By Brian J. Zikmund-Fisher, Ph.D.
Brian J. Zikmund-Fisher is Associate Professor of Health Behavior and Health Education and Research Associate Professor of Internal Medicine at the University of Michigan, as well as an Associate Editor of the journals Medical Decision Making and Medical Decision Making: Policy and Practice.
Today, the problem is NOT having too little data. In many contexts, the problem is that we have too much data, yet too little understanding of what those data mean.
Think about the breadth of types of health numbers that patients interact with (without even getting into the data that providers face). An online search turns up dozens of risk calculators that can estimate your risk for various diseases to a fraction of a percentage. The results of every laboratory test are now available on your hospital’s website. Every piece of food you buy offers dozens of nutrient values (e.g., fat grams, % of daily recommended vitamin intake, calorie counts) that we’re supposed to pay attention to in order to be healthy. Our activity trackers feed back to us steps and sleeping hours.
Are we more informed? Or more overwhelmed?
I’d argue the answer is often both.
For most of these types of data, the core problem is one of information evaluability. We may know what the number is, but we don’t know what it means without context.
The problem is: While technology is increasing our access to data, comparatively little attention is paid to designing these systems to provide the necessary context to support meaningful use.
For example, look at laboratory test results. Right now, if a patient looks up their test results in a patient portal to an electronic record systems, most likely they will see a table. A table that shows the test name (with no explanation), their result number and some units, and a “standard range” (with no explanation for what “standard range” means).
That’s it. In most systems, patients do not even get any marker (e.g., “L” or “H”) to tell them whether the result is inside or outside of the standard range. Is it surprising, then, that many people cannot figure out whether their result is within or outside of the standard range?
More recently, I have been leading an interdisciplinary team in designing visual displays of test results that use a variety of cues (colors, labels, visual spacing) to providing meaningful context. These types of displays could be incorporated into online portals to enable patients to make sense of their results. For example, our first paper showed that using these displays instead of tables can help patients become less concerned about test results that are slightly outside of the standard range but not actually clinically concerning.
The health data gathered and delivered by modern health technology has huge potential to improve the lives of patients and the efficiency the healthcare system. But, we have to know what it all means in order for society to reap those benefits.
