Using Data to Create Healthcare Narratives

I mentioned in a previous post that “important stories live in our data”.

Particularly in healthcare where tons of data abound.

The problem is that healthcare is not producing enough good stories to support decision-making in the industry.

Instead what we have is an explosion of reports, spreadsheets, charts and tables such as this:

For most people this kind of data is useless and invites lots of questions: What is this? What is happening? Why? What do we do with it?

But if you can translate it into a narrative or make it visual so that people are able to understand it then it becomes useful. Even physicians and decision-makers in the industry could use easier ways to understand this kind of data.

Something like this infographic below makes more sense (not related to above data-set):

Use visualizations not data to communicate important messages

This infographic provides a clear narrative about childhood injuries. The information is easily understandable, usable and actionable. There is no danger of jumping to the wrong conclusion.

But healthcare organizations continue to produce a ton of data hoping to glean meaningful insights that will help their decision making. Even when their own employees don’t have the skills to translate the data. Hence mountains of data heap up without a clear direction as to how it can be used.

Data Democratization

I don’t think you have to be a data scientist to handle data. Technologies such as Google Fusion, Adobe Illustrator, Tableau and others are empowering all kinds of people to interpret data and present it to others.

In fact I recently enrolled in the Knight Center of Journalism’s second massive open online course (MOOP), ‘Introduction to Infographics and Data Visualization.’ The course is taught by Prof. Alberto Cairo, author of ‘The Functional Art.’ We’re learning how to create data visualizations and to discover hidden stories from data.

This week for example I learned a simple approach to storytelling through data from Andy Kirk’s book, Data Visualization: A Success Design Process (A great resource for data enthusiasts!). Here’s the outline:

#1. Scan the data

Let’s say you’re presented with the following table of data (click to enlarge) and are asked to create a story for your audience. The table is titled, ‘Frequencies of Circulatory Diseases Among the U.S. Adult Population’:

The first thing you would do is to scan it with your eyes and see if anything interesting stands out. For example I see that hypertension is the most common type of circulatory disease in the United States.

#2. Examine the data

Next try to become more familiar with the data by examining some descriptive and statistical elements. For example there are two main variables in the table above. One for Selected Demographic and the second for Frequency of Selected disease. Also there are four demographic categories – age, gender, ethnic group and level of education. This exercise helps you to see important attributes in the data.

#3. Contextualize the subject matter

Prepare the data in a way that gives more insight. For example you could find some co-relations and do some calculations to produce percentages of frequency. In this case you discover that 25% of U.S. adults suffer from hypertension. At this point you may (or may not) start to sense the beginnings of a story.

#4. Develop editorial focus

To start shaping the story ask yourself some basic questions about the data. ‘What initially crossed my mind when I first saw the data? What kind of trends or patterns do I see? What kind of analysis might be relevant to my audience? Let your imagination roam and seek ideas that will make your content more interesting.

#5. Find a visual representation

Next you will need to focus on a data visualization that will answer these questions. It could be a bar chart, a bubble map or an infographic like the one above. The key thing is to create a visual from which the audience can perceive a story easily, efficiently and accurately.

Quick Wrap-up

No body wants to waste time pouring over meaningless data.

Time is money and (healthcare) decision makers have little of it. For them insights gained from data are more beneficial than the data itself. I believe they should explore available technologies (such as those mentioned above) to empower their employees to work with data. I know that stories resonate and visuals stick. This is one way to turn all that complex data into something people can use to make important decisions.

Your Turn: How is your organization using data to help fuel decision making? Are you first translating it into narrative form? Please share your views in the comment box below.

Leave a Reply

Your email address will not be published.