Agarwal N, Hansberry DR, Sabourin V, Tomei KL, Prestigiacomo CJ. A Comparative Analysis of the Quality of Patient Education Materials From Medical Specialties. JAMA Intern Med. 2013;173(13):1257-1259. doi:10.1001/jamainternmed.2013.6060
Given the access to a seemingly unsurpassable amount of information online, one can understand why the Internet has become one of the most commonly used sources of information, including health care–oriented resources. According to a 2011 study performed by the Pew Internet and American Life Project, 59% of Americans use the Internet to find and understand health care–oriented information.1 However, a potential problem is the difficult reading level of the patient-specific education materials. The average American adult reads at approximately a seventh to eighth grade level.2 Therefore, the American Medical Association, the National Institutes of Health, and the US Department of Health and Human Services advocate for patient education materials to be written at a fourth to sixth grade reading level.2- 4 As explored in this Research Letter, we assess the readability of patient education resources by using various readability parameters. To our knowledge, this is the first study to compare the readability of patient education materials to comprehensively assess the quality of resources provided by various medical professional organizations.
Online patient education materials from each medical specialty were downloaded in 2012. Resources from the 16 specialties were examined. For each website, material written specifically for patients was downloaded into Microsoft Office Word (Microsoft). Tables, figures, hyperlinks, and text unrelated to the patient education material, including copyright notices, disclaimers, and author information, was deleted.
Readability assessment of each article was performed using Readability Studio Professional Edition, Version 2012.1 (Oleander Software). The analysis included the Coleman-Liau index, FORCAST formula, simple measure of gobbledygook (SMOG) grading, New Dale-Chall readability formula, Flesch Reading Ease, Flesch-Kincaid grade level, Fry graphical analysis, Gunning fog index, New Fog Count, and the Raygor readability estimate. A separate analysis was performed to identify grammatical errors, cliches, and passive voice.
All readability assessments, excluding the New Fog Count, showed that patient education materials from all 16 medical specialties were too complex for the recommended sixth grade reading level (Table). The New Fog Count yielded the following scores near the recommended guidelines: dermatology, 4.3; obstetrics and gynecology, 6.0; plastic surgery, 6.1; and family medicine, 6.6. The New Dale-Chall readability formula test showed that only dermatology, family medicine, and obstetrics and gynecology were within the boundaries of the average American adult reading level. Flesch Reading Ease readability analysis showed that largely, patient resources were considered to be “difficult.” For the Flesch-Kincaid grade level readability test, family medicine was the only specialty within the parameters of the average adult reading ability. Readability scores using the Fry graphical analysis test ranged from the eighth grade level in family medicine to unclassifiable in dermatology because the complexity of the patient educational materials was beyond the 17th grade level.
Overall, across all readability analyses used to measure each of the 16 websites, the New Fog Count demonstrated the lowest mean grade level score of 9.3, whereas SMOG grading demonstrated the highest mean grade level score of 14.1. The proportion of passive voice sentences used throughout resources ranged from 4% in family medicine to 27% in neurological surgery. Obstetrics and gynecology materials contained the most cliches with a total of 40, corresponding to 5.8 cliches per 50 pages. Obstetrics and gynecology materials also contained the highest total number of indefinite article mismatches (the improper use of “a” or “an”) at 14 errors, corresponding to 1.8 errors per 50 pages.
Research conducted at the US Department of Education found that 12% of adults had proficient health literacy, 53% had intermediate health literacy, 22% had basic health literacy, and 14% had below basic health literacy.5 As a result, on an individual level, problems arise in the form of preventable recurrent hospitalizations or visits. On the national level, there are negative economic consequences: it has been estimated that inadequate health literacy is costing the US economy between $106 and $236 billion dollars annually.6
Our analysis of the level of readability across all 10 readability scales showed that none of the patient education resources provided by the 16 professional organizations met the recommended sixth grade maximum readability level or even the seventh to eighth grade reading ability of the typical American adult. As such, website revisions may be warranted to increase the level of readability and quality of these patient resources to effectively reach a broader patient population. One simple adjustment is to write more clearly, which may increase comprehension regardless of the reader’s health literacy capabilities.7 The use of pictures and videos may also be an effective way of increasing a patient’s comprehension of health information that is too complex to fully explain through pure text.8 Future studies will seek to better explain the relationship between readability and multimedia effectiveness at improving the communication of health information, which would ultimately help to improve patient comprehension and outcomes.
Corresponding Author: Dr Prestigiacomo, Department of Neurological Surgery, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, 90 Bergen St, Ste 8100, PO Box 1709, Newark, NJ 07101 (firstname.lastname@example.org).
Published Online: May 20, 2013. doi: 10.1001/jamainternmed.2013.6060
Author Contributions: Mr Agarwal and Dr Hansberry served as co–first authors, each with equal contribution to the manuscript.
Study concept and design: Agarwal, Hansberry, and Prestigiacomo.
Acquisition of data: Agarwal, Hansberry, and Sabourin.
Analysis and interpretation of data: Agarwal, Hansberry, Tomei, and Prestigiacomo.
Drafting of the manuscript: Agarwal, Hansberry, Sabourin, and Tomei.
Critical revision of the manuscript for important intellectual content: Agarwal, Hansberry, Tomei, and Prestigiacomo.
Statistical analysis: Agarwal and Hansberry.
Administrative, technical, and material support: Agarwal, Hansberry, Tomei, and Prestigiacomo.
Study supervision: Agarwal, Hansberry, Tomei, and Prestigiacomo.
Literature review: Sabourin.
Conflict of Interest Disclosures: None reported.
Additional Contributions: Chirag D. Gandhi, MD, provided guidance throughout the duration of this study.