By Gienna Shaw
Organizations are spending an "awful lot of money" on imperfect patient matching efforts, Bill Spooner (right), CIO of San Diego's Sharp HealthCare, said in a panel discussion at the CHIME13 conference in Arizona last week.
"I've been one of the minority skeptics who thinks that we need something better than what we have today," Spooner said.
Sharp HealthCare implemented a master patient index in 1991. "In early 1992 I had the experience of sending 15,000 incorrect patient bills because of patient matching problems," Spooner said. "So we learned very quickly that it isn't easy."
Although the phrase "national patient identifier" has become unpopular, the Office of the National Coordinator for Health IT recently announced a collaborative patient matching initiative to "identify common denominators and best practices" used by both private health systems and federal agencies.
In a recent interview with FierceHealthIT, former National Coordinator for Health IT Farzad Mostashari said that patient matching is a high priority at the federal level, particularly as information exchange efforts grow.
But is the cost of patient matching and identification efforts sustainable?
Sharp, which has four acute-care hospitals, three specialty hospitals, two affiliated medical groups and a health plan, has the equivalent of 10 full time employees dedicated to investigating, evaluating and cleaning up duplicates and overlays. That's roughly $1 million in salaries and benefits alone, Spooner said.
"Should every organization in the country have to continue to spend a million bucks a year? I'd really like to see a study on what it would cost to create an effective [national] algorithm."
He's also unsure, he said, of what would happen if organizations shared patient matching data with each other, especially if they are taking different approaches based on the characteristics of their own patient populations.
"The argument that always comes back is 'well, that wouldn't be perfect," he said. But imperfect is better than nothing, he added--the industry needs to find an algorithm that works for everyone.