There was a story in the Boston Globe the other day about the Massachusetts Department of Public Health’s decision to order hospitals to stop temporarily closing their Emergency Rooms, effective January 1, 2009. This is good news. I first heard about the practice of temporarily turning away ambulances from the ER — commonly called “diversion” — about ten years ago. The Globe did a series of articles on temporary ER closures, and while I can’t recall for sure, I think the culprit was “managed care.” Those stories then, like this story now, focused almost exclusively on ERs, ER personnel, ER administrators and ER clinicians.

I’m not sure that’s the best way to frame this issue. In fact, many people think ER overcrowding usually has more to do with what’s going on inside the rest of the hospital than it does with simply what’s going on in the ER. For example, I went to a meeting about eight years ago that was sponsored by Don Berwick’s Healthcare Improvement Institute — one of the most influential voices on health care quality and safety in the United States — to hear a presentation on ER diversions and ER overcrowding. Berwick was featuring a presentation on the subject by Eugene Litvak, a Professor of Operations at the Boston University School of Management.

Litvak’s contention back then was that the schedule for elective surgery had more to do with ER overcrowding than anything that went on in the ER, and his case was persuasive. His argument had three parts. First of all, he said ERs have a predictable, measurable rhythm of their own. Their volume fluctuates in a predictable way from day to day and week to week. Peak use periods and low use periods can be predicted going forward by tracking when they occurred in the past. Saturday nights after midnight, for example, are usually peak periods. Thursday afternoons are not.

Second, he said the primary reason ERs back up has to do with throughput into the hospital — operating rooms, ICUs, inpatient beds, and the like — not the activity that’s taking place directly in the ER itself. If the ORs and ICUs and recovery beds are full, there’s no place for someone who’s come in through the ER to go if they need to check into the hospital.

Third, a big piece of the OR, ICU and inpatient bed utilization issue can be managed — because it involves elective surgeries — which can be scheduled. His examination of scheduled surgical times indicated that, in many cases, peak time for them collided directly with peak time for cases coming out of the ER and into the hospital. If a hospital re-worked its elective surgery schedule — so that it maxed out when the ER demand on the rest of the hospital was low, there would be no collision, throughput would move through the ER and into the hospital, and overcrowding would go down.

Sometime soon after Litvak put that study out, I asked a friend of mine who ran a big, well-known Boston area hospital if this theory held up. He told me that he’d read the study, worked with his surgical teams to re-structure their elective surgery schedule, and then watched as their ER overcrowding problem “melted away.” He said he was amazed.

He’s not alone. Boston Medical Center, the city’s busiest ER by a wide margin, put Litvak’s theories through the reality test a few years ago, with similar results. They have the most visits, one of the shortest ER waiting times in the city, and spend far less time in “diversion” than almost all of their peers in Boston. Their re-structuring was paid for in part through a grant from the Robert Wood Johnson Foundation, which also studied and published the results of their work. This was picked up by many news outlets, including the Wall Street Journal.

BMC did some other things as well to enhance their best-in-class performance, but the reworking of their elective surgery schedule played a major role in their success.

Paul Dreyer from the MA Department of Public Health offered an interesting observation in the Globe article. He said allowing hospitals to use ambulance diversion as a release valve may have taken away their incentive to focus on the internal backlogs in their institutions. I guess this will offer us all a chance to find out if he’s right.