Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Psychotropic drugs increase the risk for falls in older persons

ACP J Club. 1999 July-Aug;131:24. doi:10.7326/ACPJC-1999-131-1-024

Source Citation

Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: a systematic review and meta-analysis: I. Psychotropic drugs. J Am Geriatr Soc. 1999 Jan;47:30-9.



What is the association between psychotropic drugs and falls in older persons?

Data sources

Published English-language studies were identified by using MEDLINE (1966 to March 1996) with the search terms accidents, accidental falls, aged, or age factors; reviewing bibliographies of papers; and contacting experts.

Study selection

Studies were selected if they assessed the association between use of psychotropic drugs, including antidepressants, neuroleptics, sedatives or hypnotics, or long- or short-acting benzodiazepines, and falls in persons ≥ 60 years of age.

Data extraction

Data were extracted on study and patient characteristics; fall assessment and definition; types of drugs; number of falls associated with benzodiazepines, antidepressants, neuroleptics, hypnotics, or sedatives; and the temporal relation between drug assessment, fall assessment, and the index fall.

Main results

1043 studies were identified, and 40 met the inclusion criteria. 70% of studies defined a fall as a nonsyncopal event that was not attributed to a violent blow, loss of consciousness, stroke, or epileptic seizure. 7 studies of community-living older persons showed a 36% annual incidence of falls and a 34% incidence of psychotropic drug use; 2 studies of persons in long-term care facilities showed a 58% 6- to 12-month incidence of falls and a 69% incidence of psychotropic drug use. 37 studies had follow-up ≥ 6 months. Use of fixed-effects meta-analysis techniques showed that both individual and combinations of psychotropic drugs increased the risk for ≥ 1 falls (Table). Neuroleptic drugs increased the risk for falls in community-living persons (odds ratio [OR] 1.66, 95% CI 1.38 to 2.00) but decreased the risk in hospitalized persons (OR 0.41, CI 0.21 to 0.82). The risk for falls did not differ between short- and long-acting benzodiazepines or between persons with ≥ 1 falls and those with ≥ 2 falls.


Psychotropic drugs are associated with a small increase in risk for falls in older persons.

Source of funding: Not stated.

For correspondence: Dr. R.M. Leipzig, Department of Geriatrics and Adult Development, Mount Sinai School of Medicine, 1 Gustave Levy Place, Box 1070, New York, NY 10029-6574, USA. FAX 212-860-9737.

Table. Weighted odds ratio (OR) for the association between various psychotropic drug classes vs placebo and ≥ 1 falls in older persons

Drug class Number of studies Number of cohort studies Weighted OR (95% CI)
Psychotropic drugs 19 11 1.73 (1.52 to 1.97)
Antidepressants 27 11 1.66 (1.41 to 1.95)
Neuroleptics 22 10 1.50 (1.25 to 1.79)
Tricyclics 12 6 1.51 (1.14 to 2.00)
Sedatives or hypnotics 22 9 1.54 (1.40 to 1.70)
Benzodiazepines 13 8 1.48 (1.23 to 1.77)


The relation between psychotropic drug use and falls has been widely studied (1). Leipzig and colleagues systematically reviewed 40 studies and found that older persons who use psychotropic drugs have about a 50% increased incidence of falls.

It is biologically plausible that psychotropic drugs could cause falls. Most of these agents produce sedation and increase postural swaying in older persons (2). In addition, studies consistently show that older persons who use these drugs are more likely to fall than those who do not. The relative risk for falls associated with psychotropic drug use usually approaches 1.5 to 2.0. A dose-response relation has been shown in several studies: Persons who use higher doses or multiple psychotropic drugs have the highest risk for falls, as shown in this systematic review.

2 lines of evidence make this association more difficult to define. First, the association lacks specificity. Despite wide variation in side effects across the spectrum of psychotropic drugs, the risk for falls is similar across all classes of agents. Second, no observational study to date has provided adequate control for "confounding by indication." Persons who are prescribed psychotropic drugs are more likely to fall because the symptoms treated by these agents place them at higher risk for falls. Studies in other areas have used a propensity score to quantify this type of confounding (3).

Given these uncertainties, psychotropic drugs should not be withheld in settings where they have shown effectiveness solely because they may increase risk for falls. Older adults with depression should be treated with antidepressants (I prefer nortriptyline as a first-line agent), and older adults with chronic psychosis should be managed with antipsychotic drugs (I prefer haloperidol as a first-line agent). The decision to use anxiolytics, hypnotics, and antipsychotics to manage symptoms of agitation, anxiety, and poor sleep in older adults must be evaluated case by case, and patients who use these drugs should be cautioned about the risk for falls.

Ronald I. Shorr, MD, MS
Methodist HealthcareMemphis, Tennessee, USA


1. Ballinger BR, Ramsay AC. Accidents and drug treatment in a psychiatric hospital. Br J Psychiatry. 1975;126:462-3.

2. Robin DW, Hasan SS, Edeki T, et al. Increased baseline sway contributes to increased losses of balance in older people following triazolam. J Am Geriatr Soc. 1996;44:300-4.

3. Rubin DB. Estimating causal effects from large data sets using propensity scores. Ann Intern Med. 1997;127:757-63.