Patient matching: The status quo is unacceptable
On Tuesday, the College of Healthcare Information Management Executives officially launched its national patient identifier challenge. The desired result, organization executives and board members shared, is to create a safe, accurate and private solution for providers to identify patients.
The word many have chosen to focus on--and with good reason--is private. The Department of Health and Human Services, of course, is barred from funding any efforts to develop a national patient ID. However, many in the healthcare industry, including government officials, believe that such a tool could be instrumental in helping providers to achieve interoperability of electronic health record systems; to that end, National Coordinator for Health IT Karen DeSalvo spoke encouragingly about the project. This is not surprising, especially considering that the agency's Interoperability Roadmap and Health IT Strategic plan both call for public and private stakeholders to work together to improve health data exchange efforts.
The word that needs more attention, however, is accurate.
At the kickoff event, CHIME Board Chair Marc Probst, CIO at Salt Lake City-based Intermountain Healthcare, commented on a question about how current accuracy levels for patient matching stack up to the 100-percent threshold that has been set.
"The numbers that we've been floating around are probably at Intermountain Healthcare we're in the high 90s--maybe 95, 96 percent of the time. But in general in the industry, within a system, it's probably about 90 percent of the time," Probst said. "When you start getting outside of your systems, so all the interoperability we're talking about, the numbers go down dramatically--maybe 50-60 percent of the time, maybe even lower."
CHIME President and CEO Russell Branzell added that "mid-80s is the norm" for federally supported health information exchanges, saying that, according to federal statistics, there are organizations reporting "far, far worse" patient matching rates.
The latter statement alone illustrates just how important such efforts are. Let's just say that "mid-80s" equates to roughly 85 percent; that means that 15 percent of patients, on average, are not matched to their correct records.
That's unacceptable, especially considering that just one mismatched record could mean the difference between life and death.
To that end, Branzell said, the healthcare industry has "an ethical responsibility" to fix the issue, with or without the help of government.
Whether the effort succeeds remains the $1 million question. Even if a tool or process is developed that, at first blush, achieves 100 percent accuracy, scaling such a system nationally will be incredibly difficult.