Health care studies can be dangerous to your health. Sometimes, researchers draw conclusions that are very misleading. Then people working to make health care better may go down the wrong path.

I saw evidence of this mistake nearly 20 years ago - and as recently as this year.

For example, a number of years ago a textbook about how to get managed care to work started out with the correct facts but drew some misguided conclusions. It said, "Most consumers purchase health care services on the assumption that the quality of care is essentially the same."

That is, patients tend to assume that all doctors and hospitals provide good care.

That's what the research shows. So far, so good.

The textbook then goes on to describe the best way to get patients to choose a given health insurance plan offered through their employer. It says to focus on issues other than how good the care is. It gives examples, such as whether the insurer's employees are courteous.

It doesn't suggest that it would be helpful to keep track of what happens to people as a result of the care that they get through that plan. Do they get better quickly? Do they get treatment after treatment and stay the same or get worse?

No, the textbook glosses over quality of care and focuses on secondary issues. Its view seems to be that the goal of insurance companies should be to sell insurance - not to improve people's health or quality of life. And that the best way to sell insurance is to focus on side issues.

Suppose that you had a choice of two insurance plans, Plan A and Plan B.

In Plan A, doctors and hospitals are located near where you live and work. When you call the customer service number, they answer on the first ring and are polite. But if you have to go into the hospital, you have one chance out of 10 that you'll pick up an infection or get some other complication that seriously hurts you.

In Plan B, you have to drive farther to get to doctors and hospitals. It takes longer to get through to a customer service agent on the phone. But if you have to go into the hospital, you have just one chance out of 100 that you'll be seriously harmed by the care you get there.

Which plan would you choose?

Given this information, most people would choose the plan that is less likely to hurt them or their family members.

But the managed care textbook suggests that quality of care isn't an important selling point. Yes, people tend to believe that all doctors and hospitals provide the same high quality of care. But that assumption is not accurate, any more than it's true that all schoolteachers do a great job teaching.

So what conclusion do I draw? Instead of using people's lack of knowledge as an excuse to focus on less important issues, it would make sense to provide people with the information they need to choose the care that will get them better results.

That's an approach that the textbook did not suggest.

More recently, the Archives of Internal Medicine published the results of a major study. It was set up to see if patients who were satisfied with their care spent less time in the hospital and lived longer than patients who weren't.

The answer? A resounding "No!" The patients who were most satisfied with the care they were given got the worst results. They had to take more prescription drugs. They ended up in the hospital more. And they died sooner.

The results of this study could lead people working in health care to say, "See? Patient satisfaction has nothing to do with good results. There's no point in checking to see if patients are satisfied with their care. Doctors do a good job deciding what care people should get. They are forced off the right path when they pay too much attention to whether patients are satisfied."

But that's not the conclusion I draw.

Many studies have shown that high patient satisfaction is related to how many services and prescriptions a patient gets. One doctor may say, "Do the exercises we talked about, ice your back, and take over-the-counter painkillers. Check back with me in two weeks."

Another doctor may say, "I'll take some X-rays today, and then here's an order for an MRI. I'm giving you two prescriptions - a muscle relaxer and a narcotic pain reliever. Come back in three days to review the results of your X-rays and MRI."

Overwhelmingly, research shows that patients of the first doctor will be dissatisfied -- they didn't walk out with even a single prescription! Patients of the second doctor will be very satisfied - look at how many tests and treatments they got! But research shows that the patients of the first doctor are likely to get better results.

Given this picture, it's no surprise at all to find that satisfied patients use more health care services and die sooner.

Does that mean that your doctor, your hospital, your insurer, and the government should stop all efforts to improve patient satisfaction, or to pay more to doctors and hospitals whose patients are less satisfied?

No, it doesn't. It means that patients need facts that they can easily use to understand the pros and cons of using different treatments, doctors, and hospitals. With this information, they are likely to be more satisfied with care that does a better job of getting them back to their normal lives.

For example, suppose that 67 people out of 100 who get exercise, ice their backs and take over-the-counter painkillers are better in three months. Suppose that 38 people out of 100 who get X-rays, MRIs, and prescription drugs are better. Which treatment would you choose?

People assume that "more is better" only because no one typically gives them the information that reveals that "more" is often "more likely to hurt you."

This fact doesn't mean that patients should be kept in the dark and their priorities ignored. It means that they should be given facts so that they can make choices that will get them the results that they care about.