• Have sleep problems? Ask your doctor about cognitive behavioral therapy

    Have you ever seen a commercial that says, “If you have insomnia, ask your doctor about cognitive behavioral therapy” (CBT)? I haven’t. But I have seen many that urge consumers to talk to their doctors about Lunesta (Eszopiclone) or Ambien (Zolpidem) or Restoril (Temazepam) or other insomnia drug.

    Does that mean the drugs are more effective than CBT? Actually, no. But they are more costly for payers and more lucrative for pharmaceutical manufacturers, generating revenue that supports expensive ad campaigns. And insomnia and other sleep problems are a huge market. About 30% of adults report some symptoms of insomnia and 10% meet all the criteria for the condition.

    I’m one of them. However, instead of turning to one of the prescription drugs, I’m trying CBT. I’ll write more about my experience with it after I’m further along in the program I’m following (designed by Dr. Gregg Jacobs and recommended by my physician). For now, I want to go over some of the underlying research.

    You can get up to speed on CBT by reading the Wikipedia entry on it. But it may suffice to know that its “a psychotherapeutic approach that addresses dysfunctional emotions, behaviors, and cognitions through a goal-oriented, systematic process” with the aim of “changing maladaptive thinking [that] leads to change in affect and in behavior.” In short, get right by thinking right.  Obviously that’s not going to be effective for everything. But for insomnia, it is. The studies say so, and some say it’s more effective than pharmacological therapy (PCT).

    For example, in a comparative effectiveness, randomized controlled trial (RCT), Morin and colleagues compared the effects of CBT, PCT (drug: Temazepam), CBT + PCT combined, and placebo in subjects ages 55 years and older. Though they report numerous results consistent with CBT leading to longer lasting improvements in sleep than other approaches, their Figure 4 (below) provides one summary. The bars represent the number of minutes subjects were awake after sleep commenced (sleep onset), according self reports (sleep diaries).* Notice that for all follow-up intervals CBT outperformed all other approaches.

    In another RCT focused on 25-64 year olds, Jacobs and colleagues found that CBT outperformed PCT (drug: Zolpidem) and placebo and was as effective as CBT+PCT combined therapy in sleep efficiency (percentage of time devoted to sleep that the subject is actually sleeping) and sleep onset latency (how long after intention to sleep the subject actually fell asleep). The charts are below.

    Based on these studies, there does not seem to be any lasting advantage to PCT over CBT. I acknowledge that these are just two studies, but they are the only two RCTs comparing CBT to PCT (alone and in combination with CBT) that I am aware of. (Update: there is a third, about which I will write another time.) I will be reading other CBT studies and will likely blog on them as I do. It’s worth noting that a limitation related to the study of CBT is that a placebo is not possible, as it is for drugs. In the trials discussed above “placebo” means a drug placebo. You can’t blind the patient or practitioner to CBT.

    Even with that limitation, since CBT will cost me very little and can be self-administered (and effectively so, according to another RCT), I consider it worth the effort for me. It’s cheaper and it seems to work better. Maybe fewer Americans should pop pills and more should ask their doctors about CBT for insomnia.

    * The authors also show results based on objective measures of sleep, which are generally consistent with the subjective-based findings. The subjective measures (sleep diaries) are not considered inferior in this type of work because self assessment of sleep (concern about insufficient sleep) is one of the key psychological drivers of insomnia. Insomnia can be relieved in a patient who thinks he is sleeping better. Put another way, subjective improvement is improvement.


    Comments closed
    • I’m not an expert on the subject, but quite relative to sleepless nights or restless sleep patterns is the fact most people suffer from exhausted adrenal gland issues.

      When the adrenal gland is exhausted, the body gets a lot of problems with cortisol, adrenaline, nor-adrenaline and melatonine levels spiking and dropping.

      People tend to ‘fix’ the problems that comes with cortisol drops with caffeine intake. Which is not the solution, rather the extension of the problem.

      Using sleeping pills won’t help in the long run either. The body doesn’t get the needed nutrition to heal itself from the problems that exists.

      First, help the adrenal gland to be healed. Next, let the body get it’s proper rest. Use herbal supplements that don’t interfere with the hormones made by the adrenal gland. Make sure your body get’s a sound sleep within the completion of the five cycles of sleep.

      My own problems have to do with sugar intake and the rushes that comes with it. Since I stopped using sugar in coffee, tea etc. and stopped using coffee alltogether my sleep has improved in a major fashion. Next I started using herbal supplements that don’t make me feel drowzy in the morning but rather give me a sound and solid sleep. I’m on my way to good rest now since decades.

    • Well said Austin – a great summary of the key points from the literature and the absurd imbalance in prescription in the context of that evidence base. However it *is* possible to subject CBT to placebo-controlled trial; we attempted a placebo-controlled RCT with our own web-based CBTi course, Sleepio. The paper has been accepted for publication by SLEEP – you can read it here:

      In short we used a placebo intervention dubbed ‘Imagery Relief Therapy’ or IRT. This passes the ‘Google test’, in that ‘imagery’ is a real therapeutic technique, as is desensitisation (the core conceptual justification of IRT), but these concepts are combined in a bogus, therapeutically-neutral way.

      Of course delivering this as a true placebo involved going to insane lengths: creating a complete alternative version of our system (which is very media-rich), complete with false online tools and sessions (take a look at the research material at http://www.sleepio.com/research). The participants in both groups were given exercises requiring comparable levels of activity, and critically both were considered equally credible by those respective groups. In fact, we had near-identical (and very high) adherence rates in both placebo and active groups, despite the significant difference in clinical improvement between groups. This is heartening: it not only suggests the placebo was a good one (ie. convincing enough for the participants receiving the bogus intervention to stick with it) but it endorses the engagement potential of the ‘wrapper’ used to deliver both interventions – the Prof character, online community and mobile support.

      Anyway, would love your views on the research paper, and good luck with overcoming your insomnia! Incidentally, I overcame my own insomnia with CBT, which is what inspired me to develop Sleepio…so hang in there!

    • Also, here’s a 2009 meta-analysis of the literature: http://www.ncbi.nlm.nih.gov/pubmed/19201632