Abstract
BACKGROUND AND OBJECTIVES: The National Health and Aging Trends Study (NHATS) is an ongoing, nationally representative panel study of older adults that collects data on health and disability, including measures on self-reported visual disability (SRVD). Prior studies assessing the association of SRVD with other measures of healthy aging have classified participants as having or not having SRVD, which does not capture the full spectrum of SRVD reported by participants. Therefore, we sought to develop and validate an ordinal indicator of SRVD to facilitate research on the impact of late-life SRVD on health and disability in NHATS. RESEARCH DESIGN AND METHODS: We used 2015 NHATS data with community-dwelling participants who answered survey questions about visual functioning and vision aid use. Based on responses, participants were categorized into one of 6 groups: blind, near and distance SRVD without vision aid use, near and distance SRVD with vision aid use, near or distance SRVD without vision aid use, near or distance SRVD with vision aid use, or no SRVD. Multivariable Poisson regression models assessed convergent validity of the ordinal SRVD scale with functional activity and well-being scores, while adjusting for demographic factors and medical comorbidities. RESULTS: Of the 7061 eligible individuals, 8.3% (n = 742) reported SRVD. Using our novel ordinal indicator of SRVD in NHATS, higher levels of SRVD were significantly associated with lower functional activity scores (p < .001 for all) and subjective well-being (p < .001), except for participants reporting blindness. Significant differences between SRVD groups were found, which could not be captured using a binary SRVD variable. DISCUSSION AND IMPLICATIONS: A novel 6-level SRVD scale in NHATS demonstrated convergent validity with functional activity and well-being scales. This scale provides a new tool with improved measurement precision to study the impact of late-life SRVD on health and disability in a nationally representative study of older adults.