Psychotropic Medications Associated With Falls in Elderly Patients

Norra MacReady

December 03, 2009

December 3, 2009 — Elderly patients who take psychotropic drugs, such as sedatives, antidepressants, and benzodiazepines, have an increased risk of falling, according to a new meta-analysis of 22 studies.

The finding of an association between sedatives and falls among elderly patients is nothing new, senior author Carlos Marra, PharmD, PhD, told Medscape Psychiatry. In fact, he was surprised that these agents are still prescribed so frequently in this population. “Even though this information has been known [for a long time], [it has] done very little to curtail the use of these drugs in the elderly and their risk of falling,” he said.

Also surprising was the finding in this analysis of no association between falling and the use of narcotics, said Dr. Marra, chairman of the Collaboration for Outcomes Research and Evaluation at the University of British Columbia, Vancouver,Canada. However, that may be attributable more to the small number of studies analyzed and the wide variability in their design than to any true lack of risk.

“I think we haven’t disproved the hypothesis that narcotics are associated with falling," he noted. "I think more work, in a prospective fashion if feasible, is needed to tease out whether or not narcotics are actually associated with falling. I think it’s premature to say that they’re not.”

Their report is published in the November 23 issue of the Archives of Internal Medicine.

Bayesian Method

Dr. Marra and his coauthors thought it was time to provide an update to a previous meta-analysis published in 1999, which included studies published between 1966 and 1996 (Leipzig RM, et al. J Am Geriatr Soc. 1999;47:30-50). That study found a small but consistent association between psychotropic drugs and falls.

In this new study, they used a method of meta-analysis known as the Bayesian method, which allows the integration of older and more recent information to arrive at an odds ratio of a particular outcome or event. “The Bayesian method allows us to ask questions like, ‘what is the probability that this drug will lead to falling?’” Dr. Marra explained.

The authors, led by John C. Woolcott, MA, identified 11,118 relevant articles published between 1996 and 2007. Of those, 22 met their search criteria and were included in the new meta-analysis. The studies included 79,081 participants older than 60 years and provided information on 9 drug classes.

To be eligible, the studies had to present original data from randomized controlled trial or cohort, case-control, or cross-sectional studies that assessed the association between medication use and falls among people aged 60 years and older, although, ultimately, no randomized controlled trials met these criteria. Using the Bayesian method, the authors then calculated the odds ratio between a given drug or drug class and the likelihood of falling.

After adjusting for confounders, sedatives and hypnotics, benzodiazepines, and antidepressants were found to be significantly related to falls in these individuals.

Table. Risk of Falls Associated With Various Drug Classes

Drug Class Adjusted Odds Ratio (95% CI)
Sedatives and hypnotics 1.47 (1.35 – 1.62)
Benzodiazepines 1.41 (1.20 – 1.71)
Antidepressants 1.36 (1.13 – 1.76)

CI = confidence interval

Although neuroleptics and antipsychotics together were significantly associated with falls in univariate analysis, the association was no longer significant in adjusted analyses. Similarly, diuretics and beta-blockers were not associated with falls in adjusted analyses, but antihypertensives were associated with a higher risk of falls in all analyses.

Stratification of studies by population and study type showed only small differences in odds ratios, the authors note. However, in studies considered to have "good" medication and falls ascertainment, sedatives and hypnotics, neuroleptics and antipsychotics, antidepressants, benzodiazepines, and nonsteroidal anti-inflammatory drugs were associated with an increased falling risk.

"Given the divergent results shown within specific medication classes, the results of our meta-analysis reiterate the need for caution when prescribing these medications to seniors," the authors conclude. "It is hoped that future research in this area can be completed with larger sample sizes in both community and long-term care facility settings and thus improve the quality of information about fall risks that is available to physicians and pharmacists when they are deciding which types of pharmacotherapy to provide."

Sedatives Overused

“This was a well-done study — they really tried to isolate the effects of medications above and beyond the other known factors,” said Mary E. Tinetti, MD, professor of medicine and public health at the Yale School of Medicine, New Haven, Connecticut. Dr. Tinetti, who was not involved in this research, pointed out that “sedatives are very much overused across the age span and particularly in the elderly.”

Often, that is because the patients themselves request the drugs to help them sleep, then become dependent on them, she explained.

Are we doing more benefit than harm in prescribing these drugs?

On the other hand, “antidepressants and antihypertensives are important drugs that treat serious disorders,” said Dr. Tinetti, who was a coauthor of the previously mentioned study by Leipzig and colleagues. In her opinion, the most important take-home message of this meta-analysis is “not whether the patient needs the drugs but about the risk-benefit ratio. Are we doing more benefit than harm in prescribing these drugs?”

Elderly people with dementia frequently are prescribed antipsychotic medications, said John Newcomer, MD, Gregory B. Couch professor of psychiatry, psychology, and medicine at the Washington University School of Medicine, St. Louis, Missouri, and medical director of the school’s Center for Clinical Studies.

“There are no drugs labeled for that indication, [but] if you’ve ever worked with this population, families of patients with dementia and people who work in that setting are well aware that this is a very knotty problem," he told Medscape Psychiatry. "You have patients who can potentially harm themselves or others or who can significantly interfere with their ability to get good care if you don’t somehow address those behavioral problems.”

Clearly there's this critical and very tricky risk-benefit consideration here.

Dr. Newcomer, who was also not involved in this meta-analysis, agreed with Dr. Tinetti that perhaps its most important take-home message is the assessment of risk-benefit. "This is a very high-risk population ... and clearly there’s this critical and very tricky risk-benefit consideration here: we’re always messaging about the downside, but I think people taking care of this population and families who are trying to manage people with dementia at home are acutely aware of the need for some kind of treatment.

"The issue is how to achieve the desired effect with behavioral interventions and perhaps some pharmacologic interventions without taking unnecessary risks," Dr. Newcomer added. "This paper is yet another reminder that our treatments are not risk free.”

The study was funded in part by the Canadian Institutes of Health Research, the Michael Smith Foundation for Health Services Research, and the Government of Canada Research Chair in Pharmaceutical Outcomes. The authors have disclosed no relevant financial relationships.

Arch Intern Med. 2009;169:1952-1960.


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.
Post as: