Open Access Open Badges Research article

Psychotropic drug initiation or increased dosage and the acute risk of falls: a prospective cohort study of nursing home residents

Murray A Echt12, Elizabeth J Samelson13, Marian T Hannan13, Alyssa B Dufour1 and Sarah D Berry13*

Author Affiliations

1 Institute for Aging Research, Hebrew SeniorLife, 1200 Centre Street, Boston, MA, 02131, USA

2 SUNY Downstate College of Medicine, 450 Clarkson Avenue, Brooklyn, NY, 11203, USA

3 Harvard Medical School, Boston, MA, USA

For all author emails, please log on.

BMC Geriatrics 2013, 13:19  doi:10.1186/1471-2318-13-19

Published: 22 February 2013



Previous studies suggest that psychotropic drug changes may signal an acute period of time whereby a person is highly vulnerable to fall. It is unknown whether certain classes of psychotropic agents are less safe with respect to the acute risk of falls. Our purpose was to compare fall rates in the 7 days following a change of an antidepressant, antipsychotic, or benzodiazepine. We also identified specific times when residents are at high risk for falls with respect to a psychotropic drug change.


Residents in our one-year study included 851 long term care residents from two nursing home facilities in Boston, MA, U.S.A. (May 2010 - May 2011). Drug changes (i.e., new prescriptions or increased dose of a previously used drug) were ascertained using the computerized provider order entry system, whereas falls were ascertained by incident reports. Negative binomial regression was used to compare the rate of falls following a drug change between medication classes. Further, we calculated the rate of falls for each of the 7 days before and 7 days after a psychotropic drug change.


Forty-eight percent of residents were prescribed a new prescription or increased dose of a psychotropic drug during the study. The rate of falls was similar in the 7 days following a change to a SSRI versus non-SSRI antidepressant (11.9 versus 14.4 falls/1,000 person years; p = 0.58), a typical versus an atypical antipsychotic (25.4 versus 17.1 falls/1,000 person years; p = 0.10), or a short versus long acting benzodiazepine (15.2 versus 13.9 falls/1,000 person years; p = 0.23). Fall risk was highest on day 4 before the drug change (19.0 falls/1,000 person days), on the day of the drug change through 2 days after the drug change (17.6-20.3 falls/1,000 person days), and 5-6 days after the drug change (17.6-19.0 falls/1,000 person days).


In the nursing home, risk of falls was similar following a psychotropic drug change of any class. We observed higher fall risk in the days before, but mostly after the drug change. We recommend that nursing home residents be closely monitored following a psychotropic drug change in an effort to reduce falls.

Falls; Nursing home; Psychotropic drug change