Healthcare Data Stewardship: The Key to Going from Information Poor to Information Rich
But how can an organization with terabytes of valuable data waiting to be tapped become information rich? There are two key components.
The first component is having ready access to the data. One of the best options is by using an enterprise data warehouse that has been specifically designed to work with a healthcare analytics system. In addition to ready access to the data, there also needs to be accountability for that data. This leads us to the second component — good healthcare data stewardship.
What Is Healthcare Data Stewardship?
Data stewardship is the aspect of healthcare data governance that focuses on providing the appropriate access to users, helping users to understand the data, and taking ownership of data quality. When there’s poor data stewardship, however, even the best infrastructures become underutilized and poorly understood by knowledge workers who could be generating value from the data every day.
Who Should Be a Healthcare Data Steward?
Can you name an organization where one person knows all the different systems in use and can speak with authority on the data within those systems? I’d be willing to bet you can’t. The point here is that, because no single person can know everything about every system, it’s best to have many data stewards — at least one for every major source of data in the organization. This may seem like a lot of data stewards, but the best person to perform this role is the person who knows the most about how and why the data is collected in each of the various source systems. One person simply can’t know all of this.
2 Tools a Data Stewards Needs to Be Successful
Once you’ve identified who should become a data steward, there are a few things they will need to succeed.
1. Enterprise Data Warehouse
First, make sure there’s a good data warehouse architecture. This will promote good data stewardship because it will eliminate the fragmented data systems.
In a fragmented IT environment, there are many diverse data systems and each system has different modes of access. Some may have ‘backend database access’ and some may have reporting platforms of their own. The trouble is, with so many different systems and modes of access, it is extremely challenging to manage access centrally. This is why the job usually defaults to the individual IT administrators responsible for maintaining the systems — the only ones who know how to give access. Unfortunately, these folks typically have about a billion other tasks in their work queue with higher priority than providing access to a data system to a user who they don’t even know.
A fragmented IT environment also makes it challenging to manage an organization-wide data dictionary. Some healthcare information systems do a good job of documenting metadata — the underlying data that communicates the origin and other details about data, much like a dictionary entry — while others do it very poorly. Having a centralized place to search for data elements and browse data structures is next to impossible in a fragmented IT environment.
A centralized enterprise data warehouse delivers a single reporting platform and a single point of entry into an organized repository of the organization’s data. The standardization of technology allows the development of more centralized access policies. Good data warehouse architecture allows a data steward to own a piece of the data warehouse that maps to their specific knowledge about a specific source system or a specific data mart created within the warehouse. This is reflected in an architecture that allows the creation of ‘source marts’ and ‘subject marts’ and allows access to be granted on the individual data mart level. The data steward is an expert in a particular data mart and is the appropriate person to decide who gets access to that area in the enterprise data warehouse.
2. Data Warehouse Tools
Once you have identified data stewards, you need good tools to let the data warehouse user community know who these data stewards are. If a user wants access to the data warehouse, how will they know who to go to for access to a particular area of the warehouse? How will a user understand what data is in a particular table, or who to go to if there is a quality issue with the data? A solid metadata tool will list data marts in the warehouse and link those data marts to data stewards. The tool will allow a data steward to share their knowledge about the data mart or marts that they are responsible for. Metadata tools that allow data stewards and users to create and edit descriptions for objects in the data warehouse provide this forum and will aid in creating a self-sufficient data warehouse user community.
Create Long-Term Value with Healthcare Data Stewardship
Once the data is unlocked in an enterprise data warehouse, the best way to create long-term value is to have a thriving user base. Data stewards are critical partners to the data warehouse team in creating a thriving user base. They are the data librarians who advise and guide users, and help them get the most value out of the enterprise data warehouse.
Users of the data warehouse will create value in different ways. Some may use the data to secure grant funding, others may use the data to squeeze extra dollars in savings through operational improvements, and still others may use the data to improve patient care and reduce costs. If more users translate to more value … make it easy to do the right thing and get these data-hungry analytics superstars connected to the data!
I’m interested in hearing about your healthcare data stewardship policies. Are you part of a health system that’s data rich but information poor? If so, have I missed anything that would help transform your system?
Would you like to use or share these concepts? Download this Why Healthcare Data Warehouses Fail presentation highlighting the key main points.