EFFECTS OF CENTRAL-NERVOUS-SYSTEM POLYPHARMACY ON FALLS LIABILITY IN COMMUNITY-DWELLING ELDERLY

Background: While central nervous system (CNS) active medications such as psychotropics and narcotic analgesics have been implicated in cont ributing to falls in older adults, the combined effect of multiple CNS -active medications has not been investigated. The purpose of this stu dy was to examine the influence, in community-dwelling elderly, of (1) taking multiple CNS-active medications on fall liability and (2) indi vidual classes of CNS-active medications (using discrete drug classifi cation) on the risk of falls after controlling for important confounde rs -age, mobility, cognition and depression. Methods: 305 community-dw elling male veterans (age: 70-104) were screened at study entry for mo bility, cognition and depression. CNS-active medications were categori zed as benzodiazepines, other sedative-hypnotics, neuroleptics, tricyc lic antidepressants, and opioid analgesics. Subjects were prospectivel y followed for 6 months to monitor falls; at the end of this time peri od, subjects were classified as fallers (at least one fall) or nonfall ers. The relationship between CNS-active drug use and falls was examin ed using multivariable analyses. Results: The risk of falls was signif icantly greater in CNS-active medication users as compared with nonuse rs. Adjusted odds ratio for one CNS-active drug was 1.54 (95% confiden ce interval 1.07-2.22) and for two or more agents 2.37 (95% confidence interval 1.14-4.94). Conclusions: In community-dwelling elderly, the use of multiple CNS-active medications is associated with enhanced fal ls liability, over and above the use of one CNS-active drug alone. Thi s apparent dose-response relationship provides support for causality.