Healthcare Data Visualization: A Clear Picture of Actionable Insights

Today’s healthcare reporting tools have incredible powers to tell stories about patient health—whether individual patients or entire populations. The same tools can also convey immediate facts about which purchases to make, drugs to use, physicians to assign—and a great deal more. Yet the present reality is that not many physicians use these tools to such an extent. There are several factors that can affect their ease of use, starting with acquiring the actual data. Read on to learn more about these and other hurdles to widespread use of data visualization tools in the clinical practice—and the way forward to clearing these barriers.

Collecting and directing the flow of information

It would be nice if physicians, researchers and other clinicians could just shepherd the needed data into a simple dashboard and quickly go about their business of curing the world of what ails it. Obviously, it isn’t quite that simple (yet).

Especially if some of the data comes from outside sources. There must be a level of trust in the quality and security of the receiving organization’s health data management systems for the data owner to be willing to share it. No one wants to risk a HIPAA violation at the hands of another entity.

Turning different data formats into one for all

Once the data has been acquired it must be made usable. And here’s where the time-tested computing principle of “garbage in, garbage out” applies. Data must be scrubbed, normalized and aggregated into a standard format all can view and manipulate.

And really, almost all data will need some sort of work in this area. Even something as basic as showing provider locations on Google Maps—addresses typically come in with different abbreviations and other formatting distinctions. It’s a mundane detail, but one that matters.

For clinical visualizations, much of the needed information may sit in unstructured notes rather than structured data fields. Organizations attempting to use that information to build visualizations are often flummoxed by how to do so.

In a related issue, deciding how to structure a report can be a time-consuming endeavor in itself. With so much data available it’s easy to get carried away, ultimately delivering a monster of a report that ends up burying the key insights under piles of factual, but irrelevant information. The larger the forest, the more difficult it becomes to find a specific tree.

Presentation. Data visualizations must be easy for business users to access. The most innovative, most well-designed dashboards in the world will be of little use if they are not formatted for the screen business users typically use to view them. That goes for mobile device users, too. While it may not be feasible to design every aspect of a report interface around a smaller screen, a version for iPads and other portable devices is a huge advantage in an increasingly mobile healthcare environment.

On that note, this is a good place to suggest some additional best practices for data visualization.

  1. Make the reports/visualizations relevant based on the user’s role, identity and concerns. Each set of users—clinical, financial, executive, IT, marketing, etc.—requires different metrics. The more specific the visualizations are, the more actionable they will be. In many cases, business users will be looking to answer a specific question or set of questions. For example, management might want to understand the financial and clinical impact of standardizing on a particular brand of knee replacement components. A visualization designed to answer that question can demonstrate the risks, rewards and success rates of various components currently being used to help the organization make the right decision.
  2. Begin with the end in mind. This may seem an obvious piece of advice, but be sure to communicate with business users what they need to see or want to accomplish in advance of structuring the report. And as a general rule, aim for no more than three to five key performance indicators.
  3. Make visualizations easily accessible by users. Circling back to our observations about today’s mobile healthcare landscape, this is especially important for physicians and nurses who are constantly on the move. While they may occasionally be at a PC or laptop, today there is a high likelihood that they will be using a tablet or smartphone to access the dashboard/report. Incorporating responsive design techniques that recognize the type of device being used will generate far better results and satisfaction than a “one size fits all” approach.
  4. Make sure you are HIPAA-compliant. It will be far easier to obtain data from outside data sources if you can demonstrate that it will be well-protected within your organization—in storage, in transit and in the way it is presented. This may mean restricting who can access certain types of data for visualizations. It almost certainly means de-identifying patient data before it can be used or shared for certain purposes. Whatever the requirement, having a strong data management mechanism in place is critical for protecting both partners and your own organization.
  5. Create reports that can lead to action. Consider a hospital that wants to reduce its readmissions. A dashboard might show readmissions rising, but unless someone looks at it, this fact will go unnoticed. If email alerts are incorporated when readmissions cross a certain threshold, however, notifications can be sent to designated people who can review the dashboard, drill down to the information behind it, and initiate actions to correct the situation. Alerts can also be as simple as a warning “light” at the top of an EHR cautioning that a patient has been given the wrong medication and the situation needs to be addressed immediately. The more information that can be acted on, the better.

Help in the cloud

There’s a place where many of the mechanisms for aggregating and normalizing a broad spectrum of data exists: the HIPAA-compliant, cloud-based data management center operated by a proven provider.

First and foremost, such a facility will have the security measures in place to alleviate the trust issues among different data partners. It also offers the psychological advantage of neutrality. Other organizations may feel more comfortable uploading their data to the provider than sending it directly to other entities involved in the data project.

What’s more, if you don’t already have internal data visualization capabilities, a good cloud services provider will have partnerships in place that can make this available to you as part of your plan—from planning and design, to delivery. After implementation, ongoing services are available that address any number of security aspects. In a facility exclusively focused on healthcare data, this will include patient de-identification and other HIPAA privacy requirements.

Taking action

There is little doubt that convenient, relevant visualization reports and dashboards will be heavily used, with the potential for previously unattainable results. Thanks to new capabilities in cloud services, those results are closer at hand than ever.