Customer satisfaction is quite the rage these days. Many stores and restaurants, many professional offices, hand out surveys, or ask customers to log onto their survey site on the Internet. As a reward, one may win everything from a free sandwich to an iPad.

I’d be interested to know how customer satisfaction played out in my home. “Kids, now you’ve had a week of school-work. Please fill out the attached survey and give your mom and me an idea how we’re doing. Frankly, if you’re not satisfied, your mom is out as teacher!”

How about politics? That would be interesting, wouldn’t it? I know, we vote. But our current system makes it difficult to remove politicians by vote, and even if it’s possible (and between money and media, it can be a stretch), we’re still stuck with them for two to four years. A “citizen satisfaction survey” would be a lot of fun. Unlike a poll, we could give it some real teeth. ‘Senator, it turns out your constituent surveys are really low. So, we’re cutting your salary by half until you bring it up by making people happier!’

Obviously, that technique has problems too, doesn’t it? Politicians can’t always please, or satisfy, everyone. And to do so would be perilous indeed. In the same way that we don’t drop “enemy satisfaction” surveys along with smart bombs. “Did you feel that the destruction of your compound was done in a professional and timely manner? Would you be willing to be bombed by the same pilot again in the future?”

Customer satisfaction has also been around in medicine for a long while, and seems to gain traction every year. Hospitals, struggling for market share, love patient satisfaction surveys and scores. For better or worse, these things often determine funding, raises, even continuation of contracts in the healthcare world. And negative scores can cause significant reprimands. It has some merit here as well. Physicians, and nurses, can’t go around being incompetent, or unprofessional. It makes for a very bad experience. And when we’re sick, or our family members are suffering, that’s the last thing we want.

However, the science isn’t always so good. For one thing, the sample sizes from which the surveys are drawn are often very small. A friend of mine worked in a hospital where the data might be drawn from one survey for a given month. If you made that one patient mad, then it was going to be a tough time!

Next, as in so many settings, happy people tend not to fill out surveys. If you like a product, you tell your friends. You go back and get another. But if you’re upset? Out comes the pen, and the boxes are checked in frustration or anger.

But lately, some physicians have been asking ‘is this really a good idea?’ A study from the Archives of Internal Medicine, published in February, suggested that physicians with very good scores may have patients who do poorly. It’s only one study, but more will likely follow. And it makes sense.

Patient has illness or injury, and desires specific test or drug. Physician feels drug or test aren’t indicated and does not provide them. Patient complains to administrator who pressures physician. Physician begins to do tests and give drugs (especially narcotics) that aren’t appropriate in order to comply with employer. I suspect this is one reason, though certainly not the only one, for the epidemic of narcotic abuse, addiction and narcotic-related deaths in America today.

I believe we should be attentive to customer service. But we have to be careful. In settings in which a high degree of expertise is necessary to make decisions, or in which grave dangers underlie poor decisions, customer service has to be balanced against knowledge and experience.

And more relevant, we often hear citizens and watch-dog groups rail against corporate interests and inappropriate influences. If a business, particularly a hospital, asks professionals to do the wrong thing in order to secure payment from the “customer,” it sure sounds like inappropriate influence to me. Likewise, the companies that push the surveys are, themselves, businesses with financial interests.

We all want to be satisfied customers. But when satisfaction has the appearance of corruption, or results in danger, maybe a little dissatisfaction would be better, and safer, in the end.

Edwin Leap is an emergency physician who blogs at edwinleap.com and is the author of The Practice Test.