While physician compensation continues to rise, productivity remained flat in 2014, according to consulting firm SullivanCotter’s Medical Group Compensation and Productivity Survey, released this summer. For hospitals, that’s an unsustainable trend, as their survival depends on getting the most—and best—work possible out of their physicians.
As such, as patient care organizations experience the shift to value-based care, they might begin to see their bottom lines decline at first. At Texas Children’s Hospital in Houston, this trend created a new urgency to fully understand its costs in relation to its revenues. At the core of the hospital’s approach involved using an enterprise data warehouse (EDW) and analytics to take a deep dive into physicians’ actual productivity.
In the past, Texas Children’s process for determining physician productivity was far too labor-intensive to support efficient decision-making, the organization’s officials say. Specifically, practice managers were required to access multiple databases on an ongoing basis to gather the data required to calculate relative value units (wRVUs)—a measure of value used in the Medicare reimbursement formula for physician services.
The effort was so resource-intensive that it took four weeks to obtain the necessary data. To compound matters, the reports lacked a common look or feel; each report was unique to the practice administrator, revenue cycle manager or data analyst that produced it. This complicated, time-consuming and inconsistent process made it very difficult—and sometimes impossible—for leadership to understand where physicians’ productivity and associated compensation stood compared to their peers internally and nationwide, says Mark Mullarkey, senior vice president at Texas Children’s.
While evaluating what was needed to make physician practice management more efficient, Texas Children’s leaders knew that they needed to have complete transparency and readily-available data to be able to put in front of administrative and physician leaders so they can see quickly if operational tweaks would create the outcome and change they were looking for, Mullarkey says. As such, they tapped into many different data sources from financial systems, its electronic medical record (EMR), patient satisfaction surveys, and other interfaces of other systems, bringing that all into its enterprise data warehouse, he says. On top of this, the organization leveraged Salt Lake City, Utah-based Health Catalyst’s Labor Productivity application to run on its existing EDW platform.
“The [EDW] enabled us to have one consistent view, rather than going to 20 or 30 different spreadsheets or systems to look things up. [We] used to spend more time looking for data and massaging it than understanding the data and doing interventions to make the needed changes. The data was at our fingertips, and that made a big difference,” Mullarkey says.
Mullarkey gives an example of bringing in EMR scheduling into the database, which allowed Texas Children’s to show in its reporting what its schedule template utilizations were. “We created such complexities in our EMR, it was not evident to us that we had the various openings in our schedules that we had,” he says. “Doctors would say to us that we want them to be as productive as possible, but they have these opening slots in their day which we need to fill in. This wasn’t about telling physicians to work harder, but about us as a team to make sure every opportunity was being maximized. We brought that right into the EDW, and we could [categorize] by specialty, by group, by physician, or by location to see where the opportunities were,” Mullarkey says. As such, the hospital was able to see in real time, through the data warehouse views that it built, if these operational tweaks were making a difference. “We saw 70 percent utilization rates increase to 90 percent. There was much greater transparency with the ability to slice-and-dice data,” he says.
Another operational improvement was made with no-show rates, Mullarkey notes. “If the physicians are there and the staff has done its job putting patients on schedules, how do you address no-shows? One would typically say to overbook patients by the no-show rate which is a fair strategy,” Mullarkey says. For example, if your no-show rate is 15 percent and you overbook by a 15 percent rate on every single provider but don’t pay attention to when you’re overbooking them, you will create havoc in the practice, he says.
“When we brought data into the EDW and created our views, we saw that one provider has a 20 percent no-show rate, while another has a 7 percent rate. So we were able to differentially overbook. The organization was also able to see that no-show rates were greater at certain times of the day and certain days of the week, allowing the staff to target overbooking so they could predict where those no-show rates could be higher. As a result, there weren’t heavy volumes of patients coming in, and there were no patient experience issues where they ended up waiting for long periods,” Mullarkey says.