Clinical Data Warehouse: Why You Really Need One
I’m hearing that question a lot these days – is data warehousing in healthcare really necessary? – from both CIOs and IT directors. It usually follows a presentation or a panel discussion. And it’s not that the IT professionals asking the question don’t perceive value in a healthcare data warehouse. It’s usually that their organization already has other tools in place that come with some degree of data integration capability. “Isn’t what I have already enough for what I want to do?” tends to be the next question.
These are great questions, important ones, too. I’m excited because these questions mark a shift in the way hospitals are thinking about their data. Today, physicians, nurses, executive leadership, and state and federal agencies are sending the message to healthcare CIOs that data IS important, that measurement and key performance indicators are vital, and that their organization needs to be more agile, flexible and fluent with its data.
But still, from an IT side, those “Do I really need …” questions hint at a reluctance to employ the right healthcare enterprise data warehouse solution that to achieve the organization’s goals. And that’s something I’m hoping to change.
What Happened to the First Wave of Clinical Data Warehousing?
It wasn’t all that long ago that data warehouses were cutting edge. They were the original “big data.” They spawned an entire industry of extract/transform/load (ETL) tools. All the cool kids were getting them. So what happened and why aren’t they everywhere in healthcare?
- Because many healthcare organizations tried. But their data warehousing projects didn’t deliver the value that was being recognized in other industries, including retail, manufacturing, etc.
- Healthcare CIOs knew they had plenty of tasks to accomplish and a growing list of vendor partners who were ready to help them with those tasks. Taking the time to build a clinical data warehouse didn’t have the perceived ROI of tackling tasks one at a time.
In all honesty, we should have seen the lack of healthcare-analytics-tool adoption coming. Few individuals − and no vendors − had figured out a way to reliably integrate the cornucopia of data sources in healthcare, and the healthcare industry lacked the right partners to supply the people, the process and the technology to do it right. Until now.
Clinical Data Warehouses Enable Healthcare Analytics
Healthcare business intelligence tools are a great way to get a rudimentary level of data integration functionality. But today’s technology, data, and even regulations scream for more – an analytics solution. A Healthcare EDW can offer this.
Here is what a hospital should be looking for:
IT teams are fielding more requests than ever to use the data that their hospital already collects. That makes reporting requirements paramount in an analytics solution. The solution chosen should include:
- Tools that measure performance to meet the following conditions:
- Chief Clinical Officers’ requests for monthly summaries of their health system’s value, defined as outcomes per dollar spent.
- Repeatedly and reliably delivered information that combines clinical, financial performance, quality, cost, and patient experience data.
- Highlight your organizational performance relative to peers and national benchmarks.
- Tools that efficiently extract critical data currently locked in EMRs, claims and billing systems in order to make data acquisition repeatable.
- “Bullet-proof” tools that automate the integration of disparate data sources so a hospital’s most important asset – its people – spend their time analyzing rather than acquiring its second most important asset – its data.
Organizations have a choice about where in the technology stack they place the integrated view of data. Reporting tools often tout this capability – but then the data model, data transformation business rules, or sometimes even the data themselves are locked in the reporting layer, which may not be the wisest place.
Consider this: more and more powerful data visualization tools become available every year. When down the road you want to leverage a different tool in your BI layer, you may need to plan for a significant capital investment if you already made heavy use of your current BI tool’s data integration functionality. This is because migrating the data, business rules, sophisticated calculations, and semantic information from one BI tool to another is (human) resource-intensive. If your organization seeks flexibility in its reporting or BI tool choices, keeping this layer de-coupled from your integrated view of data gives you more options.
A more flexible architecture – and the one that we recommend – is to push the integrated data to a lower level of the “stack.” In this case, that’s the relational database. Relational databases implement tried and true technology that has been around for 30+ years. They provide a lot of the following flexibility for an organization to work with:
- Want to migrate to a different relational database platform in about 15 years? Probably won’t be a big deal. The next database will also have tables and views, just like they’ve have had since the early 1980s.
- Want to invoke a Big Data tool in the future? No problem – most database vendors in the marketplace have Big Data on their product road maps. Big Data plays well with relational databases. Both approaches find it in their best interest to complement each other.
- Want to layer on a different, more capable, less-“spendy” BI tool in the future? Adding on is pretty simple, too, when integrated data is stored in a relational database.
Very likely, an organization’s data will outlive its choice of reporting tools, so storing integrated data in the database layer is a way to design for the future. Organizations have to execute for today while also planning for tomorrow and the relational database has been proven successful in numerous industries over several decades. It’s adaptable, flexible, and it makes good architectural sense.
Getting Healthcare Data Throughout the C-suite
CEOs have an intuitive sense of what they need from their data: quality, cost, patient experience, and revenue information all in one place. Savvy leaders use broad views of their business’ data to better understand how moving a given lever within an organization could affect something else. But it’s still not uncommon for a healthcare executive with the vision to steer his or her organization up the Healthcare Analytics Adoption Model to be lacking consistent access to the breadth of information needed to truly realize that vision.
This, fortunately, is a very solvable problem. A growing number of health systems consistently give their CEOs the ability to quickly see across their organization’s data and make informed decisions. These companies may have different reporting tools, but the best out there rely on a robust, comprehensive, and accessible clinical data warehouse platform.
Take, for example, a clinical data warehouse developed with a Late-Binding™ architecture, which we at Health Catalyst believe is the right tool for the job. First, it’s been proven and tested at multiple healthcare organizations. Second, it has a track record of rapid time to value. And, third, it has the ability to address the demands of accountable care organizations.
Health Analytics Solutions for Everyone
Back to the question: “Do I really need a data warehouse for my clinical operation?” When I hear this, I know this organization wants to access its data, and that its entire team – from clinicians to executives – love what they do, understand the difference they make in their patients’ lives, and are passionate about providing the highest quality patient care.
That one simple question tells me the organization wants tools that are integrated, views that are consolidated, information that sits at their fingertips, all of which will help their organizations thrive and grow and their patient care to be the best possible. They want a broad view of their hospital’s data and they want it in a way that’s efficient to access and easy to use.
My answer isn’t just yes. It’s “You need a solution that gives you near real-time answers to your questions. You need a solution that gives you flexible access to data so it can be sliced and diced in numerous ways. You need a solution that allows you to maintain your data’s integrity and use it wherever, whenever and however your organization. You need a late-binding clinical data warehouse and analytics applications that empower you to make informed decisions.”
And that’s why we invariably hear, from our customers, a slightly different question than the one posed in this article:
“How did I get by so long without this?”