Abstract
Purpose: :
To compare error patterns of the MNread test with a similarly formatted test using random words (SKread).
Methods: :
177 eyes of 96 consecutive patients in a retina practice had visual acuity (VA) measurement, continuous text reading (MNread Chart) and random word reading (SKread Chart) performed. The new test utilized the same 60 character per font size format as the MNread chart but instead of sentences, used random words designed to impede cognitive correction of errors as does the Pepper Visual Skills for Reading Test. Types of errors committed were categorized into misidentifications (MIS) and omissions (OM). Eyes not able to read the largest (8M) text were excluded. Tested eyes were categorized as normal (NL, n=65) or diagnosed disease (DD, n=112). Age median/range was NL= 56/28–85, DD= 74/31–96. Median education was NL 16 and DD 16 years. Mean daily reading time was NL 3.7 and DD 2.3 hrs/day. English was the first language of 79% of patients.
Results: :
Median VA was NL 20/20 and DD 20/50. Mean error rate (errors/60 character block) for continuous text was NL 0.02, DD 0.20 and for random words was NL 0.50, DD 1.45. Mean error rates (both MNread and SKread) were not significantly associated with age, daily reading time or education level and only weakly associated with visual acuity (r2 =0.08 and 0.21 respectively). The 24 worst perfoming of 112 eyes with pathology (21.4%) made two or more errors per block on random words compared with 0.01% for continuous text. DD mean error rate was significantly different (Mann Whitney test) compared with the NL group. Looking at the error types for random words, misidentification rate (misidentifications/60 character block) was NL 0.39, DD 0.75 and omission rate was NL 0.11, DD 0.67. DD misidentification and omission rates are significantly different (Mann Whitney test) compared with the NL group.
Conclusions: :
Patients with eye disease make more errors when reading a random word version compared with continuous text version of the MNread. Misidentifications and omissions are more frequent when reading random words. Error rates with random words and continuous text are only weakly associated with visual acuity.
Keywords: low vision • reading • quality of life