Data Governance: Lessons Learned from the Front Lines

Print Friendly, PDF & Email

It’s no surprise to anyone that the healthcare industry has been behind the pack when it comes to technology adoption. Many U.S. hospitals are starting to implement stronger data analytics tools as compared to traditional models, and with those implementations comes the need for strong data governance.

That was the exact case for us at Covenant HealthCare. We had just adopted a data analytics solution, but had been struggling with figuring out what the single source of our data’s truth was. For example, I would find myself asking for a certain number or metric from our analytics dashboard and end up getting four or five different numbers in response from different team members. You can imagine how inefficient – and frustrating – that can be.

To mitigate the problem, we decided to add a data governance capability to our analytics platform that would validate what the true metrics were and help restore our team’s confidence in the numbers they were receiving. Now, as I said earlier, healthcare is known to be a challenging industry when it comes to IT adoption – and a data governance capability was no exception. While we did end up seeing immense value from our data governance initiative, that’s not to say that there weren’t a few bumps along the way.

If your organization is planning to implement a data governance strategy in the near future, here are a few lessons that we learned at Covenant that may help you on your journey to improved trust and integrity in your data:

  1. The right outputs start with the right inputs – At Covenant, our primary reason for adopting a data governance capability was to assess our patient readmissions. The initial information that we received, however, seemed way off par. Our IT team went back to the source of the data – our EHR – to examine the numbers that had been entered into the system. We found that those numbers were also incorrect, and were the root cause of the governance inaccuracy. Moral of the story – you can’t have the right outputs without the right inputs.
  2. A single – and consistent – source of truth is a must – It is critical that your organization commits to creating shared definitions and measures across departments to ensure that everyone is on the same page. At Covenant, we took it upon ourselves to engage with different stakeholders and discuss what criteria we were, and weren’t, going to be using for our definitions. Not only do consistent definitions and measures boost your organization’s confidence in data, but it also ends up saving you time from having these definition debates in the first place.
  3. The culture around leveraging data must also change – The healthcare industry is at a point where we can’t solely rely on the IT team to know how to leverage data anymore; it needs to be an enterprise-wide initiative in order to be successful. When everyone across the organization is data literate, departments that were not effectively using the insights will start to make more informed data-driven decisions. You’ll also notice teams beginning to proactively ask questions about what other data elements they can be leveraging. It may be a slow transition, as with any culture shift, but the end-result will be a more data-driven, and prosperous, organization overall.

There’s no question that a strong data analytics foundation is critical in today’s healthcare ecosystem. If your organization is considering adopting a data governance capability to support your analytics efforts, keep these best practices in mind to ensure that your data is consistent, accurate and trusted across the entire organization.

 

Contributed by: Ken Arnold is Analytics Manager at Covenant HealthCare

 

Sign up for the free insideAI News newsletter.

Speak Your Mind

*