Mortality and Cancer Risks, Which Pills to Avoid & Better Alternatives
Last Revised April 2019
Read this eBook: Chapter by Chapter | Single Page | Printable PDF
Mortality and Cancer Risks, Which Pills to Avoid & Better Alternatives
The Benefits of Hypnotics
I have described the dark side of hypnotics and described the alternative correction of habits and attitudes, because these are the most important points about sleeping pills. I did not describe any sleeping pill benefits until this Chapter 4, because in my view, the risks of death, cancer, depression, and infection with sleeping pills, besides the behavioral impairments and accidents, are much more important than any small benefits. Besides, use of sleeping pills seems to cause insomnia, at least after withdrawal.
A laborious and somewhat misplaced effort has employed sleep researchers over the years to measure the small amount by which sleeping pills increase sleep. I will not bore you with the details. The effort is misplaced, in the sense that the prescription sleeping pills increase sleep only very little, so that the exact size of the tiny benefit is trivial.[47] In fact, at the low drug doses that the FDA recommends for safety, popular sleeping pills do not reliably increase sleep at all. The manufacturers of Ambien and Belsomra (zolpidem and suvorexant) have written the FDA that these low doses are “ineffective.” Likewise, the official information on Sonata (zaleplon) stated, “a significant difference from placebo on sleep duration was not demonstrated,” which means that zaleplon generally did not help people sleep more than a dummy pill. I think the recommended doses of Lunesta (eszopiclone) and Rozerem (ramelteon) are similarly ineffective.
In most sleep laboratory studies, sleeping pills given to insomniacs increase their self-reported sleep only 20-40 min. or even less. The EEG-laboratory recorded benefit of sleeping pills is still less than what patients report. These are only trivial increases, when we consider that many people who sleep only five hours do not complain of insomnia, whereas there are people who report sleeping nine hours or more who feel their insomnia is severe. As I have mentioned above, although 20 min. increases in sleep may be statistically significant (which means statistically reliable), they are not functionally significant, since sleeping pills usually produce no measurable improvements in daytime performance or health.
Ramelteon (Rozerem) may offer little risk (we did not have enough data in our epidemiologic study for ramelteon), but it also offers little benefit. In short-term studies, Rozerem produces a small decrease of time to objective EEG sleep of seven to 16 minutes, which is trivial. After six months, Rozerem increased total sleep by one minute compared to placebo! According to the NDA data at the FDA web sites, in many of the company studies, patients who received Rozerem did not think they were sleeping better than those receiving placebo. However, if many patients taking ramelteon do not feel they are sleeping better, why buy the stuff? I have found that many patients do not like Rozerem. We do not know about mortality, but some indications suggest that ramelteon might cause depression, infection and cancer.
“The European Committee for Medicinal Products for Human Use (CHMP) has issued a negative opinion on the use of the melatonin receptor agonist ramelteon in insomnia, due to its unfavorable risk-benefit balance.”[48] They thought melatonin itself might have a better benefits/risk ratio for treating insomnia.
I agree with the European opinion.
Whereas most sleeping pills increase sleep a few minutes for the first few nights of use, it is unclear how long the benefits last with continuous nightly usage. In our Dalmane-midazolam study, the benefits were gone in less than seven days as compared to placebo,[49] and in the triazolam-flurazepam study, the benefits were gone after three weeks as compared to placebo.[50] Unfortunately, for many years, almost all laboratory studies have used placebo baseline recordings as the control, without counterbalancing the order of placebo and hypnotic. The studies where hypnotic and placebo are given in parallel (at the same time to randomly-assigned volunteers) suggest that participation in laboratory experiments (and spontaneous recovery) led to improvements in sleep. After two to four weeks, the improvement seen in a drug-treated group as compared to baseline may be due to the time-related placebo-effect improvement rather than due to any real drug benefit. Manufacturers’ advertising often deliberately confuses this point.
Even with tiny increases in sleep that they provide for a few days, hypnotics do not improve an insomnia patient’s daytime function. More often, the pills make daytime function measurably worse. Patients often seek improved function, but they usually do not receive it. Further, although we hear colleagues mention that perhaps a patient will be healthier if the patient sleeps better, our research found that patients taking sleeping pills were more likely to develop new medical disorders than matched control patients who avoided sleeping pills. I have located no reliable evidence that any sleeping pill improves general health, but there is much evidence of serious harm to physical and mental health.
Endnotes for Chapter 4
47. Buscemi, Nina et al. The Efficacy and Safety of Drug Treatments
for Chronic Insomnia in Adults: A Meta-analysis of RCTs. J.Gen. Intern Med.
2007; 22(9)
: 1335-1350. [return]
48. Ramelteon: application withdrawn. Ramelteon in insomnia: withdrawal of marketing application in patients’ best interests. Prescrire Int. 2009 Jun;18(101):114. [return]
49. Kripke, DF, Hauri, P, Ancoli-Israel, S, and Roth, T. Sleep evaluation in chronic insomniacs during 14-day use of flurazepam and midazolam. J.Clin.Psychopharmacol. 10(Supplement 4):32S-43S. 1990. [return]
50. Mitler, MM et al. Comparative hypnotic effects of flurazepam, triazolam, and placebo: a long-term simultaneous nighttime and daytime study. J.Clin.Psychopharmacol. 1984;4:2-15. [return]
Table of Contents