A Different Way of Ranking Hospitals

Not by how well they replumb hearts or replace joints, but by how quickly patients go home.


Consumer Reports, the venerable publication from Consumers Union that rates everyday products, from cars and computers to TVs and laundry detergents, has now turned its analytical eye on hospitals, yesterday releasing first-time rankings of 2,857 centers. Access is free. An article will appear in the July issue. (Disclosure: I was the magazine's Washington editor in the early '80s.)

Unlike the magazine's trademark tables with little filled-in colored circles showing how well hospitals perform various procedures or deliver types of care, these rankings show how aggressively or conservatively—longer or shorter stays, more or fewer tests and specialist visits—patients are treated at a hospital relative to all other hospitals. The rankings are based on Medicare patients in the last two years of their lives who had been hospitalized any number of times during that period for any of nine chronic conditions—heart failure, dementia, and coronary artery disease are three. Adjustments were made to compensate for some patients being sicker than others.

It's an interesting angle. High-intensity care is crucial for some patients, of course, but not across the board. Aggressive treatment can make patients sicker. A study by the Dartmouth Institute for Health Policy and Clinical Practice, Consumer Reports' partner in the rankings venture, revealed that Medicare patients aggressively treated for colon cancer, heart attack, and hip fractures were less satisfied with their care, functioned no better, and died more often than patients who got conservative care. So it is in your interest, if you or a family member or friend must enter a hospital, not to blindly accept every test or doctor visit.

Working with the Dartmouth team, the just-created Consumer Reports Health Ratings Center created a scale for each hospital from zero (most conservative) to 100 (most aggressive), factoring in the total number of days in the hospital over the two years before the subjects died and, of that, the number of days in intensive care and the number of physician visits, with those from specialists judged as more aggressive than those from primary-care doctors.

From one end of the scale to the other, the differences staggered me. We're not talking about a couple of days and a few visits. The most-aggressive hospital anywhere, NYU Medical Center in New York, averaged 54 days of care for these chronically ill patients, 12 of them in the ICU, and 131 doctor visits, the vast majority from specialists. At the conservative end of the scale, patients at Scott & White Memorial Hospital in Temple, Texas, stayed an average of 16 days, 4 in the ICU, and had 45 visits from doctors, more than half from primary-care physicians. Data on Medicare costs also was collected, although it was not factored into the ratings. The average at NYU Medical Center for two years of care was more than $105,000. At Scott & White, it was $44,000-plus.

The extremes between hospitals across the country surprised John Santa, director of the ratings center. But he was "stunned," he told me, by the large differences between hospitals in each other's backyard. Me, too. I can accept that in an expensive place like New York, costs are far higher than in Temple, Texas. But a difference of more than 2 to 1? And are chronically ill Medicare patients so much sicker in New York that they need more than three times as many hospital days? Doubtful. The average stay at New York Presbyterian Medical Center, another Manhattan behemoth, is more than two weeks shorter than at NYU. And it is even more doubtful that NYU patients are dramatically worse off than patients at New York Presbyterian.

As fascinating as I find these rankings, they seem to me less a consumer tool than a prod to the public consciousness. They send a message of great and expensive imbalance, that part of our national healthcare problem, as Shannon Brownlee writes in her recent book, Overtreated, is the enormous weight of care that is simply unnecessary.

I wish users could compare hospitals from around the United States, but rankings are state by state. This also means that anyone living in a metropolitan area that crosses state lines—such as New York City, Philadelphia, or Washington—must call up rankings for each jurisdiction. But the power of the message more than makes up for this. In its infancy during the 1930s, Consumers Union successfully inveighed against radioactive strontium-90 in the nation's milk supply. I wonder whether the new rankings might at least give the floundering healthcare system a nudge in a new direction.