Readability of websites containing information about prostate cancer treatment options.

Abstract

e16544 Background: 90 million adults have literacy skills that test below a high school reading level. Websites written above this level can pose a challenge for those seeking more information about prostate cancer treatment options. In this study, we determine the readability of selected websites, using a systematic search process and validated readability formulas. Methods: We identified the three most popular keywords (“prostate cancer,” “prostatectomy,” and “prostate cancer treatment”) from 513 terms related to prostate cancer treatment options. Three physician searchers (CE, AP, AK) systematically collected 270 websites from the top three search engines, representing 95 % of all internet searches (Google, Yahoo, and Bing). We excluded sites that were non-English, not primarily text, irrelevant and/or duplicated, and then categorized the remaining sites by source (government, academic practice, nonacademic practice, commercial, nonprofit, news, health website, other) and treatment type. We used validated readability instruments, the Flesch-Kincaid (FK) grade level and Flesch reading ease, to calculate scores for each site. We also calculated an intraclass correlation coefficient, in order to assess inter-rater agreement among our three searchers. Results: Sixty-two unique websites were available for analysis. The mean FK grade level of all sites was 11.4 (SD +/- 0.95, range 8.0-12.0), indicating that on average, passages were written at an eleventh grade reading level. The mean Flesch reading ease score on all sites was 35.6 (SD +/- 15.7, range 0.0-65.5), indicating that on average, passages were slightly easier to read than passages more suited for university graduates (0-30). Only three sites (4.8%) were written below a high school reading level (< 9.0). The intraclass correlation coefficient between the three searchers was 0.96 for grade level and 0.97 for reading ease. Conclusions: Few websites with discussions on prostate cancer treatment options are written below a high school reading level. This is problematic for patients with a low literacy level who search online for more information. Clinicans should be aware of this finding and guide patients to websites written at their patient's reading level.