Benzodiazepines and risk of Alzheimer’s disease

BMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g5312 (Published 09 September 2014) Cite this as: BMJ 2014;349:g5312
  1. Kristine Yaffe, Roy and Marie Scola endowed chair and professor of psychiatry12,
  2. Malaz Boustani, Richard M Fairbanks professor in aging research34
  1. 1University of California at San Francisco, San Francisco, CA, USA
  2. 2San Francisco Veterans Affairs Medical Center, San Francisco, CA, USA
  3. 3Indiana University Center for Aging Research, Indianapolis, IN, USA
  4. 4Regenstrief Institute, Indianapolis, IN, USA
  1. Correspondence to: K Yaffe, 4150 Clement St, Box 181G, San Francisco, CA 94121, USA kristine.yaffe{at}ucsf.edu

Prescribers and patients need a proper surveillance system for cognitive side effects

A growing number of observational studies have shown the critical role of potentially inappropriate medications for increasing the risk of cognitive impairment. In a linked paper, Billioti de Gage and colleagues (doi:10.1136/bmj.g5205) extend the pharmacoepidemiological research on the adverse cognitive effects of benzodiazepines with an investigation of their link with Alzheimer’s disease.1 Their results suggest that long term exposure to benzodiazepines might be a modifiable risk factor for this condition.

The authors conducted a nested case-control study of about 2000 older members of a public drug plan in the province of Quebec, Canada. They observed a cumulative dose-effect association between exposure to benzodiazepines (at least 90 days) and risk of developing Alzheimer’s disease and found that exposure lasting more than 180 days was associated with a nearly twofold increase in risk. In further analyses, they showed that longer acting benzodiazepines were associated with greater risk of developing Alzheimer’s disease compared with shorter acting benzodiazepines, adding support for a causal association.

The interpretation of these findings is strengthened by rigorous methods. For example, the authors minimized selection bias by randomly selecting both the case and control subjects from a well defined and generalizable cohort of older adults living in the community. They also limited recall, temporal, and indication biases associated with drug dispensing by using exposure windows of at least five years before the index diagnostic date for cases and their controls matched on age, sex, and follow-up. Findings were adjusted for many potential confounders with multivariate conditional logistic regression analysis. Confounders included underlying medical conditions such as depression and anxiety that commonly trigger use of benzodiazepines and are also considered risk factors for Alzheimer’s disease. Finally the study’s sponsor had no influence on the design and the conduct of the study.

It is not surprising that benzodiazepines are associated with adverse cognitive effects. In 2012, Tannenbaum and colleagues published a systematic review that found an association between benzodiazepines and cognitive impairment in 38 out of 39 studies, but the impairment was presumed to be transient and reversible.2 The findings of the present study challenge such a reversible cognitive effect and suggest that the adverse cognitive outcomes of benzodiazepines might include neurodegenerative disease. Others have suggested that benzodiazepines and other potentially inappropriate drugs are commonly prescribed for older adults and are associated with worse cognitive outcomes over a long follow-up.3 4 These studies, however, also remind us of the complex relation between dementia (including Alzheimer’s) and use of benzodiazepines. Older adults who received a diagnosis of dementia were more likely to be taking benzodiazepines and other potentially inappropriate medications five to six years before, raising the possibility of reverse causation as benzodiazepines can be used to treat prodromal symptoms of dementia including anxiety and sleep disturbances.

As life expectancy increases, the developed world is facing a rapid rise in the number of older adults with multiple chronic conditions.5 Managing multiple conditions requires, on average, treatment with at least five drugs,6 and one of the most critical negative consequences of polypharmacy is the development of adverse drug effects.7 Some side effects, such as renal injury and anaemia, are well documented, readily identified, and easily monitored with objective laboratory tests. Other side effects, including cognitive side effects, are much harder to recognize because there is no routine laboratory test or standardized post-marketing surveillance system.7

In 2012, the American Geriatrics Society updated its list of inappropriate drugs for older adults to include three classes of drugs with adverse cognitive effects: anticholinergics, H2 antihistamines, and benzodiazepines.7 Despite evidence of such cognitive side effects, a review of published literature indicates that up to half of older adults continue to use these.8 Currently, there is no standardized approach to help to identify and monitor the cognitive side effects of drug treatments. As a result, potential long term consequences of drugs such as benzodiazepines remain hidden and contribute to a growing burden of cognitive impairment among older adults. Identification of drugs with adverse cognitive effects requires dedicated monitoring of cognitive function, which is often absent from routine hospital and ambulatory clinical care.9

To fill this gap, we support the development of a structured reproducible approach to the identification and accurate monitoring of the adverse cognitive effects of all drug treatments used by older adults with multiple chronic conditions, particularly by those at risk of Alzheimer’s disease. Ideally, this would establish a global surveillance system for adverse cognitive effects that uses readily available clinical factors and drug exposures from a growing number of electronic medical record systems. Better surveillance for cognitive side effects could improve therapeutic decisions among doctors treating high risk older adults with multiple chronic conditions and hopefully, eventually, help to reduce the burden of cognitive impairment worldwide.


Cite this as: BMJ 2014;349:g5312


  • Research doi:10.1136/bmj.g5205
  • Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: KY has received fees as a board member for Takeda, an NIA sponsored study, and the Beeson Scientific Advisory Board and as a consultant for Novartis and Pfizer. KY’s institution receives funding from NIA, NIMH, NIDDK, the Alzheimer’s Association, Bright Focus Foundation, the California Department of Public Health, and the Department of Defense and Veterans Administration for studies in which she is an investigator.

  • Provenance and peer review: Commissioned; not peer reviewed.


View Abstract

Sign in

Log in through your institution

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial