Case Challenge

'I Just Can't Sleep': The Hazards of Treating Insomnia in the Elderly

Douglas S. Paauw, MD


September 26, 2016

The Many Downsides of Treating Insomnia

Dementia. Zolpidem has received a lot of negative publicity in the past few years. The effects of zolpidem use on the development of dementia or Alzheimer disease was evaluated in a retrospective, population-based, nested, case-control study involving 8406 patients with dementia and 16,812 control subjects enrolled in the Taiwan national health insurance database from 2006 to 2010.[1] The investigators reported that zolpidem use was associated with an adjusted odds ratio (OR) of 1.33 (95% confidence interval [CI], 1.24-1.41) for the development of non-Alzheimer dementia after controlling for a wide range of variables, including age, sex, coronary artery disease, stroke, diabetes, depression, and use of antihypertensives, antipsychotics, anxiolytics, antidepressants, or benzodiazepines. There was not an association with Alzheimer disease. They did report an interesting dose-response relationship; the adjusted OR for patients receiving mid-range doses (cumulative exposure doses between 170 and 819 mg/year) was 1.65 (95% CI, 1.08-2.51; P=.0199). There was no association reported in patients receiving lower or higher doses. The authors concluded that although the data did not support an association with Alzheimer disease, there was an association with non-Alzheimer dementia.

Mortality. A retrospective cohort study involving over 100,000 patients seen in primary care practices in the United Kingdom looked for a relationship between anxiolytic and hypnotic drugs and premature mortality.[2] The population included 34,727 patients aged 16 years or older who were prescribed anxiolytic or hypnotic drugs, or both, between 1998 and 2001 and matched for age, sex, and practice setting with 69,418 patients with no prescriptions for these agents. The age-adjusted hazard ratio (HR) for mortality for use of any study drug in the first year after recruitment was 3.46 (95% CI, 3.34-3.59). After adjusting for other potential confounders, the HR was slightly lower at 3.32 (95% CI, 3.19-3.45). After excluding deaths that occurred in the first year of use, the researchers calculated that use of one of these agents results in four increased deaths per 100 patients followed over 7.6 years.

However, another study, also a retrospective cohort study conducted in Taiwan, suggested that zolpidem does not increase mortality.[3] In this study, benzodiazepines were reported to have a higher HR for death (HR, 1.81; 95% CI, 1.78-1.85), but zolpidem appeared to exhibit a lower risk for mortality in adjusted models (HR, 0.73; 95% CI, 0.71-0.75).

Falls. While not conclusive, what is known is that several studies have suggested that zolpidem use may be associated with worsening balance and an increase in fall and fracture risk.[4,5,6] A well-designed retrospective cohort study provides stronger evidence about this risk. The investigators looked at inpatients who had zolpidem prescriptions.[7] This was a captive audience because they were in the hospital, so there was good documentation of receipt of the medication. Because the prescriptions for zolpidem were written as PRN, the researchers compared the fall risk in patients who actually got the drug (n=4962) compared with those who, while they did have a prescription, did not receive it (n=11,358). The fall rate among patients who were prescribed and received zolpidem (3.04%) was much higher than in those who did not receive it (0.71%; P<.001).

Conclusion. Bottom line: What we know to date suggests that evidence linking zolpidem use and non-Alzheimer dementia is pretty weak. We also do not have conclusive data suggesting that zolpidem increases mortality. We do, though, have good data on the risk for falls. Given zolpidem's widespread use, it's likely that we will all face questions from patients about potential use. It is a great opportunity to get older patients off of a medication that is higher-risk.

Editor's Recommendations


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.