Date of Award

January 2013

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

James C. Tsai

Subject Area(s)



Purpose: To compare the effects of uncorrected refractive error (URE) and nonrefractive visual impairment (VI) on disability measures.

Design: Cross-sectional population-based study.

Participants: 2469 individuals with binocular presenting visual acuity (PVA) of 20/80 or better who participated in the first round of the Salisbury Eye Evaluation study.

Methods: URE was defined as binocular PVA of 20/30 or worse, improving to better than 20/30 with subjective refraction. VI was defined as post-refraction binocular best corrected visual acuity (BCVA) of 20/30 or worse. The visual acuity decrement attributable to VI was calculated as the difference between BCVA and 20/30 while that due to URE was taken as the difference between PVA and BCVA. Multivariable

regression analyses were used to assess the disability impact of 1) vision status using the group with normal vision as reference and 2) one-line decrement in acuity due to VI and URE.

Main Outcome Measures: Objective measures of visual function obtained from timed performance of mobility and near vision tasks, self-reported driving cessation, and self reported visual disability measured by the Activities of Daily Vision (ADV) scale.

Results: Compared to individuals with normal vision, subjects with VI (n=191) demonstrated a significant decline in all objective and subjective metrics of visual function (p<0.05) while subjects with URE (n=132) demonstrated slower walking speeds, slower near task performance, more frequent driving cessation, and lower ADV

scores, but did not demonstrate slower stair climbing or descent speed. For all disability metrics evaluated, the impact of VI was greater than the impact of URE. The impact of a one-line decrement in visual acuity due to VI was associated with greater deficits in mobility measures, driving cessation, and self-reported visual function when compared to a one-line visual acuity decrement due to URE.

Conclusions: VI is associated with greater disability than URE across a wide variety of functional measures, even in analyses adjusting for the severity of vision loss. Refractive and non-refractive vision loss should be distinguished in studies evaluating visual disability, and should be understood to have differing consequences.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.