Do Sleeping Pills Kill?

I couldn't resist.  I was drawn, like moths to a light, to the same hysteria-causing headlines surrounding sleeping pills last week.  Actually, the Wall Street Journal and CBS News did a better job by pointing out that a recent observational study published online in BMJ Open only links use of sleeping pills to an increase risk of death rather than demonstrating any cause & effect.  What really surprised me was the commentary among the physician readership.  Some were outraged that something like this could pass for research and be published by a reputable journal.  Others were happy that some light was finally being shed on this class of medications.  Still others were concerned about how the public might react to 10 second sound bites.

So what's the fuss all about?  The authors, one of whom may (or may not) have a bone to pick, evaluated electronic medical records from Geisinger Health System, a rural integrated health system serving 2.5 million lower-socioeconomic, less than high school educated people in Pennsylvannia.  Medical records for close to a quarter million individual patients were abstracted if they had been seen between January 2002 and September 2006.  10,531 of these case patients received at least one prescription for a sedative-hypnotic during this 4+yr period of time.  They were then matched to 23,674 controls who never received a sedative-hypnotic during this same period of time.  The authors took into account the usual suspects for finding control patients: sex (60+% female), age +/- 5yrs (avg 54yo), ethnicity (>94% white), marital status (>50% married), smoking status (>40% never), alcohol use & body mass index.  They even went so far as to look at whether comorbid conditions were present prior to initial prescription or diagnosed later on.

Not surprisingly, zolpidem (Ambien & Ambien CR) was the most widely prescribed sedative-hypnotic, followed by temazepam (sold as Restoril), a benzodiazepine.  What I did find surprising was the increased mortality associated w/sedative-hypnotic use regardless of comorbidities and confounders.  And the more one used these drugs, the greater one's risk of death.  In fact, use of just 1-18 pills in a year was associated with 3x risk of death compared to non-users.  Those in the highest tertile of use had 5x greater risk of death compared to non-useers.  To top it off, sedative-hypnotic use was also linked to greater risk of cancer!

Now, smarter people than I have attempted to analyze the data and make sense of it (or tear it apart).  What's fascinating is that this isn't the first study making this link between sedative-hypnotic use and increased mortality.  Apparently there have been at least 2 dozen other studies.  I should also point out that for what ever reason, one of the authors apparently has made it his quixotic quest to inform the world of the "dark side of sleeping pills".  Regardless, while correlation does not prove causation, I think it would be wise to consider the risks, benefits & alternatives to sedative-hypnotics when it comes to managing insomnia.  

Follow @alvinblin
Tweet