We report on a subset of data from a larger study exploring patients’ decision making to accept or decline an ICD for a primary prevention indication from 3 sites in southern Ontario, Canada.20
Ethics approval for the study was obtained from all 3 study sites. During interviews with participants who had accepted or declined an ICD for primary prevention, participants were asked to identify and share with the interviewer any written material they had obtained about the ICD before or after their decision to accept or decline the ICD. The analyzed materials were documents available to participants (and to the public-at-large) in ICD waiting room areas and/or given to them by their physicians or nurses at some point before or after their decision to accept or decline the ICD. Pamphlets related to ICDs were gathered during the recruitment process from waiting areas of the 3 sites and retrieved from the recruitment sites after the interviews by a research assistant. We conducted readability testing and content analysis focusing on the themes represented in the educational material and the rhetoric implied in its messages.
We were conscious that each member of the investigative team had preexisting beliefs, interests, and experiences that could affect the analysis. We began by surfacing these in team meetings and adopted a dialectical approach to the analysis. Meeting and analysis notes were kept to record issues, questions, and decisions at each phase of the analysis process. Analysis was conducted by P.H.S. (a nurse scientist with expertise in qualitative methods), S.d.L. (a master’sprepared research assistant), and 2 fourth year nursing students (G.K.T. and K.V.).
Two readability tests, the SMOG (“simple measurement of gobbledygook”)21
methods, were used to cross-check the grade-level calculations. The SMOG method samples 30 sentences combined from the beginning, middle, and end of the document to provide an overall average grade level. The Fry method plots 3 samples of 100 words on a grade-level graph to determine readability scores. Both of these testing methods were developed in the late 1960s to facilitate readability testing, and both have been used extensively in the generation and evaluation of medical education material, with SMOG being endorsed by the National Institutes of Health in preparing government documents.23
Two analysts (K.V. and G.K.T.) independently conducted SMOG and Fry analysis on all documents, and we met to discuss and critically review issues that arose in applying the readability assessment tools.
Because readability cannot be measured by syllable count and sentence length alone, we also developed analysis criteria () based on the recommendations from several authorities on the development and evaluation of patient education material.12,13,24,25
Print material evaluation criteria
In addition, we analyzed the text for content and rhetorical tone. Content analysis informed by Krippendorff26
structured this aspect of the methodological approach. Following Krippendorff, we questioned the material, incorporated relative context, and generated inferences about the data. Thematic content analysis of the 21 print documents was done by 3 team members (P.H.S., G.K.T., and K.V.). We began by developing an a priori list of themes based on a scan of the documents. Two of the analysts (G.K.T. and K.V.) independently coded each of the documents using the a priori list as a guide. Several subthemes were identified and the team met to review the analysis, discuss and challenge coding decisions, establish interrater reliability, and reach consensus. Finally, 12 overarching content themes were identified.
We then used a rhetorical lens to analyze and explore language and images in the documents for persuasive tones, thereby extending the analysis beyond descriptive content analysis. Whereas content analysis incorporates reflection on the meaning and messages being conveyed, rhetorical analysis aims to highlight the literary and visual symbols used to persuade readers and viewers to the author’s intended message. The texts were reviewed in their entirety for persuasive tones that appeared mainly in metaphors, in bolded statements, in adjective and adverb use, and in the juxtaposition of images with text. The attribution of meaning to text and images is culturally constructed and can be misunderstood depending on personal and historical circumstances.27
Three team members (P.H.S., S.d.L., and G.K.T.) reviewed the texts independently for rhetorical messages and tone. Before and at points during the analysis process, the analysts encouraged and challenged each other to reflect on their interpretations and surface preexisting biases. After several rounds of this process, consensus about the rhetorical meaning was achieved.
The participants from the larger decision-making study represent the demographics of study individuals who were offered an ICD for primary prevention. We highlight relevant demographics in .
Overall, 21 educational documents were reviewed. All material was intended for patients who had primary or secondary indications for an ICD. The materials reviewed had publication dates ranging from 2005 to 2008, with 1 document from 2000. Twelve documents were published by ICD manufacturers, 6 were sitespecific hospital publications, and 3were published by 2 different cardiac support organizations. Hospital documents tended to be easily reproducible letter-size pages or booklets that contained few (if any) colors and minimal illustrations. Manufacturer and support organization publications were more often full-color, glossy pamphlets with photographs and illustrations throughout. Generally, the information that related to anatomy, physiology, pathophysiology, device function, and therapeutic intent was consistent across all the materials. Inconsistencies that occurred tended to be in relation to perioperative care. Device manufacturer documents tended to focus on anatomy, pathophysiology, the purpose and function of the ICD, and impact of the device on lifestyle. Hospital-generated material focused more specifically on function of the ICD and perioperative management.
Overall, documents published by device manufacturers had lower readability scores and more simple design layouts. However, these documents also contained more persuasive language and images than did those published by hospitals and support organizations. Consistent messages were found across all the documents and contradictions were minimal; however, the depth and breadth of the content in each document varied.
Language, Word Use, and Readability
Although readability testing considers the number of polysyllabic words per text sample (number of words or number of sentences), it does not make visible the use of scientific jargon or the use of active and passive voice. All but 5 documents contained complicated scientific jargon, and an active voice was used in all documents. Scientific jargon was commonly used to describe clinical symptoms such as ventricular tachycardia or bradycardia. In most cases, these terms were followed by an explanation, but not always a simple one. In other cases, jargon was used unnecessarily. For instance, one pamphlet stated that the hospital will “try to interrogate the device” if an ICD recipient died at some point in the future. Because jargon may use few syllables, and active/passive voice cannot be measured in readability tests, the scores listed in are only a partial measurement of the documents’ readability. The analysis we undertook extends further than traditional readability by incorporating the document design features and language use.
Readability scores of select ICD patient education material
In some cases, a large discrepancy in readability scores existed within a document; this can be attributed to the randomness of the selected text sampled by the independent analysts. We subsequently calculated an average readability score for the final measure. There was a large variation in the readability of the materials; the reading level ranged from grade 9 to 15 and from 9 to 19 years of age or more. Ninety-five percent of the documents in our sample scored a higher than a grade 8 reading level. Half of the documents’ readability scores were in the upper high school, lower university years range (grade 11 and higher). All 3 author types (manufacturers, hospitals, and support groups) are represented in the half of documents requiring higher reading abilities. The 3 lowest scoring documents were from a device manufacturer, hospital, and cardiac health support organization respectively.
All but 1 document (a hospital-manufactured, photocopied information pamphlet that appears to have been reduced in size during reproduction) used 11 or 12 font type. Three hospital-manufactured documents used 14-point font. The majority (66.6%) used a Serif font similar to Times New Roman. All but 3 incorporated typographic cues such as bullets and numbering to facilitate directing readers, and all used subheadings to help organize content.
Graphics, Illustrations, and Tables
All but 3 of the documents provided illustrations/images on the pamphlet’s front cover. None of the covers contained an ICD in its main picture. A total of 6 documents, 4 from device manufacturers, did not have any internal images. Generally, internal images did not have captions explaining their representations, and more often, the images did not correspond to the nearby text. For example, one device manufacturer repeatedly included images of male seniors playing with a child, presumably a grandchild, or playing golf, next to text about ventricular tachycardia or warning signs of SCD risks. Six documents contained images of the ICD, the heart muscle, or both that provided a visual representation of (a) the ICD’s implant location in relation to the heart, (b) the parts of the heart, or (c) parts of the ICD.
Layout, Space, and Paper
Only 5 documents, all from device manufacturers, used glossy paper. The others used uncoated, matte paper. Hospitals were least likely to use full color; the 6 documents reproduced in black and white were from this source. All of the documents incorporated white space in their design, avoiding filling up all areas of the page with text and illustration that could tire or overwhelm the reader. All but 2 used visual cues such as shaded boxes and arrows to help guide readers through the material. Contrast between the text, images, and paper was also used successfully in almost all cases.
Audience Relevance and Appropriateness
The documents did contain material relevant to their intended audienceVmainly older adults. The language and words used, although of a high literacy level, were neutral in terms of being directed at any specific audience. There was no mention of age or age-related activities, neither were specific terms relating to any one cultural group used. The accompanying images in the documents that included images of people (mainly those from device manufacturers) were culturally diverse and age specific: older adults of various ethnic backgrounds.
The patient education documents were analyzed for content. Twelve themes, listed in , were identified that reflected the descriptive content of the documents. The themes were used to subsequently analyze the consistency of and rhetoric in the document messages.
Content themes of implantable cardioverter defibrillator patient education material
Rhetoric and Language
Metaphors and persuasive language were used throughout the documents from all 3 publication sources (manufacturer, hospital, and cardiac organization). Compelling lifesaving aspects of the ICD, monitoring functions, and promotional messages were presented with strong language and exclamation marks. Manufacturers of ICD compared the device to “It’s like having a paramedic with you at all times.” Hospital-generated publications explained that “ICDs have saved the lives of hundreds of thousands of people around the world” and that the device “allows you to lead a life that is as good as or better than you could before ICD treatment.” Other statements similar to “Talk to your doctor today!” and the ICD “stops certain deadly heart arrhythmias before they kill” were found in the manufacturer and support organization documents, respectively. Furthermore, the images included in many of the documents, mainly those from ICD manufacturers, were persuasive in that they presumably represented ICD recipients living active, healthy, happy lives. Even potentially frightening or negative information was presented in a positive light. For instance, one device manufacturer pamphlet about coping with an ICD includes the heading “Pay attention to feeling alive and well.”