3 Phases of Healthcare Data Governance in Analytics
Healthcare data governance. It sounds like it could be a dry topic, right? The truth is that data governance is a very exciting and complicated challenge in healthcare analytics. The purpose of this commentary is to:
- Show the challenges involved with adaptive healthcare data governance
- Demonstrate how effective data governance evolves throughout the lifecycle of a health system’s analytics program
- Describe the three phases most healthcare organizations pass through as they establish an effective system of data governance
What Data Governance Is
First, I want to quickly address what data governance is by using a comprehensive definition from Gartner:
Data governance is “the specification of decision rights and an accountability framework to ensure appropriate behavior in the valuation, creation, storage, use, archiving, and deletion of information. It includes the processes, roles and policies, standards, and metrics that ensure the effective and efficient use of information in enabling an organization to achieve its goals.”
Pretty broad, right? The scope of data governance includes data stewardship, storage, and technical roles and responsibilities. But it also requires leadership and processes to get the most out of an investment in analytics. As such, healthcare data governance is similar to an analytics implementation because it requires the same type of leadership buy-in and processes.
Because leadership buy-in is so important, I’m going to focus on the important role a core group of leaders — the key stakeholders who form the healthcare data governance committee — will play in setting, growing, and sustaining a successful analytics program.
If you are a VP or director-level analytics/business intelligence (BI) leader, you understand how important it is to ensure the right group of leaders is engaged in the governance of the BI program as a whole. The best data governance committees help to ensure the efficient, equitable use of a health system’s information, and enable the organization to achieve its goals of higher-quality, lower-cost care.
Effective Healthcare Data Governance Is Adaptive
Data governance is hard and frankly, very few health systems feel they have every aspect of effective healthcare data governance figured out. One of the reasons it is so difficult is that no single template for data governance can be applied to every organization. Data governance approached as a rigid, idealized plan — outlined at the beginning of your analytics implementation and unchanging despite the volatility of the healthcare environment — often ends up being scrapped.
But before we get too deep into healthcare data governance, I want to make it clear that I don’t consider myself an expert on the topic. Instead, I am deeply grateful for the opportunities I have had to work with some truly incredible teams, for both health systems and as a vendor, because of the perspective they’ve given me. And through the years of working with these teams, one thing has become clear to me: an adaptive approach to data governance is best suited to healthcare. I say this because I’ve had the opportunity to see the success that healthcare organizations have experienced by adapting data governance to fit their needs.
What I want to share with you now are the three phases of adaptive data governance that characterize successful analytics implementations. But rather than outlining a detailed framework for you, I’ll describe who should be involved in each stage of data governance, and what those individuals should focus on during each stage.
1. The Early Stage of Healthcare Data Governance
When starting the early stages of planning for adaptive data governance, it’s useful for the sponsor of the health system’s data governance initiative to have some authority within the organization.
Ideally, these leaders are those who have a passion for using data, and whom are generally described as being both exciting and decisive. Official “HR” titles matter less than credibility and effectiveness in their role. They may be executives, directors, influential managers in the quality department, nurses, or physicians. They are the people who can serve as a catalyst for moving the analytics initiative forward in the organization. Importantly, they should represent both analytically “mature” areas of the business (like finance) and chronically underserved areas: the people who have been crying out for more and better access to data.
The personalities and temperaments of these leaders are important factors in the success of this early data governance committee. They will need to make important decisions based on limited information, so they must be comfortable with that. They need to be able to deal with ambiguity. Importantly, they have to be willing to let the analytics initiative evolve quickly and organically instead of clinging to preconceived, grand visions of perfect governance.
One of the most important questions this group will address at this point is this: where should we start with our analytics initiative? Where should we focus our resources first? For example, the group may decide to begin by implementing analytics in Cardiology rather than in Labor and Delivery. In addressing this question, the group’s main role will be to keep the peace, and they must make sure everyone in Cardiology understands what is being prioritized and why. They must also gain buy-in from those who will be affected by the initiative.
The group will have to ensure the relevance of the initiative by communicating in terms of business and clinical decisions rather than in technological terms. It is their job to protect the integrity of the initiative: to make sure they aren’t just building an impressive technological infrastructure but something that will drive real quality and cost improvement.
It’s important to note that at this point in the data governance process, it may be too early to bring up issues of data quality or be too dogmatic about data stewardship. It’s also possible that no one will listen quite yet because not enough people are relying on the data for it to be a priority for anyone but the governance team. When you reach a certain critical mass of adoption, it will be time to address these issues.
2. The Mid-Term Stage of Healthcare Data Governance
As your implementation of analytics continues, the nature of the decisions made by the healthcare data governance committee will change. Fewer “big” decisions will be required once you have a structure in place, and the program starts to operate as a system. The decisions made by this committee are still vital, but not as sweeping in scope. For example, instead of the decision to work first with Cardiology over Labor and Delivery, the committee will decide whether to work on heart failure or heart rhythm disorders within the cardiology group.
It’s possible some of the initial data governance committee members will lose a little bit of interest at this stage, and may find reasons to decline meeting invites. Which makes sense: many of these initial members tend to be passionate, entrepreneurial, visionaries who like to get projects started but who view the maintenance of the program as more of an operational problem. It’s fine to let them walk away — the organization may need their drive elsewhere to function smoothly. The key is to work with them to figure out who the right person is to take their place and help move analytic governance to the next phase.
At this point, the committee will begin to implement the solution in new areas of the enterprise. For example, if Cardiology’s needs are being met, you could branch out to serve Labor and Delivery.
The group will also be responsible for monitoring the progress of existing initiatives. They will ask questions like:
- How are things going?
- Who is using the system?
- What additional training or tools are needed to increase utilization?
- What should we keep doing/stop doing/do more of?
Finally, the group will be actively planning for the next phase of the initiative’s development. Now, utilization is typically increasing but the demands are still manageable. It’s a good idea to use this “honeymoon” period to plan for the coming deluge of requests for more content and functionality. Strategy discussions will revolve around topics such as:
- Prioritization: when we start getting new requests for additional data content, how should we make good decisions?
- Staffing: do we want to align resources with any specific areas of the business?
- Communication: how will we continue to communicate with those currently using the system?
- Data quality: how do we ensure a good strategy for data stewardship and management? What would be the process for communicating with the frontend systems? How do we inform them of the problems we’re encountering and drive change?
3. The Steady State of Healthcare Data Governance
In this stage, the data governance committee may want to start calling itself something more interesting, like a governance council. Let it happen. Only a few of the initial leaders will still be involved, but the council itself will continue to evolve.
The most effective agents of governance at this point are those who are the “ethical politicians” of the organization. These are the people with a talent for keeping the peace and building consensus. Much of their work will now be done outside of official council meetings. They will have a natural, one-on-one style of leadership that allows people to approach them in the hallway and air their concerns. At this point, they won’t be solving big problems in real time; rather, they’ll be addressing common concerns that they hear frequently.
Let’s face it, if the analytics effort is going well, there will be dissatisfied customers. That sounds contradictory, but it’s true. Some customers will be dissatisfied because they haven’t received as much support as other areas in the organization. Leaders’ role will be to hear these customers’ concerns, make their voices heard, help them feel there’s hope for improvement, and then keep them in the loop until their concerns are addressed.
An important role of this group is simply to stay the course. This is the phase of the project where groups that aren’t getting what they need may be vocal and call for a different approach. But if the analytics initiative is working for most of the organization, this group will have to actively work to pull any disenfranchised constituencies back from the ledge. This is where the political savviness of this group will be enormously helpful in allowing the technology team to focus on their work. The last thing the initiative needs is for the technology team to get wrapped around the political axle.
Finally, as in the previous stage, this team will continue to be responsible for monitoring the success of the initiative and prioritizing efforts. This prioritization will be more crucial now than ever, because there will not be more demand for data than the technology team can possibly address. The governance council will have to decide where the need is greatest based on the organization’s goals.
How to Know If Healthcare Data Governance is Working
To know if data governance is working, explore using data to measure the effectiveness of your data governances if you can. Here are some of the key metrics you could look at:
- If governance is successful, these metrics will trend up:
- # of users
- # of requests
- # of queries/reports/page views
- # of success stories from your users!
- # of key stakeholders who are aware of your group and what you do
- These metrics should stay solid, or trend upwards in rare cases:
- User/customer satisfaction.
- Technology/analytics team satisfaction
- These metrics should hopefully trend down:
- Time and resources it takes to answer common analytic questions
- # of requests you get to evaluate competing analytic systems
At various points in the lifecycle of an analytics initiative, you may be tempted by the simplicity of a “one size fits all” framework for data governance. In my experience, healthcare data governance is most effective when it is allowed the flexibility to adapt and change. If you are prepared to allow your data governance strategy to evolve as analytic maturity develops, your organization will be well positioned for success.
I’d love to hear from you — how have you seen data governance adapt? Have I missed something about adaptive healthcare data governance that you’ve found useful at your health system?
Would you like to use or share these concepts? Download this healthcare data presentation highlighting the key main points