Improving bone mineral density reporting to patients with an illustration of personal fracture risk

Stephanie W. Edmonds, Peter Cram, Xin Lu, Douglas W. Roblin, Nicole C. Wright, Kenneth G. Saag, Samantha L. SolimeoBMC Medical Informatics Decis. Mak 2014 · DOI

TL;DR

This study investigated how patients best understand their fracture risk, a crucial but often poorly communicated aspect of bone mineral density (BMD) reporting. Researchers found that a vertically oriented bar graph with a stoplight color system was the most preferred and easily understood illustration, suggesting that visual aids tailored to patient preferences can significantly improve health communication.

What Problem Does This Paper Address?

Osteoporosis (OP) is a common condition leading to fragility fractures, which significantly increase morbidity, mortality, and decrease quality of life. While tools like FRAX® can calculate a patient's 10-year probability of fracture, effectively communicating this risk to patients is a challenge. Many patients do not understand their risk, which can hinder their adherence to preventative measures. This lack of understanding is often exacerbated by low health numeracy skills, where even seemingly straightforward percentages can be misinterpreted. Existing methods of risk communication, including verbal and written reports, often fall short. This paper aims to address this gap by exploring patient preferences and comprehension of different visual illustrations designed to convey fracture risk.

Methods

The study employed a mixed-methods approach. Four distinct illustrations of fracture risk were developed based on published literature and expert input:

  • FACES: A pictogram of 100 faces, with some colored red and frowning to represent risk.
  • ARROW: A horizontally oriented, directional graph with red, yellow, and green "stoplight" colors.
  • BAR: A vertically oriented bar graph with a graduated stoplight color system.
  • STOPLIGHT: A rectangular depiction integrating stoplight colors, scaled to 100% risk.

These illustrations were tested on a convenience sample of 142 adults aged 50 and older at two medical clinics in the US. Participants were shown identical FRAX® risk results (21% risk) using each illustration in a randomized order. They were then asked to rate their preference, ease of understanding, and perceived susceptibility to fracture for each. Qualitative interviews explored the rationale behind their ratings. Quantitative analyses included Friedman's test for preference ranking and logistic regression for comprehension.

Datasets

Not discussed in the paper. The study used simulated FRAX® risk data (a 21% risk of major osteoporotic fracture) to present to participants across the different illustration formats.

Key Insights and Ideas

  • Visuals Matter: The study strongly supports the idea that visual aids are crucial for communicating complex health risks like fracture probability, especially to individuals with lower health numeracy.
  • Stoplight System is Key: The use of stoplight colors (red, yellow, green) to denote risk levels (high, moderate, low) was consistently favored and understood across different illustration formats. This aligns with common cultural associations of these colors with danger and safety.
  • Bar Graphs Excel: A vertically oriented bar graph, particularly when combined with a stoplight color system, emerged as the most preferred and easily understood format. This suggests that a clear, structured visual representation is more effective than abstract pictograms.
  • Pictograms Fall Short: The FACES pictogram was the least preferred and most difficult to understand, likely due to its lack of clear risk stratification and perceived "childlike" nature.
  • Numeracy Not the Sole Factor: Interestingly, average numeracy levels were not significantly associated with preference, ease of understanding, or correct risk identification for any of the illustrations. This implies that visual design and clarity can be more impactful than inherent numerical ability in this context.

Key Results

  • Preference: The BAR illustration was the most preferred (37%), followed by STOPLIGHT (24%), FACES (22%), and ARROW (17%). FACES was ranked least favorite by a significant majority (52%).
  • Ease of Understanding: BAR was also reported as the easiest to understand by the largest proportion of participants (34%).
  • Comprehension: ARROW and STOPLIGHT were best at conveying "high risk" (72% and 62% correct perception, respectively). ARROW was significantly better than BAR and FACES in conveying correct risk perception.
  • Worrisome Factor: STOPLIGHT and ARROW were rated as the most worrisome illustrations, likely due to their direct implication of worsening risk.

Limitations and Future Work

  • Limitations:
    • The study used a convenience sample, which may limit the generalizability of findings to the broader population.
  • Presenting illustrations sequentially in a randomized order might have led to some priming or reinforcement effects.
  • The study tested four distinct illustrations, but not all possible variations or combinations.
  • The sample was drawn from a general clinic population, meaning some participants might have had prior knowledge of OP, while others did not.
  • Future Work:
    • The findings will inform the development of an intervention for the PAADRN randomized clinical trial, which aims to assess the efficacy of a direct-to-consumer mailed DXA reporting intervention using these illustrations.
  • Further research could explore how these illustrations impact patient motivation and behavior change related to bone health.
  • Investigating the impact of color-blindness on the effectiveness of stoplight color systems is also a consideration.
  • Exploring the preferences of subgroups with very low educational attainment or numeracy could provide further insights.

Summary updated

BibTeX
@article{edmonds2014improving,
  author = {Stephanie W. Edmonds and Peter Cram and Xin Lu and Douglas W. Roblin and Nicole C. Wright and Kenneth G. Saag and Samantha L. Solimeo},
  title = {{Improving bone mineral density reporting to patients with an illustration of personal fracture risk}},
  journal = {BMC Medical Informatics Decis. Mak.},
  year = {2014},
  doi = {10.1186/s12911-014-0101-y},
}
Figure