Background
Biological underpinnings of falls in older adults are not well established. Objectives: To examine the validity of selected oxidative stress and inflammatory biomarkers for predicting incident falls in community-dwelling older adults. Design: Prospective cohort study. Setting and participants: 266 non-demented and ambulatory community-dwelling older adults (mean age 78 years, 55% women). Measurements: Oxidative stress (malondialdehyde) and inflammatory (interleukin-6 [IL-6]) biomarkers were selected based on associations with fall risk factors, and values were log-transformed to account for non-normal distributions.
Conclusion
Oxidative stress predicted falls in a community-based cohort, and should be further examined as a fall risk biomarker as well as a potential target to prevent falls.
Results
Over a mean follow-up of 20.5 ± 10.1 months, 119 participants fell. In Cox proportional hazards models, each one standard deviation increase in baseline log-malondialdehyde levels predicted incident falls (Hazard ratio (HR) adjusted for age, gender, education, comorbidity count, medications, log-IL-6 levels, prior falls, depressive symptoms, cognitive status, gait velocity, and balance 1.53, 95% CI 1.11-2.16). Log-IL-6 levels were not associated with falls. Participants in the highest log-malondialdehyde quartile at baseline had increased risk for incident falls than those in the lowest quartile (HR 2.47, 95% CI 1.41-4.34).