Much of the early research on readability has focused on developing tools for objectively assessing the readability of texts in order to match them to the reading skills of readers. This effort has resulted in the development of many readability formulas, which have been widely used. The Flesch-Kincaid formula (1975) is one of the most popular and heavily tested formulas and it correlates 0.91 with comprehension as measured by reading tests.6 There are many individual articles evaluating the readability of specific documents using this formulae as well as the Flesch Reading Ease score (1948), FOG index (1952), Fry Readability Graph (1963), Simple Measure Of Gobbledygook (SMOG) (1969) formulae and several others (refer to examples 7-9). These formulae evaluate documents based on Natural Language Processing (NLP) and rely on measures such as sentence length, word length and number of polysyllabic words, which in a health environment may not necessarily be useful as the propensity for polysyllabic words in common use eg diabetes, may confuse the evaluation.
These formulae have been widely used to evaluate a variety of patient information pamphlets and instruction sheets which generally have concluded that the reading age of the material is inappropriate for the intended audience. However, almost none of these studies evaluate the effects of visual or design factors that could influence readability and comprehension. A study in 2008, by Kandula and Zeng-Treitler 10 in an attempt to create a ‘gold standard’ for the readability measurement of health texts, evaluated 324 documents noted that a document’s style - such as font, sub-headings, and bulleted lists affected the document’s readability but did not evaluate further.
On the whole very little has been written about the impact of layout of text on readability of pathology or healthcare related reports. This may be a result of the perception that style (text layout) is primarily about aesthetics and marketing (branding) with minimal impact on readability. In a study by Frost in 199911 on the “importance of format and design in print patient information”, a survey of 44 oncology healthcare professionals identifying characteristics that they believed important for effective print educational materials was undertaken. Style (text layout) was initially perceived to be the least important factor impacting readability by the surveyed oncologists; however it in fact made a major impact on acceptance and a nine-fold increase in distribution of the redesigned material.
In the last decade there appears to be a slowly increasing recognition that style (text layout) may be a major contributor to the overall predictor of readability. A study by Arnold et al in 2006 in their Assessment of Newborn Screening Parent Education Materials developed the "User-Friendliness Tool (UFT)”12 which focussed on issues such as layout, illustrations, message, information, and cultural appropriateness. They viewed layout as “critical” to reading ease. Many very recent assessments of readability of educational and instructional material (examples 13-14)have used one of the abovementioned readability formulae’s in conjunction with the UFT. The UFT has taken the recommendations of several manuals and texts and distilled this into general layout recommendations such as font size (≥ 12 point), type (avoiding all-capital letters, italics, and specialty fonts), white space (ample), paragraph size (<4-5 lines) and the presentation of manageable information (use of bullets and boxes to break up text).
In a 2009 MIT study 15, Bufuka, recognising that current readability measures depend heavily on text analysis (NLP) but neglected style properties (text layout), found that style (text layout) is in fact an important predictor of a documents readability. His study developed a program to analyse images of a document and then review the style (text layout) and predict readability. This program, evaluating style (text layout) alone, provided a closer correlation to human expert assessments of readability than the Flesch-Kincaid and other NLP methods15.
One of the more recently developed readability formulae, the PMOSE/IKIRSCH formula16 in 1998 recognised the role of structure/design on readability. It was developed to evaluate overall document complexity based on structure (design) (e.g., simple list, combined list, and nested lists), density (e.g., number of labels and items) and dependency (whether or not any important information is to be found outside the document).