Brandel France de Bravo, MPH, Cancer Prevention and Treatment Fund
When we hear “sleeping pills,” most of us think of prescription drugs such as Ambien (generic name zolpidem), Restoril (temazepam), and Lunesta (eszopiclone). While prescription sleep medications are big business — consumers spent $2 billion on them in 2010 — many people with trouble sleeping turn to over-the-counter antihistamines such as Tylenol PM and Benadryl, which are also considered hypnotic drugs. But the use of these hypnotic drugs may take a nosedive in light of the findings of a study published in February 2012 in the prestigious British Medical Journal. Led by researchers at the Scripps Clinic Viterbi Family Sleep Center in California, the study shows that people who take hypnotic drugs are significantly more likely to be diagnosed with cancer or to die within the next two and a half years than people who don’t take them. Author Dr. Daniel Kripke estimates that these popular sleep medications could have caused 320,000 to 507,000 deaths in 2010.
The researchers looked at 10,529 primary care patients who were prescribed hypnotic drugs between 2002 and 2007 and compared the health of each of them to at least two very similar patients without such prescriptions who were the same sex, ethnicity, marital status, smoking status, and had similar health conditions, alcohol use and body mass index (a combination of height and weight). All patients, who were followed for 2.5 years on average, were from a Pennsylvania clinic that serves a mainly low-income population.
Patients who were prescribed sleeping pills were at least three to five times more likely to have died during the study than were the patients not prescribed sleeping pills. Even the patients who were prescribed fewer than 18 pills per year were at higher risk of dying: 3.6 times higher. Patients who were prescribed more than 132 pills a year were more than five times as likely to die.
The researchers were careful to exclude from the study patients who were diagnosed with cancer before the study or very early in the study. In spite of this precaution, they found that patients who were prescribed more than 18 pills a year had an increased cancer risk, with the heavy users (over 132 pills prescribed per year) having a 35% greater risk than those with fewer pills prescribed. Among those with prescriptions for sleeping pills, the increased risk of their developing lymphoma, lung cancer, colon and prostate cancer was greater than the risk from being a current smoker.
Before this study, there were at least 18 other studies showing an increased risk of death for people taking sleeping pills, and several also showed an increased risk of cancer. However, this study is especially well-designed and the only one that includes the newer, short-acting class of popular sleeping pills known as nonbenzodiazepines . These were generally believed to be safer than previous generations of sleeping pills because they wear off more quickly. In fact, until this study, the scariest side effect was seemingly inexplicable weight gain due to night time raids on the refrigerator while sleep walking.
Among study participants, the most commonly prescribed sleeping pill was zolpidem (a nonbenzodiazepine marketed as Ambien, Edluar, or Zolpimist), followed by temazepam (a benzodiazepine). However, prescriptions for the use of any hypnotic drug as a sleep aid was associated with a significant increase in the risk of death, including eszopiclone (”Lunesta”), zaleplon (”Sonata”), and barbiturates, as well as prescriptions for diphenhydramine, an antihistamine used in many over-the-counter sleep aids. The average age of patients was 54, but the study found harm associated with sleeping pill use in every age group.
All the sleeping pills showed a similar increased risk of death except Lunesta , which showed a more than 500% increased risk compared to any of the other sleeping pills. However, Lunesta was a relatively new drug at the time of the study, and relatively few people took it. For that reason, it is not possible to say whether the risk of Lunesta is really higher.
One shortcoming of the study is that getting a prescription for a sleeping pill is not the same as taking sleeping pills. It is possible that some of the people with prescriptions, especially for small numbers of pills, never took any of them. It is also possible that people who did not have prescriptions for sleeping pills took over-the-counter antihistamines to help them fall asleep, instead of the prescription version of the same pills. However, those shortcomings would tend to underestimate the risk of sleeping pills, rather than over-estimate the risks.
What could possibly explain these increased risks? Are people who are prescribed sleeping pills more anxious or stressed out? There is evidence that they are more likely to have car accidents or to fall down, probably because of the residual effects of the drugs during the day. Other studies show an increase in infections among people taking sleeping pills, and that can also increase the risk of cancer and death from other causes. These other studies all suggest that sleeping pills really do increase the risk of dying and there are no logical explanations to explain away the substantial increased risks found in this study, especially the increased risk of cancer.
While the researchers can’t say for sure that the sleeping pills caused death or cancer, many people who used to take these medications should think about these new research findings and consider other, safer ways to fall asleep. The sleep specialists who conducted the research suggest that since hypnotics have limited benefits, old-fashioned sleep aids like warm milk, as well as cognitive-behavioral approaches that can be taught and used for the rest of your life, would be excellent alternatives. If you decide to toss your sleeping pills, be sure to see our article Drugs in the Drinking Water for tips on safe medicine disposal.
- IMS data cited in The Wall Street Journal. Dawn of A New Sleep Drug. July 19, 2011.
- Kripke DF, Langer RD, Kline LE. Hypnotics’ association with mortality or cancer: a matched cohort study. British Medical Journal. Open 2012;2:e000850 doi:10.1136/bmjopen-2012-000850