Abstract
Purpose::
To evaluate change in reading performance over time in an elderly population-based sample and to determine which vision variables are associated with this change in reading.
Methods::
Smith-Kettlewell Institute (SKI) Study participants completed an extensive battery of vision tests at three points in time over the course of 9.2 years. The mean age of the participants at baseline (BL) was 75.5 years (stdev: 9.3, range: 58.4-101.9 yrs). The test battery included high and low contrast acuity, low luminance, low contrast acuity, low contrast acuity in the presence of disability glare, contrast sensitivity, glare recovery time, stereopsis, and standard and attentional visual fields. Additionally, reading performance was assessed with the Pepper Reading Test (Corrected Reading Rate [CRR]=words read correctly per minute). In longitudinal studies, measures obtained for each individual participant over time are correlated. Longitudinal analysis techniques allow modeling of change while taking into account these "within-participant" correlations, and also permit the use of all available data in the longitudinal dataset. Therefore, random coefficient analysis was used to evaluate the association between CRR and other vision variables over time, while controlling for age, sex, and years of education.
Results::
Of the original 902 participants seen at BL, 596 returned for the second wave of testing (mean=4.4 years after BL), and 452 returned for the third test (mean = 7.0 years after BL). CRR means for the three tests were 81.5, 74.7, and 70.3 wpm, respectively. The random coefficient analysis revealed that declines in several vision measures were significantly associated with change in CRR. Decrements in high contrast acuity, low contrast, low luminance acuity, low contrast acuity in disability glare, glare recovery time, stereopsis, standard and attentional fields were all significantly associated with decreases in CRR.
Conclusions::
In this elderly sample, changes in several vision measures were significant independent predictors of reading performance change. These longitudinal results confirm and extend our previously reported cross-sectional findings. Data collection on the fourth wave of SKI Study testing will be complete before ARVO, and these results will be included in the final analyses.
Keywords: aging: visual performance • reading