The Quantitative Impact of Visual Function on Accelerometer-measured Physical Activity in Older United States Adults: A Nationwide Cross-sectional Analysis

Abstract

ObjectiveTo explore the impact of objective vision measures on novel metrics of objectively-measured physical activity (PA) in a nationally representative sample of US older adults. DesignCross-sectional analysis using data from the National Health and Aging Trends Study. ParticipantsAdults had their distance and near visual acuity (VA) and contrast sensitivity (CS) tested. Any objective vision impairment (VI), defined as any VI in distance VA, near VA, or CS, was the primary exposure. PA data were collected using the Actigraph CentrePoint Insight Watch worn for 7 days. MethodsMultivariable regression models were used to investigate the association between vision and PA measures. All analyses accounted for the survey design and models were adjusted for age, sex, race, living arrangement, education, and comorbidities. Main OutcomesPA metrics included: 1) total daily activity (active minutes per day, number of active bouts, and mean length of active bouts), 2) activity fragmentation, and 3) time until 75% activity. An active bout was defined as ≥1 consecutive active minute. Activity fragmentation was defined as the probability of an active minute being followed by a sedentary minute, with higher values indicating more fragmented activity. Time until 75% activity was defined as the time taken to complete 75% of daily PA starting from their first active bout. ResultsAmong 723 participants, sampled from 10,443,338 older adults in the US, 30% had any objective VI. Any objective VI was significantly associated with lower number of active minutes per day (7.8% fewer, [95% CI: −1.7% to −13.6%]), shorter active bouts (7.0 % shorter, [95% CI: −1.4% to −12.3%]), and greater activity fragmentation (β= 2.5%, [95% CI: 0.8% to 4.2%]), while no associations were found with number of active bouts. Time until 75% activity did not significantly differ between adults with any objective VI and those without (p=0.34). ConclusionsOlder US adults with any objective VI displayed lower total daily activity, as well as more fragmented, shorter periods of PA, despite having a similar number of active bouts compared to their normally sighted counterparts. Implementing interventions that increase bout duration may help promote physical activity in adults with VI.