Joel Berman is in an enviable position for a chief medical information officer.
Six months into a slow-roll implementation of a computerized physician order entry system at 220-bed Concord (N.H.) Hospital, Berman has had not one medical staff delegation show up at his door with flaming torches. In the CPOE business, that is not damnation with faint praise.
On the contrary, a recently completed survey of medical staff physicians gives Berman cause for optimism, and in one case, surprise.
We had implemented CPOE in September 2007, and six months later we wanted to find out our providers point of view, Berman said. What was working well? What did they like? What did they want different?
So far, about 80 of the 300 or so physicians with privileges at Concord are using McKesson Corp.s Horizon Expert Orders CPOE system, Berman said. Thirty-three of them answered questions in the online survey, with solid majorities (between 72% and 94%) rating the system very easy or easy to use for finding and entering orders for medication, laboratory results, diagnostic tests, and support orders for dietary, physical and occupational therapy.
Not surprisingly, given those relatively high marks for functionality and ease of use, 81% of the doctors surveyed rated their CPOE training as excellent or good while 88% gave the same positive ratings for on-going support. Fifty-three percent of physicians concluded that using the system yielded a significant (6%) or slight (47%) improvement in the quality of care.
What was unexpected, Berman said, was the perceived impact on efficiency.
In general, physicians are not the most happy stakeholders in the equation because CPOE requires them to do order-entry at a detail that previously they hadnt, Berman said. Providers are used to (Microsoft) Windows functionality and so they expect to be able to minimize screens and to right click and get definitions and I dont know of any (CPOE) system that has that. CPOE is not time-neutral, especially early on.
Although Berman said Concord has not put a stopwatch to specific order writing, for many physicians plugging away at an unfamiliar system, a common perception is that for certain tasks it takes longer with CPOE than with paper. Even so, Berman said he was heartened by physician responses to questions about the impact of the system on efficiency of care.
While just 6% of physicians found the system significantly increased their efficiency, 34% responded efficiency increased slightly and another 12% reported no change. And while 41% reported it decreased efficiency slightly and another 6% significantly, Berman is more than satisfied with those numbers so far.
Within the first six months, That surprised me, Berman said. We were thinking there would be a larger percentage that would have rated it unfavorably.
Berman said he believes the physicians favorable responses, so far, have been because many of them are taking a broader view of efficiency. If physicians looked only at how long it took them to write an order as opposed to entering it in the system, theyd probably rate the new CPOE system as less efficient, he said. But if physicians use a systemic definition of efficiencyand he believes many at Concord havethat includes a reduced number of callbacks from the pharmacy due to poor handwriting or contraindications, many will conclude they make efficiency gains, Berman said.
We've been gratified by the overall level of acceptance by the medical staff, Berman said. The question of time efficiency is a bell-shaped curve, as you might expect, depending on the user's facility with, and attitude towards electronic applications. Medication reconciliation remains a huge challenge, as I sense it is everywhere else. But in general, a majority of providers are reasonably satisfied with the system, and a little over half have been able to climb the learning curve to time neutrality or better.
Concord had set the IT table for CPOE, having already implemented a suite of McKesson programs including a clinical data repository, a nursing management suite, a picture archiving and communication system/radiology information system and a physician portal. But physicians still used paper for order-entry and notes.
The hospital also prepared the staff by selling the benefits. It created a game show, sort of a stump-the-staff in PowerPoint, projecting on a screen some of the Concord physicians own orders and asking staff members to decipher their own handwriting. While the physician scribe was not identified, people had to vote and we had to show them what the correct result was, Berman said. We started out humorously and got more serious. The last slide, Berman said, was an example of a physician who overdosed a girl with Dilantin by a factor of 10.
Another possible positive influence on physician perception of system efficiency has been Concords benchmarking efforts pre- and post-implementation of CPOE.
We had some experience with our outpatient EMR that taught us to do that, Berman said. The first of what are now nine physician offices owned by the hospital went live with an EMR in 1995. That was our primer on how to effect organizational change.
With the CPOE rollout, the hospital used 40 different metrics to document progress, including the 11 metrics for diabetes care promoted by the Bridges to Excellence program, Berman said.
For example, Documentation of diabetic foot exams went up from 20% to 80% in six months, Berman said. There were prompts at the point of care along with population reports, so everybody could see how each provider was doing.
Concord also documented a one-day observational survey comparing its post-go-live deep-vein thrombosis prophylaxis rate in December 2007 to the pre-CPOE rate in October 2006. DVT prophylaxis rates for eligible surgical patients increased from 92% to 100% while rates for medical patients rose from 77% to 100%, according to Berman.
Berman credits Concord President and Chief Executive Officer Mike Green as the real champion of this. (He) accurately identified CPOE as the single-most challenging thing in his 15-year career. He decided two years ago that we could no longer continue to do paper-based orders.
Workflow is paramount in the hierarchy of needs as much as we want to improve quality. The other corollary is, since this is such a complex change, you cannot predict the downstream impact on what youre going to do when you introduce CPOE.
This is why Berman said he advocates a phased implementation of CPOE, not the big bang approach in which the system is turned on across the enterprise. At Concord, internist and physician champion Paul Clark became the first CPOE user.
He wanted to test it, and on day one we found some things worked very well and some other things had consequences downstream, messages were sent to the wrong place, Berman said. Then his colleagues joined him in his (three-physician) internal medicine group. Then we rolled it out to the hospitalist group in September. By then, it was a pretty smooth running system. But every time we introduced a new medical specialty, we found new orders and new problems. We found it (slow roll) more effective than big bang.
Berman said he doesnt think the hospital will ever be able to push all orders through CPOE. For now, the goal is for 90% of physician orders to come through CPOE within six months.
Were about halfway there now, he said. Overall, they are as a group pretty satisfied with it, which to me is a huge accomplishment.
What do you think? Write us with your comments at [email protected]. Please include your name, title and hometown.