over the long-term. On the other hand, Cognitive Behavioral Therapy for Insomnia is safe and equally effective in the short-term, more effective than sleeping medications in the long run. Even if someone does not respond to CBT-I there are other behavioral treatment options that might work better for them than medications. I know which treatment modality I prefer and I hope that after reading this article you feel the same.

Here is a nice graphical representation of the pros and cons of CBT-I versus sleeping pills that I played no role in creating.

Here is a link to my list of digital CBT-I programs.

Here is a link to two databases of behavioral sleep medicine clinicians who can do CBT-I with you: https://www.behavioralsleep.org/index.php/united-states-sbsm-members AND https://cbti.directory/

ABOUT THE AUTHOR:

Dr. Michelle Jonelis is the founder and medical director of Lifestyle Sleep, a lifestyle-focused sleep medicine clinic in the San Francisco Bay Area. Michelle is board certified in sleep and lifestyle medicine with additional training in behavioral sleep medicine. Her clinical focus is on the non-pharmacologic management of sleep disorders using techniques such as Cognitive Behavioral Therapy for Insomnia, bright light therapy, circadian rhythm optimization, and Lifestyle Medicine. Michelle is passionate about advocating for improved patient access to Cognitive Behavioral Therapy for Insomnia and other evidence-based treatments in sleep medicine and has served on several professional committees through the American Academy of Sleep Medicine and the Society for Behavioral Sleep Medicine on these topics. When she is not practicing medicine, Michelle enjoys spending time with family, hiking, biking, yoga, cooking and spending as much time outdoors as possible.

--

--

Michelle Jonelis

I am a sleep medicine physician in Marin County, CA. My clinical focus is on the non-pharmacologic management of sleep disorders using techniques such as CBT-I.