By Gerald T. O'Connor, Ph.D., Sc.D.
On March 15, 1958, the cover illustration on the Saturday Evening Post depicted Dr. Donald Campbell in his Stockbridge, Mass., office. A young boy was standing on a chair reading the doctor's diploma while waiting, with a bared bottom, for an injection. This Norman Rockwell painting has retained its popularity for nearly half a century because it emphasizes the doctorpatient relationship and captures the essence of why people devote themselves to careers in health care—they want to take care of and help individual patients.
The practice ofmedicine has changed in important ways during the 49 years since then. The explosion of knowledge and technology has greatly benefited patients. But the price of this progress has been an enormous increase in the complexity of the systems of care.
In 1987, Dr. Stephen Plume, a Dartmouth cardiothoracic surgeon, introduced me to the complexities of cardiac surgery. TheHealth Care Financing Administration (HCFA) had used insurance claims data to develop hospital mortality rates, and the rates for cardiac surgery had been reported in local and national newspapers. There were few clinical databases at that time, so many heart surgeons first learned their mortality rates in the press.
Trust: Steve Plume and I were concerned that important variables were missing from HCFA's model. So in April 1987, he convened a meeting of the staffs of five New England cardiac surgery programs to propose developing a shared cardiac surgery database. The group, now known as the Northern New England Cardiovascular Disease Study Group, collected data that revealed opportunities for improvement. As trust was built among the group's members, they began to collaborate, visit each other's institutions, observe processes, and use what they learned to make changes. Over the years, these changes have resulted in substantial reductions inmortality rates. The group is still active and recently celebrated its 20th anniversary.
During these 20 years, I have worked on process improvement with dozens of cardiac surgery teams in the U.S. and Europe. Typically, a team includes from 50 to 100 members representing many different specialties. These are highly trained, caring people who want, both professionally and personally, to do what's right for their patients.
Observing hundreds of cardiac surgeries has taught me some lessons. Sometimes cardiac surgery is practiced as though it is a solo performance, and sometimes as if it is an ensemble effort. The latter, team approach produces better results. The only way to improve clinical care is by changing the processes of care. Admonitions to "try harder" never work and are disrespectful of the practitioner. We should assume good intent and fix processes rather than assign blame. These ideas are well known by those involved in clinical quality improvement.
Both the processes and outcomes of cardiac surgery are wildly variable. Sometimes this is because there is no clearly defined approach. Sometimes, however, there are systems failures, such as a lack of explicit responsibility and authority for each team member.
Interactions: Why does this happen? Clues can be found from watching how team members interact. I have attended many cardiac surgery team meetings and have observed that some meetings are very positive and productive. Others, however, are dysfunctional and chaotic. There is no clear agenda. Stream-of-consciousness conversation rules. Stereotyping of clinical specialties is common. Team members are often interrupted or publicly cross-examined. Their body language reveals unspoken embarrassment, hurt feelings, and sometimes anger. Such meetings end with no clear consensus, only a hazy picture of the team's intent. And such behaviors may not be the only problems that lead to poor care.
Sometimes a single, simple interaction is revealing. For instance, at an excellent Midwestern medical center, a prophylactic antibiotic to prevent surgical wound infection was administered in a timelymanner to only a third of cardiac surgery patients. A quality improvement team, created to remedy the problem, held meetings, collected data, and put reminders in place for the anesthesiologist to give the prophylactic antibiotic 30 to 60 minutes before the skin incision was made. But these efforts yielded only small improvements.
Revealing: Then a pharmacist shared a revealing story at a team meeting: "Last week I was in the cafeteria and I saw an anesthesiologist having lunch alone, so I joined him.We discussed the antibiotic issue and he said, 'I really try hard to do a good job as a cardiac anesthesiologist. I care a lot about my patients, but there is no part of anesthesiology practice that requires the use of antibiotics. If the surgeon believes that hemight contaminate the operative field, then he should give the prophylactic antibiotic. If, for some reason, he cannot do so then he should ask me. If he asked me, I would do it.'" The anesthesiologist was not being difficult but was just expressing a very human need—to be treated with respect.
Aretha Franklin, in her hit song "R-E-S-P-E-C-T," got it right: "I got to have (just a little bit). A little respect (just a little bit)." If all caregivers could be more respectful of all their colleagues, it would improve our daily work and benefit our patients.
The Grand Rounds essay covers a topic of interest to the Dartmouth medical faculty. O'Connor, a professor of medicine and of community and family medicine, is DMS's associate dean for the evaluative clinical sciences. He is also the research director for the internationally recognized Northern New England Cardiovascular Disease Study Group, whose work he mentions above. A voluntary collaborative of cardiac surgeons, cardiologists, nurses, administrators, and other medical professionals, the group tracks the management of cardiovascular disease in eight Northern New England medical centers.
If you'd like to offer feedback about this article, we'd welcome getting your comments at DartMed@Dartmouth.edu.
This article may not be reproduced or reposted without permission. To inquire about permission, contact DartMed@Dartmouth.edu.