My journey into the world of clinical quality measures began seven years ago…and it continues today.
At the time, in 2011, my organization had just received a grant to improve care and reduce cost for people with chronic obstructive pulmonary disease (COPD). The terms of this grant required us to report four separate clinical quality measures from our COPD population.
Fortunately, we already had 95% of primary care providers in our region ready to participate in the project. Yet, even with this overwhelming buy-in, we still faced an uphill battle. Why? For starters, most of our practices had only recently implemented their first electronic health record, and thus had little experience capturing clinical quality measures in this way. To add to that, these 36 primary care practices were using a total of 11 different electronic health record systems.
In short – they were just starting to learn the language, and not all of them were learning the same one.
In the end, we performed manual audits at each practice. The various electronic health records did not have the capability to report the necessary quality measures. These audits were very labor intensive, and we realized the process needed to be automated, but how?
Ultimately, my organization decided to invest in a population health management tool, which accepts data from multiple electronic sources and aggregates it. Clinical quality measures can be obtained from a such a tool, as well as gaps in care for any individual patient.
Optimistic that this tool would be our solution, I was shocked to discover how difficult it was to get the right data from our electronic health records into the right place in the population health management tool. It was a slow, painstaking process.
We initially thought we could obtain all the data we needed from the individual health records. We had hoped to build the capacity to share clinical quality performance with our providers and appeal to their competitive nature. Come to find out, we had lofty goals – in two years we were unable to generate performance reports.
Although this failure was painful, we learned some valuable lessons:
- Don’t count on an electronic health record to yield all the data you need; you’ll likely need other sources of data to report certain clinical quality measures.
- Implementing a population health management tool requires an adequate number of staff, and you must have a clearly-defined vendor and internal deliverables and roles
The next leg of my clinical quality measure “journey” came with two new projects to support healthcare transformation. These projects had a total of nearly 70 clinical quality measure performance requirements, and our providers would now be paid based on their performance.
Most of the measures were claims-based, meaning performance measurement would be delayed. This delay makes it difficult to identify care gaps in a timely manner, and to truly improve their performance, our providers needed more real-time, actionable information. Clinical data was required to effectively close care gaps.
By this time, electronic health records had improved their capability to report measure performance and patient care gaps. Our team worked with practices to build reports and develop processes to more effectively manage their patients care gaps. We still had the goal of producing aggregate reports for our providers, so that they could see how they measure up to their peers.
Here we go again! We purchased our second population health management tool in 2016, and this time we were more prepared to acquire measurement data from the electronic health records. Our first experience had taught us the key questions to ask vendors and to assure roles and responsibilities were clearly defined.
We began to receive data at a much faster pace than our previous attempt. We are now able to report some measures for some providers and we continue to improve the quantity and quality of the data. We’re also able to analyze claims to obtain cost data in our population health management tool.
The lesson of my seven-year journey with clinical quality measures is this – don’t try to boil the ocean!
Set reasonable goals and time frames for your project. Accept that you may not get each data element you need from one source and recognize that the functionality you plan for may not be the functionality you end up with. Just like with all journeys, be persistent and take lessons from your setbacks.
By Charles McArthur
Charles McArthur is the Quality Analyst at the Fort Drum Regional Health Planning Organization in Watertown, New York. He has a Bachelor of Arts in Biomedical Science from St. Cloud State University and an associate degree in Respiratory Therapy from the University of Chicago. He previously worked in healthcare quality improvement as a Six Sigma Green Belt in the Mayo Health System. Charlie is also the author of many scientific articles, reviews, and practice guidelines, and is the former chair of the diagnostic section of the American Association for Respiratory Care.