The use of hypnotic sleep aids was associated with a three- to five-fold higher mortality risk compared with the risk for nonusers, even when the prescription was for a small number of pills, investigators reported.

A prescription for 0.4 to 18 doses per year was associated with a mortality hazard ratio of 3.60 compared with patients who had no prescriptions for hypnotics.

The hazard jumped to 5.32 for patients prescribed more than 132 doses a year, investigators reported online in BMJ Open.

“Rough order-of-magnitude estimates ... suggest that in 2010, hypnotics may have been associated with 320,000 to 507,000 excess deaths in the U.S. alone,” Daniel F. Kripke, MD, of the Scripps Clinic in La Jolla, Calif., and co-authors wrote. “From this nonrandomized study, we cannot be certain what portion of the mortality associated with hypnotics may have been attributable to these drugs, but the consistency of our estimates across a spectrum of health and disease suggests that the mortality effect of hypnotics was substantial.”

Patients who used hypnotics most often also had an increased risk of cancer, with an overall cancer increase of 35% among those prescribed high doses. More than 30 years ago, investigators in an American Cancer Society-supported study showed that both cigarette smoking and hypnotic use were associated with excess mortality (Arch Gen Psychiatry 1979; 36: 103-116). But the link to hypnotics was largely discounted because the study was not designed primarily to examine these drugs, Kripke and colleagues wrote. Subsequently, at least two dozen studies examined the mortality risk associated with hypnotic use, and two-thirds of the studies demonstrated significant (P<0.05) associations. Lack of uniformity across the studies precluded a meta-analysis, but 22 of the reports showed a mortality hazard ratio that exceeded 1.0, the authors continued. Previous studies had several notable limitations, including limited information on the specific types of drugs, confounding with tranquilizers, lack of monitoring of the quantities of drugs provided to patients, and limited data on newer short-acting hypnotics, such as zolpidem, zaleplon, and eszopiclone (Lunesta). To address some of the limitations of previous work, Kripke and colleagues performed a matched-cohort study based on longitudinal data from a large U.S. health system. A query of the database identified 10,531 adult patients who had at least one prescription for a hypnotic drug from Jan. 1, 2002 to Sept. 30, 2006. Using the same database, the authors matched the hypnotic-user group with 23,674 patients who did not have a prescription for a hypnotic during the period studied. Three-fourths of patients prescribed a hypnotic had an explicitly stated sleep-related indication in their records. Women (mean age 54) accounted for 63.9% of hypnotic users. Hypnotic users and the control group had been followed for about 2.5 years. The users had a mean morbidity score of 1.53. Zolpidem was the most commonly used hypnotic (4,338), followed by temazepam (2,076). Overall, 6.1% of hypnotic users died during observation, compared with 1.2% of the nonusers. Hypnotic use was associated with a significantly increased mortality risk (P<0.001). The magnitude of the hazard ratio increased with the number of pills prescribed per year (P<0.001 for all comparisons versus nonusers): HR 3.60 for 0.4 to <18 pills HR 4.43 for 18 to 132 pills HR 5.32 for >132 pills Separate analyses of the two most commonly used hypnotics showed a similar increase in the mortality hazard. For zolpidem the hazard increased from 3.93 for patients who took 5 mg/year to 130 mg/year, to 5.69 for patients who had prescriptions for >800 mg/year (P<0.001). Patients with temazepam prescriptions totaling 1 mg/year to 240 mg/year had a mortality hazard of 3.71, increasing to 6.56 for >1,640 mg/year (P<0.001). Overall, only patients whose hypnotic use fell into the top two categories (18 to 132 pills and >132 pills) had an increased cancer risk (HR 1.20, P=0.022; HR 1.35, P<0.001). For individual drugs, only patients in the top category of zolpidem use had an increased cancer risk (HR 1.28, P=0.023), whereas the two top categories of temazepam use were associated with an increased mortality hazard (HR 1.44, P=0.024; HR 1.99, P<0.001). The authors acknowledged limitations to this research, most notably that residual confounding could not be fully excluded "due to possible biases affecting which patients were prescribed hypnotics and due to possible imbalances in surveillance." They also pointed out that cohort studies may demonstrate an association but do not necessarily imply causality. However, "the preferable randomized controlled trial method for assessing hypnotic risks may be impractical due to ethical and funding limitations," they said. "The meager benefits of hypnotics, as critically reviewed by groups without financial interest, would not justify substantial risks," the authors wrote. "A consensus is developing that cognitive behavioral therapy of chronic insomnia may be more successful than hypnotics. "Against meager benefits, it is prudent to weigh the evidence of mortality risks from the current study and 24 previous reports, in order to reconsider whether even short-term use of hypnotics, as given qualified approval in National Institute for Clinical Excellence guidance, is sufficiently safe," they added.