Health systems invest significant time and resources in improving HEDIS (Healthcare Effectiveness Data and Information Set) performance. Quality teams analyze care gaps, conduct patient outreach, review measure specifications, and implement targeted interventions designed to improve outcomes.
Yet many organizations continue to overlook one of the most consequential drivers of HEDIS performance: healthcare data quality, particularly the consistency and completeness of the clinical data used to identify eligible patients.
The result is a challenge that often goes unnoticed: health systems may be working diligently to close care gaps for the patients they can identify while simultaneously missing opportunities among patients who never entered the quality workflow at all.
How clinical data quality affects HEDIS performance
When HEDIS results fall short of expectations, quality improvement efforts naturally focus on improving clinical performance – or HEDIS metric numerator improvement. If mammography rates are low, increase patient outreach. If diabetes measures are underperforming, improve care management. These are all reasonable, clinically valuable responses.
But often little is done to address the HEDIS denominator gap. Patients are frequently absent from measure populations because diagnosis, laboratory, medication, procedure, or problem list data is fragmented across systems or represented inconsistently.
Consider a health system attempting to identify all patients eligible for diabetes-related HEDIS measures. Within the enterprise electronic health record (EHR), many patients are documented using specific ICD-10-CM codes such as E11.9 (Type 2 diabetes mellitus without complications), E11.22 (Type 2 diabetes mellitus with diabetic chronic kidney disease), or E11.65 (Type 2 diabetes mellitus with hyperglycemia).
At the same time, newly acquired practices may use local diagnosis descriptions such as “DM2” or “adult onset diabetes,” while laboratory feeds contain HbA1c results stored under non-standard laboratory codes.
Although all of these patients belong in the same clinical cohort, the data lake may not consistently recognize them as such if diagnoses, laboratory results, medications, and problem list entries have not been standardized and linked appropriately. Hundreds or even thousands of patients may be excluded from diabetes measure populations, care gap outreach programs, and quality reporting workflows – not because the care was absent, but because the underlying clinical data was fragmented across source systems before entering the data lake.
The challenge grows as health systems grow
This challenge is becoming increasingly common for a simple reason: health systems are becoming more complex. Over the past decade, most organizations have expanded through acquisitions, physician alignment initiatives, ambulatory growth, and partnerships with external providers. In parallel, they have invested heavily in enterprise analytics platforms, population health solutions, data lakes, and interoperability initiatives.
Each of these investments creates value. Each also introduces additional data sources. Variation is inevitable. And while individual discrepancies may appear insignificant, thousands of them materially affect quality reporting.
HEDIS performance is often determined long before HEDIS season
One common misconception is that quality reporting begins when teams prepare annual submissions.
In reality, many of the opportunities to improve performance occur months earlier.
The first quarter is typically focused on finalizing and reviewing results from the previous measurement year. Valuable lessons emerge from that exercise, but opportunities to influence prior performance have largely passed. The more important question becomes: what can be done differently this year?
The most effective organizations use the spring and early summer to evaluate where patients were missed, where attribution logic broke down, and where data standardization issues affected reporting.
Health systems should routinely ask:
- Were eligible patients excluded from measure populations?
- Are laboratory results being standardized consistently?
- Have acquired practices been fully integrated into quality workflows?
- Are diagnosis and procedure mappings current?
- Do active problem lists accurately reflect chronic conditions?
Actions taken during April, May, and June provide the greatest opportunity to influence current-year performance because there is ample time to identify patients, close care gaps, and correct data quality issues.
That said, the window does not close at midyear. Early Fall remains a critical period for many organizations. Patients who were previously missed due to inconsistent diagnosis coding, fragmented laboratory data, incomplete problem lists, or data lake attribution issues can still be identified and incorporated into quality workflows. For measures involving annual screenings, laboratory testing, medication adherence, or chronic disease management, September and October often provide enough runway to complete interventions before year end.
The challenge is that available options begin to narrow as the calendar progresses. A patient identified in May may have several opportunities to complete recommended care. A patient identified in October may have only one.
This is why leading health systems treat data standardization as a continuous process throughout the measurement year, with the greatest emphasis placed on the first half of the year and early Fall. The earlier eligible patients are identified, the greater the likelihood that care gaps can be addressed before reporting deadlines arrive.
Book a demo to see how standardized, complete data could help your organization identify eligible patients and improve HEDIS performance.
