PS 4-7
DESIGNING DATA VISUALIZATIONS TO SUPPORT SHARED DECISION MAKING ABOUT BLOOD PRESSURE
Recent research has identified the important role of home blood pressure measurements, considered at least equal to clinic blood pressure (BP) measurements in their predictive value (Pickering et al., 2008). However, raw home BP data may currently be underutilized—due to the potential for information overload—as there exists a lack of clarity regarding how best to integrate this data into the shared decision-making process. This may also be a contributing factor to the phenomenon of clinical inertia. Current EMR systems lack adequate capabilities for informative visualizations of BP data. Our goals were to determine: (1) the key information for both patients and physicians when discussing patient BP data and (2) the best practices for the display of that data.
Method:
We used a rapid prototyping process in which candidate displays were iteratively refined based on regular feedback from patients and physicians. Several focus groups—three with hypertensive patients and three with family and internal medicine physicians—were held at primary care practices at an academic research hospital system in the Midwest to gather feedback on candidate displays.
Results:
Based on transcriptions and materials generated by focus group participants, several important design elements were identified. Displays needed to be designed to present both home and clinic BP measurements, with features to easily distinguish between the two data sources. Design considerations were made to allow for rapid assessment of whether BP measurements are within goal range; however, additional flagging of values outside of goal range did little to further improve understanding. Both patients and physicians wished to annotate the display and several designs were created to accommodate for this while minimizing display clutter. A medication timeline (http://inspiredehrs.org/timeline/) was introduced to allow for the visualization of the effect of medication on BP measurements, which both groups found intuitive. A sample of these features is shown in the accompanying figure. Several other considerations, such as providing patients and physicians with summary statistics of patient BPs, and the development of these various data visualization features will be presented.
Conclusions:
Understanding how stakeholders evaluate BP data visualizations and which factors help simplify and improve understanding allowed for the development of a simple approach to the visualization of home BP measurements and potential improvement to the shared decision-making process.
