A Content Analysis of Visual Cancer Information: Prevalence and Use of Photographs and Illustrations in Printed Health Materials

Abstract

Researchers and practitioners have an increasing interest in visual components of health information and health communication messages. This study contributes to this evolving body of research by providing an account of the visual images and information featured in printed cancer communication materials. Using content analysis, 147 pamphlets and 858 images were examined to determine how frequently images are used in printed materials, what types of images are used, what information is conveyed visually, and whether or not current recommendations for the inclusion of visual content were being followed. Although visual messages were found to be common in printed health materials, existing recommendations about the inclusion of visual content were only partially followed. Results are discussed in terms of how relevant theoretical frameworks in the areas of behavior change and visual persuasion seem to be used in these materials, as well as how more theory-oriented research is necessary in visual messaging efforts.

ACKNOWLEDGMENTS

The author thanks Susan Morgan, Jake Jensen, Jim Tyler, Dorothy Teegarden, and Abigail Selzer King for feedback on previous drafts of this article. Additionally, thanks to Lisa Guntzviller, Ashley Arend, Caroline Bostwick, Brit Davis, and Scott Oliver for their help with coding. Finally, thank you to representatives of the public health agencies that assisted in providing materials for the project.

Notes

¹ The four organizations were selected because each offered materials on a wide variety of topics and cancer sites, utilized advisory boards in some capacity in the creation of some or all of their materials, were assumed likely to use best practices, given that most of the organizations fund or participate in research to improve the production and reception of printed materials, and offered the materials to the general public for free either online, through local outreach offices, or by ordering over the phone.

² It is worth noting that during this article’s review process, research published by Bol and colleagues (2013) examined cognitive versus affective illustrations on website satisfaction and recall of cancer-related information. For interested readers, image content of people and objects would likely be considered affective illustrations, whereas data-oriented content would likely be considered cognitive illustrations.

³ Image size was measured by recording the length and width of each image in millimeters (mm). For images that were not square, coders were instructed to sketch straight lines, like a frame, around images. Length and width values were multiplied to determine the total surface area of each image (mm²). Reliability was high initially (Krippendorff’s alpha = .98) and at the midpoint check (Krippendorff’s alpha = .98).

⁴ Coders measured the length and width of each panel in millimeters (mm), and then squared those numbers to determine the overall surface area of each panel (mm²; Krippendorff’s alpha = .97).

⁵ Number of panels referred to the total number of faces within each item. An information sheet would have two panels (front and back side of one piece of paper). A traditional pamphlet would have six panels (one piece of paper folded twice). A booklet with 20 pages would have 40 panels (front and back side of 20 pages). Intercoder reliability was high (Krippendorff’s alpha = .99). Pamphlet-level coders also counted the total number of panels that contained images. Reliability was also high for this dimension (Krippendorff’s alpha = .99).