Abstract
Purpose: :
Outcome measures are needed for clinical trials to determine which low vision reading programs are most cost–effective and the dose of rehabilitation needed to achieve a treatment response. This study compares the responsiveness of reading performance and self–report measures to the effects of low vision rehabilitation.
Methods: :
The Pepper VSRT and VA LV VFQ–48 were administered pre and post rehabilitation to 72 veterans with macular diseases who were participating in a comprehensive blind rehabilitation program at Hines VA Hospital. The MNREAD was also administered to 24 veterans from the original cohort.
Results: :
There is a large treatment effect, defined as Cohen’s d= .8 or larger, for the change in difficulty patients self–report performing reading activities on the VA LV VFQ–48 (d= 2.44, p= 4.1x10–29 for paired t–test) and the change in number of words read correctly on the Pepper VSRT (d= 1.18, p=5.26x10–8 for paired t–test). A medium treatment effect, defined as Cohen’s d=.5 or larger but less than .8, exists for the increase in reading speed on the Pepper VSRT (d= .77, p=.0003) and the decrease in critical print size read on the MNREAD (d=.5, p=.04 for paired t–test).
Conclusions: :
The reading performance measures are responsive to the effects of rehabilitation, but the self–report measure shows the largest magnitude of change. Effect sizes should be considered when choosing outcome measures and computing sample sizes for clinical trials.
Keywords: clinical (human) or epidemiologic studies: treatment/prevention assessment/controlled clinical trials • low vision • reading