On February 27, 2012 the online medical journal BMJ Open published article titled: “Hypnotics association with mortality or cancer: a matched cohort study”. This study concluded that "receiving hypnotic prescriptions was associated with greater than threefold increased hazards of death even when prescribed <18 pills/year”. Online news outlets and social media sites have erupted with commentary regarding this study. According to BBC News, “Report suggests sleeping pills linked to early death". WABC in New York ran an article titled "New sleeping pill dangers". KHOU.com in Houston Texas reported that a “New study findings big risk for sleeping pills". Social media sites are filled with people commenting on their worries about sleeping medication. Advertisements for alternatives to sleeping pills are populating Facebook, Twitter and other social media outlets. The big question on the minds of many sleeping pill users is:
Are sleeping pills going to kill me?
The BMJ Open study does not answer that question.
Every medical study must be carefully evaluated before it is applied to medical practice or is used to change recommendations in the lives of individual patients. Yesterday's article by doctors Kripke, Langer, and Kline published by the British Medical Journal provokes important discussion but the study is significantly limited. When a medical study is reviewed, the structure of the study, the patient population, and confounding factors need to be considered. The BMJ Open study had a significant number of patients (34,205). However, these people were not individuals seen by the researchers. This was a retrospective chart review study. This study did not compare the efficacy and safety of sleeping medication. This study compared patients who were taking sleeping medication to patients who were not treated with sleeping medication. Although the groups were matched for some confounding factors (age, gender, smoking, body mass index, ethnicity, marital status, alcohol use and prior cancer) other important factors were not included. The study did not match case and control patients for other medical illnesses or psychiatric problems such as depression which could have accounted for a significant portion of the increased death rate.
Most significantly, this study did not control for insomnia itself. Previous studies have shown that insomnia itself is associated with an increased risk of death. The Wisconsin Sleep Cohort Study is an ongoing population based study which examines the natural history of sleep and sleep disorders on significant health outcomes. The Wisconsin Sleep Cohort Study has reported that patients with persisting symptoms of insomnia were more than twice is likely to die compared to people who did not have insomnia. Other studies have indicated that people suffering from depression are more likely to die than people who do not have depression. It is also well known that people with chronic medical illnesses are more likely to have disrupted sleep. Furthermore, it is important to note that people who have medical and psychiatric disease are more likely to have insomnia. People with obstructive sleep apnea often present with insomnia and obstructive sleep apnea is associated with a two-fold increase in cardiovascular disease events. Medical and psychiatric diseases themselves are associated with increased risk of death.
Since people with medical and psychiatric disorders are more likely to have insomnia as a consequence of their conditions, and people with insomnia are more likely to take medication to promote sleep, it is reasonable to conclude that people who take sleeping medication are more likely to have medical and psychiatric disease. Concluding that people who take sleeping medications are more likely to die is similar to concluding that people who drive to work are more likely to have a motor vehicle accident than people who walk to work. We cannot implicate the medications themselves based on this study. The conditions resulting in the use of the medications need to be considered.
The important conclusion one can draw from the BMJ Open study is that it is that prescribers of sleeping medications need to identify the underlying causes of sleep disturbances and treat them appropriately. Medical and psychiatric disease which resulted in insomnia should be managed appropriately. Disturbed sleep should be investigated and appropriate sleep hygiene and sleep behaviors implemented. Although it may be true that people who take sleeping medication are more likely to die than people who do not, this study does not prove that the medication is the cause of death. This study chart review suggests that the people with sleep disturbances treated with sleeping pills are more likely to have conditions which increase the risk of death. The practice of prescribing sleep medication should be coupled with investigation into the cause of sleep disturbances.
Darius Zoroufy, M.D.
