Insomnia is one of the most common complaints at the doctor’s office. It is estimated that chronic and/or severe insomnia affects as many as 20% of all adults, and perhaps even more, since insomnia is often left undiagnosed. According to a 2002 Sleep Foundation poll, 35% of adults reported suffering from insomnia every night, and close to 60% reported trouble sleeping at least a few nights per week. Chronic insomnia is characterized by difficulty initiating or maintaining sleep, often resulting in impaired daytime functioning, overall health, and quality of life.

The cause of insomnia is sometimes hard to pin down and even more difficult to correct. Work schedule, depression , anxiety , and caffeine, alcohol, and nicotine intake may all play a role. In many cases the cause is never determined. Consequently, treatment for insomnia tends to focus on symptoms—not causes. Sleep medications, known as hypnotics, are the most widely used treatment method. Newer drugs, such as zolpidem (Ambien), appear to have fewer drawbacks than some of the older drugs, which have been associated with dependency, daytime sedation, and forgetfulness. Still, none of these medications is a cure, nor are any of them free of side effects. Insomnia may return or even become worse after stopping a medication. Guidelines recommend only short-term or intermittent use of sleeping pills, but long-term use is not uncommon and, unfortunately, long-term side effects have not been thoroughly studied.

Given the drawbacks of medications, several behavioral therapies have been developed to treat insomnia. Cognitive-behavioral therapy (CBT), for one, shows promise, presumably because it addresses a range of possible causes of insomnia, including attitudes, beliefs, and sleep-related behaviors.

Researchers from Harvard Medical School and Beth Israel Deaconess Medical Center in Boston compared CBT and drug therapy in both the short- and long-term. Their findings, published in the September 27, 2004 issue of the Archives of Internal Medicine , demonstrate that CBT may actually be more effective than medication at relieving insomnia.

About the Study

This study included 63 participants, ages 25-64, who had experienced insomnia for six months or longer. For the purposes of this study, insomnia was defined as at least a one-hour delay in falling asleep on three or more nights per week, resulting in at least one daytime impairment (fatigue, diminished mood, or impaired performance). Participants were excluded if they had psychoses, major depression, alcohol or drug abuse, and/or any coexisting sleep disorders (e.g. sleep apnea or restless leg syndrome).

Approximately a quarter of the participants were assigned to each of the following treatment groups for a duration of eight weeks:

  • CBT
  • Drug therapy
  • Combined CBT and drug therapy
  • Placebo (no treatment)

Participants who received drug therapy took zolpidem at high doses during the first four weeks of the study. Doses were tapered down during the final four weeks so as to test whether the drug therapy group would continue to benefit over the long run, after active treatment was suspended.

Participants undergoing CBT received four 30-minute, in-person sessions with a psychologist, as well as a final 15-minute session by phone, all during the first six weeks of the study. CBT treatment was suspended during the final two weeks of the study in order to test if it remained effective after active treatment was suspended. The cognitive component of the CBT sessions was designed to assist patients in recognizing and challenging stressful, distorted thinking about sleep. The behavioral component included a set of instructions for patients:

  • Use the bedroom for sleep and sex only
  • Go to bed only when drowsy
  • If unable to fall asleep within 20-30 minutes, leave the bedroom and engage in a relaxing activity until drowsy
  • Repeat the previous step as often as necessary and for awakenings in the middle-of-the-night

The participants kept sleep diaries before they began treatment, at mid-treatment, and just after treatment ended. “Nightcap” sleep monitors (which record and analyze eyelid and head movements) were used to objectively compare sleep quality on three nights before and after treatment. Daytime functioning scales were also administered before and after treatment to measure changes in mood and energy level.

The researchers used these data to assess sleep latency (the amount of time it took participants to fall asleep), total sleep time, and sleep efficiency (the percentage of time allotted for sleep that participants were successfully able to sleep).

The Findings

Because this study was so small, there were few statistically significant findings. From pre-treatment to mid-treatment, both the CBT and combination groups showed a 44% reduction in sleep latency, the drug therapy group showed a 29% reduction, and the placebo group showed a 10% reduction, but this difference did not quite reach statistical significance. There was also no significant difference in total sleep time between the CBT and drug therapy groups. What was significant was the finding that CBT participants experienced the greatest improvement in sleep efficiency by mid-treatment, possibly due to better planning and time management regarding sleep. Combined treatment had no advantage over CBT alone for any of the variables.

By post-treatment—after both drug therapy and CBT counseling had been suspended– nearly 60% of participants who received CBT achieved “normal” sleep latency rates, which means they took less than thirty minutes to fall asleep (as measured by Nightcap recordings). Also, this group’s overall reduction in sleep latency increased to 52%, while the reduction for the drug therapy group dropped to only 14%– and these findings did reach statistical significance. In other words, drug therapy did not have lasting effects after treatment was suspended, whereas CBT remained effective in the long run, even after counseling stopped.

Finally, there were no significant differences among groups on changes in daytime mood before and after treatment. The researchers point out that all participants showed only minimal levels of depression and had fairly normal energy level scores before treatment, which may not have allowed much room for improvement in overall daytime functioning. However, this finding does limit the scope of claims to be made for CBT based on this study.

How Does This Affect You?

This study suggests that cognitive-behavioral therapy should replace drug therapy as the first-line intervention for chronic insomnia, or at the very least it should be considered a viable option, even for quick relief from current symptoms. The findings clearly demonstrate the benefit of CBT over medication in the long-term, and they also suggest that CBT may be more effective in the short-term—though most findings were not statistically significant. (Future, larger studies may be better able to demonstrate CBT’s short-term effects.) Interestingly, there appeared to be no advantage to combining drug therapy with CBT, a practice that is currently common.

Doctors and patients alike need to be more aware of the value of CBT for insomnia. They also need to be less reliant on drug therapy, especially due to the potential for adverse effects and long-term abuse. In addition to its superior long-term success rate, CBT for insomnia seemed to be reasonably cost-effective, demonstrated by the fact that only five brief sessions were necessary to return patients to normal sleeping patterns in this study. But, while CBT may be cost-effective, time and availability will continue to be major obstacles to its widespread use. Pills, unfortunately, will always be easier to prescribe and simpler to take.