Why cardiology leaders should be careful about the quality metrics they focus on

"You might hold them in the emergency department or put them in observation, anything to avoid admitting them back into the hospital," Sauer said. "When, in reality, the best place for that patient may be back in the hospital."

Sauer said hospitals and cardiology programs should avoid allowing performance measures to become objectives in themselves.

"Data should simply be a measure of a broader organizational objective," he said. "When the data becomes the goal, that's where organizations can get into trouble."

Too many metrics can be a problem

Sauer urged healthcare executives to streamline the number of metrics they routinely review, concentrating only on information that is meaningful to clinicians and managers and that leads to measurable improvements. He described sitting through staff meetings where participants review extensive dashboards containing dozens of statistics that generate little discussion because they lack relevance to daily clinical operations. Rather than collecting every available data point, organizations should identify measures that support decisions and drive improvements in patient care, he said.

"The things we're measuring should be meaningful to the leadership of a department, an organization or a service line," Sauer explained.

Physician compensation also influences behavior

Sauer also cited physician productivity metrics as another example of Goodhart's Law in practice. Many physician compensation plans rely heavily on work relative value units (wRVUs), effectively rewarding clinical volume above all else. While productivity is important, Sauer said an exclusive emphasis on wRVUs may unintentionally discourage physicians from participating in activities that improve quality but do not generate direct reimbursement.

"If we've basically told physicians that the only thing we value is clinical activity, then that's naturally what they're will focus on," he said.

That can reduce participation in quality committees, administrative meetings and other initiatives that contribute to long-term organizational performance but are not reflected in productivity metrics.

AI may help turn data into useful information

The rapid growth of cardiovascular information systems (CVIS), enterprise analytics platforms and artificial intelligence (AI) is making the discussion about patient data and quality metrics even more relevant than ever, Sauer said. Healthcare organizations now have access to enormous volumes of operational and clinical data, but converting those numbers into meaningful information remains a challenge.

"We have no shortage of data points. What we're short on is information," Sauer said.

He believes AI holds promise for helping healthcare leaders filter through what he described as a "tsunami of data" to identify insights that are truly actionable for managers, clinicians and frontline staff.

Meaningful clinical context remains essential when looking at data

Even as analytics become increasingly sophisticated, Sauer cautioned against interpreting performance metrics without understanding the clinical context behind them.

For example, dashboards may show that one echocardiographer or cardiac CT technologist completes fewer studies per day than peers. Without additional context, that individual could appear to be underperforming. In reality, the slower workflow may reflect that the technologist consistently receiving the most complex patients who require additional imaging time.

Ultimately, Sauer said healthcare organizations should pursue a balanced approach to performance measurement—using analytics to guide better decisions while ensuring that improving patient care remains the primary objective.