Why Doctor Ratings Are Misleading

There is danger in making much of physician-rating sites.

Like many of the over 800,000 physicians in the United States, I regularly receive emails concerning my online reputation. In just the past week, I have received three that open with:

“Dear Dr. Gunder. Do you know your doc score?”

“Congratulations, RG. You are eligible for the Hippocrates Award.”

“Hello, Dr. Richard. How many bad reviews do you have?”

The messages come from online vendors whose services include compiling and publishing patient satisfaction scores, providing free medical advice to unseen patients, and helping physicians manage (read: improve) their online reputations.

Never mind the fact that most physicians have never heard of these companies, that most recipients of such messages never requested a reputational biopsy, or that many of these solicitations convey all the sincerity of a bulk mail notice from a bank bearing congratulations on the news that you have been pre-approved for a credit card.

Some naive patients may take such ratings seriously, but very few of my colleagues in medicine do. For one thing, these ratings are often based on a very low number of reviews. A physician that treats hundreds of patients might have less than a handful of ratings. This poor participation rate lowers the probability that the online rating truly reflects the aggregate views of the physician’s patients, and necessarily exaggerates the influence of even a single disgruntled or highly laudatory reviewer.

In most such online ratings, there is no guarantee that the person submitting the review was even cared for by the physician. Like other online ratings for services as diverse as lawn care, college professors, and automobile dealers, there is no practical way to verify that any particular rater knows what he or she is talking about. And the more influential such ratings become, the greater the incentive to manipulate such scores artificially.

Major aspects of a physician’s overall quality tend not to be readily apparent to patients. There is a natural tendency to overrate aspects such as accessibility, affability, and bedside manner. Friendliness of office staff and even ease of parking may figure prominently. Though important, these features do not tell the whole story. Other vital aspects of care, such as the physician’s fund of knowledge, technical skill, and professional judgment may be difficult or impossible for most patients to evaluate thoroughly.

Good outcomes do not necessarily reflect good medical care, and the same can be said conversely for bad outcomes. A patient with a minor and self-limited viral infection might be very satisfied that a physician ordered several diagnostic tests and prescribed antibiotics, despite the fact that such measures did nothing to hasten recovery. Conversely, a patient with an incurable disease might express great dissatisfaction, despite receiving the very best care possible under the circumstances.

Most disturbingly, it is possible to be highly rated by such online services and yet be a remarkably bad physician. In one celebrated case last year, a New Jersey cardiologist who admitted to employing unlicensed personnel, allowing poorly qualified staff to treat patients, and defrauding insurers of nearly $20 million as also found to have online scores of “very good” to “excellent” and had received three quality awards from one of the most prominent online physician rating services.

This is not to say that patient satisfaction and physician reputation do not matter. To the contrary, satisfied patients are more likely to return to their physician, practice, or hospital for care. They are also generally more likely to follow a physician’s recommendations, and may even enjoy better health outcomes. Everyone knows that one of the most important sources of referral for physicians is the personal recommendation of a satisfied patient.

Yet there is real danger in making too much of online doctor ratings. Doing so may lead physicians to do things that conflict with their professional judgment—such as providing unneeded medications and procedures—in the interest of improving scores. It may also lead them to spend too much time worrying about and buffing up their reputations, rather than focusing their attention on taking the best possible care of their patients.

To top it all off, there is evidence that satisfied patients are not the best cared for or healthiest. A March 2012 study in the Archives of Internal Medicine showed that patients with the highest satisfaction scores were more likely to be taking prescription medications, visit doctors’ offices, and enter the hospital. They were also likelier to be in poor health and die in the ensuing years.