Concordance between Self-Reported Visual Difficulty and Objective Visual Impairment: The National Health and Aging Trends Study.

ABSTRACT

PURPOSE: To examine the performance of self-reported visual difficulty (VD) in predicting objective visual impairment (VI) in older adults and explore factors that influence discordance. DESIGN: Cross-sectional analysis of the National Health and Aging Trends Study (2022). METHODS: Participants reporting blindness or difficulties with distance or near vision were characterized as having VD. Presenting binocular distance visual acuity (VA), near VA, and contrast sensitivity (CS) were assessed. Objective VI was defined as having VI in distance VA (worse than 20/40), near VA (worse than 20/40), or CS (worse than 1.55 logCS). Receiver operating characteristic analysis was used to compare performance of VD in predicting VI. To investigate factors that influence discordance, we limited our sample to adults with VI and used a multivariable logistic regression model to identify factors associated with not reporting VD. Similar analyses were performed to explore factors associated with reporting VD in adults without VI. MAIN OUTCOME MEASURES: Discordance factors. RESULTS: Four thousand nine hundred ninety-nine adults were included in the 2022 cohort. Visual difficulty achieved an area under the curve (AUC) of 56.0 (95% confidence interval [CI], 55.2-56.9) in predicting VI, with a sensitivity of 15.8 (95% CI, 14.2-17.5) and specificity of 96.3 (95% CI, 95.5-96.9). Characteristics associated with not reporting VD in adults with VI included female gender (odds ratio [OR], 0.64 [95% CI, 0.42-0.99]), Hispanic ethnicity (OR, 0.49 [95% CI, 0.31-0.78), higher income (≥75 000, OR, 1.99 [95% CI, 1.14-3.45]), ≥4 comorbidities (OR, 0.46 [95% CI, 0.29-0.72]), and depressive symptoms (OR, 0.49 [95% CI, 0.25-0.93]). Factors associated with self-reporting VD in the absence of VI included Hispanic ethnicity (OR, 2.11 [95% CI, 1.15-3.86]), higher income (≥$75 000, OR, 0.27 [95% CI, 0.12-0.63]), and anxiety symptoms (OR, 3.05 [95% CI, 1.56-5.97]). CONCLUSIONS: Self-reported VD is a distinct measure assessing disability and has limited ability in predicting objective VI. Caution is advised when using self-reported VD as a surrogate measure for objective VI in epidemiological studies, although it may still be an effective way to capture risk of current or future disability. FINANCIAL DISCLOSURE(S) Proprietary or commercial disclosure may be found after the references.