Is the use of benzodiazepines associated with incident disability?

OBJECTIVES: This study examined the association between benzodiazepine use and incident disability with an emphasis on elucidating whether the underlying health conditions that result in benzodiazepine use (confounding factors) or intrinsic adverse effects of benzodiazepine use were responsible for functional decline.
DESIGN: Cohort study with follow-up of 4 to 5 years.
SETTING: A health maintenance organization (HMO) in western Washington. PARTICIPANTS: Individuals aged 65 and older from a random sample of HMO enrollees who participated in a health promotion intervention trial (n = 1,519). MEASUREMENTS: Benzodiazepine use was ascertained from computerized pharmacy records. Self-reported functional status was assessed using a six-item physical function scale ranging from vigorous activity to self-care activities of daily living (ADLs). Two outcomes were examined: decline in overall physical function and limitations in self-care ADLs. Multivariate models were examined that included demographic characteristics, health status, and health behaviors that were likely to be confounders. Several analyses were conducted to examine whether benzodiazepine use or confounding factors were responsible for functional decline.
RESULTS: Benzodiazepine use was significantly associated with incident loss of physical function (hazard ratio (HR) = 1.51, 95% confidence interval (CI) = 1.02-2.24) in the fully adjusted model. Although use of benzodiazepines was associated with limitations in ADLs, it was not significant when adjusting for other factors (HR = 1.71, 95% CI = 0.87-3.34). Several of our findings suggest that the health conditions leading to benzodiazepine use may partly or fully explain these associations: (1) use of anxiolytic benzodiazepines (HR = 1.95, 95% CI = 1.24-3.07), but not hypnotic agents (HR = 1.21, 95% CI = 0.73-2.00), was associated with functional decline; (2) adjustment for health status variables minimized these associations; and (3) there was little evidence of dose response.
CONCLUSIONS: A modestly increased risk for decline in physical function was associated with benzodiazepine use, especially of anxiolytic agents. The health conditions that result in benzodiazepine use may be more important in the pathogenesis of disability than benzodiazepine use itself. Although there are many reasons for avoiding benzodiazepines in older adults, it is still unclear whether use contributes independently to functional decline.