A concerted effort to share data and compare routines for coronary bypass graft surgery resulted in a sudden sharp drop in deaths at the five hospitals performing the procedure in three New England states.
The result followed nine months of feedback on surgery outcomes, training in techniques of continuous quality improvement and the hospitals' observation of one another's departments in action.
In the nine months immediately after that "intervention" period, the hospitals collectively posted a drop of more than 20% in mortality, compared with the expected rate of deaths given the condition of patients prior to surgery.
That lower rate was maintained up to the end of the study period in mid-1993, said Gerald O'Connor, lead author of the report and an epidemiologist at Dartmouth-Hitchcock Medical Center in Hanover, N.H.
The analysis was published in the March 20 issue of the Journal of the American Medical Association.
From April 1991 through July 1993, 234 deaths were observed, 74 fewer than expected from a total of 6,488 operations performed at Eastern Maine Medical Center in Bangor, Fletcher Allen Health Care in Burlington, Vt., Catholic Medical Center in Manchester, N.H., Maine Medical Center in Portland and Dartmouth-Hitchcock.
The collaboration among 23 surgeons and their attendant clinical staffs was critical to developing an atmosphere for clinical improvement, O'Connor said.
In isolation, each surgeon and staff may see only a few operations a year that result in the patient dying because of any number of reasons, and they may not experience the same type of complication for years, he said.
In addition, the practices developed during surgery can become so ingrained that they aren't questioned. "We get so used to the system around us that we don't notice it," he said.
By comparing cases and watching each other work, the clinicians were able to revise any number of small processes and anticipate potential hitches, creating a system that made fewer mistakes, O'Connor said.
In particular, the surgical teams improved the points where responsibility for a patient moved from one specialist to another. "Handoffs are an absolutely critical time in healthcare, where errors are likely to occur," O'Connor said.