Comparing Reading Speeds for Reading Standardized Single Sentences and Paragraphs in Patients with Maculopathy

Introduction: The purpose of this study was to compare reading speeds (RS) of either paragraphs of text or single sentences in patients with maculopathy by investigating the repeatability of RS to decide which text type to use in studies on patients with maculopathy. Methods: RS was measured for standardized texts in 25 participants with a central scotoma due to maculopathy (mean age 77.8 years +/- 9.9 SD, mean binocular visual acuity 0.65 logMAR (+/- 0.85 SD), median magnification requirement 3-fold). Reading 3 single sentences taken from the Radner reading charts (sample #1, #2, and #3) of 14 words and reading 3 paragraphs of International Reading Speed Texts (IReST, sample # 3, #6, and #10) with a mean of 132 (+/- 3.2 SD) words, each in German, were compared. The 6 texts were read aloud in random order from a closed-circuit TV system, with size adjusted according to the individual magnification requirement. Reading time was measured by stopwatch, and speed was calculated in correctly read words per minute (wpm). Differences in RS depending on text length (single sentence vs. paragraph) and text sample were calculated by the Bland-Altman analysis. Results: The mean RS showed no significant difference between 2 charts of the same kind (sentences: 93 wpm +/- 37 SD; paragraphs: 95 wpm +/- 38 SD). RS differences between 2 charts were lower in paragraphs than in single sentences. Highest correlations of RS between all 6 texts existed between the 3 IReST text samples (r = 0.98, 0.98, and 0.98) compared with the 3 Radner sentences (r = 0.89, 0.81, and 0.90). The inter-chart reliability (coefficient of repeatability) was smaller for the paragraphs (12.9 wpm) than for the single sentences (36.4 wpm). Conclusion: In patients with maculopathy, single sentences are well suited for single measurement of RS. For repeated measurements (e.g., monitoring the course of a reading disorder or assessing effects of interventions), paragraphs are preferable because of their lower variability of RS between the paragraphs.