Despite physicians' dislike of online rating sites, a survey finds patients are increasingly relying on them when shopping for a new doctor.

A survey conducted by researchers at the University of Michigan Medical School found that when choosing a doctor, physician rating sites weren't ranked as high as other factors such as word of mouth from family and friends or whether a physician accepts the patient's insurance.

But there is evidence the rating sites have become an important tool. And use of them is likely to continue growing.

The survey of more than 2,100 U.S. adults finds 59 percent think physician rating sites are somewhat important (40 percent) or very important (19 percent) when choosing a physician. A third of respondents reported selecting a physician based on good ratings, and 37 percent had avoided a physician with bad ratings. The findings were published online Feb. 19 in the Journal of the American Medical Association.

"The amount of awareness, to me, was somewhat surprising," said Dr. David A. Hanauer, associate professor in the Department of Pediatrics at the University of Michigan Medical School, and lead author of the study.

When compared with review sites in other industries, such as cars, movies or electronics, awareness of physician review sites is lower. But usage of the physician rating sites was similar to the other industries, according to Hanauer.

Online physician ratings have been the focus of much scrutiny and criticism from the medical community since they started gaining popularity several years ago. The American Medical Association has long been critical of physician rating sites saying they should not be a patient's sole source of information about a physician.

AMA passed policy in 2010 calling for legislation that would require review site operators to prominently state whether they have been approved or sanctioned by an organized medical association or regulatory agency.

Because more patients use the ratings than contribute to them, the concern many physicians have is that the ratings might not be an accurate representation of the overall patient experience. A PwC report published in April 2013 found that 48 percent of patients have consulted online reviews, but only about 25 percent have written one. Only 5 percent of those surveyed by Hanauer and his colleagues had written a review.

A 2011 study in the Journal of Medical Internet Research found that the majority of reviews left for physicians are positive. But because so few people actually write a review, one or two bad reviews could have a significant impact to a physician's overall rating.

"If you only have a few ratings and one or two of them are negative, then I think that really is potentially harmful to that clinician if other people are actually using it to make decisions," Hanauer said. And because of confidentiality issues, physicians have little recourse when it comes to responding to negative reviews, he said.

But, a 2010 study in the Journal of General Internal Medicine found the majority of negative reviews were based on nonclinical complaints that physicians could respond to without breaking any confidentiality agreements. It found most of the complaints were based on things such as parking or long wait times.

Hanauer acknowledges physicians have an extreme dislike of the review sites, but use of them is likely to continue to increase.

"At a certain point, while people may not want to embrace it," he said, "physicians have to accept it because it is a new reality."