Abstract
Purpose: :
To assess the impact of bilateral visual impairment (VI) on subsequent falls among older persons.
Methods: :
Of 3654 baseline participants of the Blue Mountains Eye Study, 2334, 1952 and 1146 were re-examined after 5, 10 and 15 years. Bilateral VI was defined as habitual visual acuity (measured without best refractive correction but wearing current spectacles if participant owned a pair that was used regularly) <20/40 in the better eye, determined at each examination. The incidence of falls was determined by self-reported history of two or more falls in the a 12 month period prior to the next examination following the examination when bilateral VI was present. Discrete logistic models with time-dependant variables were used to analyze the associations between VI and sthe ubsequent incidence of falls (outcome) in 5 years, adjusting for age, gender, walking disability and the presence of co-morbidities (diabetes, hypertension, hypercholesterolemia, arthritis, and cardiovascular disease) that were present at the same examination when bilateral VI was present.
Results: :
Of the 3654 baseline participants, 367 (13.5%) had bilateral prevalent VI at baseline, and 230 (12.4%), 205 (14.6%) and 113 (13.6%) participants had bilateral VI at the 5-, 10- and 15-year follow-up visits respectively. The incidence of newly reported two or more falls was found in 193 (10.4%) participants at the 5-year visit (20.7% of persons with VI at baseline vs 7.3% of persons without VI had a fall), 192 (13.7%) participants at the 10-year visit (17.4% of persons with VI at the 5-year visit vs 9.4% of persons without VI had a fall), and 82 (9.8%) participants (19.0% of persons with VI at the 10-year visit vs 12.8% of persons without VI had a fall).After adjusting for potential confounders, the presence of bilateral VI in the previous visit was associated with incidence of having two or more falls in a period 12 months prior to the next visit in 5 years (odds ratio (OR) 1.7, 95% confidence interval (CI) 1.2-2.5). We did not find significant associations of the use of multifocal glasses, recently (<1 year) acquired glasses or having cataract surgery with incident falls either within the next 5 years or in 5 or more years.
Conclusions: :
Longitudinal data from this older Australian sample reveal that the presence of bilateral VI was significantly associated with subsequent falls in five years, after adjusting for potential confounders.