About This Project

When it comes to our bodies, data abounds. We all have a blood pressure, weight, cholesterol levels, A1c, BMI, and more. We have risks, too. We might have or be at risk for cancer, or heart disease, or have a higher risk of experiencing a side effect of a medication or treatment than someone else.

In theory, this data can help us make better decisions about our health. Should I take this pill? Will it help me more than it hurts me? How can I reduce my risk? And so on.

But for individuals, it’s not always easy to understand what the numbers are telling us. And for those communicating the information – doctors, hospitals, researchers, public health professionals — it’s not always clear what sort of presentation will make the most sense to the most people.

That problem is the inspiration behind Visualizing Health, a project of the Robert Wood Johnson Foundation and the University of Michigan Center for Health Communications Research. This site contains 54 examples of tested visualizations – that is, graphic displays of health information that we’ve evaluated through research among the general public. Our objective was to create a gallery of beautiful and easy-to-make-sense-of graphs, charts, and images that effectively communicate risk information. Health data that makes sense.

These visualizations are distributed via a Creative Commons license, which allows anybody – academics, healthcare organizations, even for-profit businesses — to adapt them for their own objectives. Please use them – and tell us how you’re using them by emailing us at vizhealth@rwjf.org.

Here’s what we’ve developed:

How to Use This Site

The central tool of Visualizing Health is our Wizard. This tool lets users choose images based on various concerns or needs. Not all graphs work for all purposes. The graphical style you choose will probably depend on the goal of your risk communication. So we've built the Wizard to help clarify your goals.

The Wizard will ask you a few simple questions and then it will lead you to the category of risk graphics that most closely matches yours.

Note that the Wizard requires users to identify a primary goal: the main point of your risk communication.

How We Created Our Images

We started with about a dozen different, but common, kinds of risk communication problems – scenarios where an individual might be faced with health data. We call these “use cases.” Among them:

  • Tables of side effect risks
  • Translating test results into risk
  • Visualizing health scores
  • Racial disparities in rates of disease
  • Putting outbreaks of disease into context
  • Visual displays of side effect risks
  • Personally tailored data about side effect risks
  • Risk calculator: More than a number
  • Showing how side effects change over time
  • The benefits of risk reduction
  • Years of life saved by taking a drug
  • Icons to show severity of side effects

A summary of the use cases is available.

We developed these into specific scenarios and sent them out to four teams of data designers who proposed several concepts for each use case. Our researchers evaluated these visualizations for accuracy, but the style and approach was left to the artists.

We then began testing the visualizations to see which ones made the most sense to ordinary individuals – not health professionals. We used three tools to test our images.

Google Consumer Surveys

is a new service for collecting data online. It can be a very quick (days, not weeks) and inexpensive for getting small amounts of data on specific questions. If you only need to ask one or two questions, you can get that data for a fraction of the cost of other online surveys.

Survey Sampling International

is an online survey panel that includes millions of people who are willing to take surveys. Qualtrics is an online survey tool that enables you quickly develop online surveys. We programmed and posted a survey in Qualtrics, and then used SSI to recruit people who have particular characteristics.

Compared to GCS, it takes a little longer to get data, but we get more data because SSI participants take longer surveys.

Amazon Mechanical Turk

MTurk has people ("Workers") who agree to be part of MTurk and do small jobs (HITs, "Human Intelligence Tasks") for others who need it done ("Requesters"). In return, they get paid.

The fact that MTurk participants treat surveys like jobs is both good and bad. It means they do their best to answer our questions. But it is not a true proxy for how people might react in real life.

How We Identified the Best Designs

wizard Janus is trying to choose the best visualization.

Each survey instrument provided results that indicated how well these different audiences were able to comprehend the information contained in the visualizations. Our researchers compared across each survey tool to establish how each image performed in the tests. All of the images we tested are included in our gallery section. We include them all because seeing a range of different designs can help you to recognize key design features. However, sometimes certain images performed demonstrably better than other images in our tests. When that happened, we have put a star on one or more of its labels so that you will know on what dimension that graphic did particularly well. If you want to know more about our testing methods and results, you can download a testing summary document from each image page that describe our methods and analysis in detail.

We also adapted some extra images based on the winning designs, so you can see how these concepts might look in different contexts.

Who We Are

Visualizing Health is a project of the Robert Wood Johnson Foundation and the University of Michigan Center for Health Communications Research. It was conceived by the Foundation’s Entrepreneur in Residence, Thomas Goetz, in collaboration with RWJF Program Officer Andrea Ducas.

At the University of Michigan, the overall project principal investigator was Victor Strecher, PhD. Research activities were directed by lead investigator Brian J. Zikmund-Fisher, PhD, with input from a multi-disciplinary research team that included:

The design effort was led by Tim Leong, with visualizations by the Italian-based firm La Tigre, the Spanish-based firm Lamosca, Luke Shuman, Lauren Manning, Walter Bauman, and Jan Avendano.

The website was designed by the University of Michigan Center for Health Communications Research: