For adults older than 65, falls are a leading cause of injury — and they occur often. Approximately one-third of all older adults fall in a year. As prevalent and debilitating as falls are, the risk for falls may be up to two to eight times higher for those with dementia (Pharmacotherapy 2019;39:530–543).
Medications that affect the brain or “psychoactive medications” have been shown to increase the risk of falls, especially when multiple medications that affect the central nervous system (CNS) are prescribed and when higher doses are used (J Clin Pharmacol 2012;52:947–955). However, although evidence exists that medications such as antidepressants, benzodiazepines, and antipsychotics may contribute to falls — and many of these medications are listed in the tools that identify potentially inappropriate medications, such as the Beers Criteria — at the individual patient level, some people do require high-risk, CNS-active medications to achieve quality of life or health-related goals.
Which Psychoactive Medications Are Safest?
Because studies assessing falls risk with medications that affect the CNS provide inconclusive evidence about which medications are safest, clinicians may rely on judgment and select less sedating medications at low doses to treat older adults with a history of falling. Although this type of rationale is sensible, given the limitations in the medical evidence, the largest recent systematic review and meta-analysis on falls risk, conducted in 2018 by Lotta Seppala, MSc, and colleagues of the eEUGMS Task Force and the Finish Group on Fall-Risk-Increasing Drugs, may help to assess the relative safety of common psychoactive medications (J Am Med Dir Assoc 2018;19:371.e11–371.e17).
In their analysis of 248 studies, the researchers noted that the use of selective serotonin reuptake inhibitors (SSRIs) is associated with the highest increase in falls risk — about a twofold increase in risk. This is close to the risk with long-acting benzodiazepines at 1.8 times higher. Other antidepressants, such as tricyclic antidepressants and antipsychotics, had between a 1.4 and 1.5 times increase in risk. Short-acting benzodiazepines had a 1.27 times increase in falls risk.
Ms. Seppala and colleagues noted that SSRIs are thought to be less sedating than tricyclics and may be considered safer, but they emphasize that the elevated risk of SSRIs and falls is consistent across several studies. This effect may be due to prescription bias: prescribers may turn to SSRIs when an already frail patient with a high risk of falls needs an antidepressant.
The researchers used pools data from individual studies to show a consistent increase in falls risk associated with CNS-active medications. However, this meta-analysis approach leaves gaps when it comes to clinical practice. How many CNS medications and what doses ultimately pose the highest risks for falls?
Evidence-Supported Approach to CNS Medications
To address the clinical questions, another research group, headed by Joseph Hanlon, PharmD, and colleagues at the University of Pittsburgh, studied serious falls and medication use along with dose data for specific drugs in a post-acute and long-term care setting. Dr. Hanlon’s group used Medicare claims data to identify which medications at which doses were associated with severe falls as identified by insurance claims codes (J Am Geriatr Soc 2017;65:1183–1189).
For over 1,800 older adults in long term care, they found that about 65% were taking a CNS medication. However, the falls risk was 1.83 times greater when patients were taking three or more CNS-active medications at daily doses consistent with a geriatric pharmacotherapy handbook. Interestingly, their analysis showed no statistically detectable difference in taking more than zero or less than three CNS-active medications.
Notably, their finding that three or more CNS medications, including opioids, increases the risk of falls is in agreement with the most recent version of the Beers list, the trusted guide developed to give evidence-based support to clinicians assessing medication safety (J Am Geriatr Soc 2019;67:674–694). This threshold of three medications enables providers to address the clinically appropriate need for a psychoactive medication while heeding calls for vigilance and frequent monitoring for efficacy and side effects as well as maximizing nonpharmacological interventions.
For CNS-active medications that are no longer effective, evidence-based algorithms are increasingly available for safely decreasing and discontinuing these medications. These include the clinician tools available at the US Deprescribing Network (deprescribingresearch.org
) and at the PIMSPlus website (pimsplus.org
Ongoing efforts are needed to enable tools embedded within our workflow and encourage collaboration among team members to promote safe and judicious use of medications that affect or treat neurologic and/or psychiatric conditions.
Dr. Lee is the Geriatric Pharmacotherapy Fellow at the Peter Lamy Center on Drug Therapy and Aging at University of Maryland Baltimore. Additionally, he works as a clinical pharmacist at the MedStar Center for Successful Aging.