Invasive Procedures: Less is More ... And Better

Invasive Procedures: Less is More ... And Better

By Donald M. Berwick, M.D.
Newsweek, October 16, 2006

The modern hospital is the cathedral of our time—gleaming, mystical, intimidating, even majestic. It seems to contain miracles and, sometimes, it does; bioscientific breakthroughs have changed the course of illness, curing cases of leukemia, fixing hearts, transplanting organs. It is easy to be in awe. We want whatever the magic of health care can give us—the more, the better.

So it comes as a surprise to most people that the facts often suggest otherwise, as researchers at Dartmouth Medical School—led by Profs. John Wennberg and Elliott Fisher—have been showing for years in the so-called Dartmouth Atlas project. They sort the United States into 307 geographically defined "hospital service areas," and study how often Medicare patients in those areas get specific medical and surgical procedures.

For many procedures, the variation is stunning. Compared with the lowest-use areas, people in the highest-use areas get 10 times as many prostate operations, six times as many back surgeries, seven times as many coronary angioplasties and 10 times as many hospital days if they have heart failure. It all raises an obvious question: if medical technology is being used so differently, who's right?

A lot of the variation depends not on clear-cut scientific evidence that one treatment is better than another, but on the beliefs of specialists in the area and the supply of doctors and hospital beds. The use of specialist visits varies by 660 percent, and what best predicts the rate is the number of specialists per capita. The more doctors, the more doctor visits. The more hospital beds, the more days spent in the hospital.

Are you lucky if you live in a high-use region? Not necessarily. Landmark studies by Fisher showed that high use did not mean better quality of care and outcomes. In fact, for many measures, quality and outcomes were best in the low-use areas and worst in the high-use areas. The less, the better.

Fisher and Wennberg suggest that the rates of use of health-care procedures and treatments should, indeed, vary a lot, but not according to where you live. Rather, variation should reflect the preferences of patients ("Would you rather wait a bit to see if your back pain gets better, Mr. Smith, or try surgery now?") instead of the local habits of doctors or, worse, the local supply of specialists. When patients are actually invited to participate in decisions about their care—"shared decision-making"—both costs and rates of use of expensive, invasive procedures tend to fall, and outcomes and satisfaction improve. The Dartmouth team calls this "preference-sensitive care," and it thinks health care driven by necessity rather than supply could be both more responsive to our needs and, overall, far less costly.


A shortlist of "high variation" surgical procedures in the Dartmouth Atlas includes gallbladder removal; coronary-artery bypass and coronary angioplasty (in people with minimal symptoms); hip replacement; carotid-artery surgery; radical breast surgery (instead of lumpectomy) for localized breast cancer, and prostate removal for benign enlargement of the gland. If you or a loved one is scheduled for one of these, you might pause and ask some questions. Do you understand the alternatives to surgery? Would a period of "watchful waiting" help? Is the decision to operate based on your own assessment of your options? Or does it reflect mainly the local habits of the medical-care system? Surgery may be the right choice, but that ought to depend on the patient's preferences as well as the scientific facts.

Berwick is clinical professor of pediatrics at Children's Hospital and Harvard Medical School, and CEO of the Institute for HealthcareImprovement. For health information from Harvard, go to health.harvard.edu

Read article at orignal source