A Canadian analysis of advice from two popular health shows has stirred controversy with its conclusion that roughly half of the celebrity doctors’ recommendations are not based on solid evidence.

The researchers warn that recommendations on such programs may not be supported by high-quality medical evidence and often lack “enough balanced information to adequately inform decision making.”

"A lot of the responses (to the study) seemed to feel that we were saying that the doctors on these shows are charlatans and that is not true," said Dr. G. Michael Allan, the study’s senior author. "All we are saying is that people need to be skeptical of what they hear."

Allan, director of the Evidence-Based Medicine Program at the University of Alberta in Edmonton, told Reuters Health he is disappointed at how critics have “jumped all over” the study.

His team did the analysis because their patients often would ask if they should follow the advice from “The Dr. Oz Show.”

"So that got us thinking about it, about what kinds of things are being recommended," Allan said.

No researchers have really looked into medical information on popular TV talk shows featuring doctors, he added.

"So again, our study was not meant to generate bad publicity for the shows but simply to question what was being said and if there was evidence to support it," Allan said.

His team reviewed information and recommendations on two of the most popular medical talk shows, “The Dr. Oz Show” and “The Doctors.” Both shows are internationally syndicated and are watched daily by millions of viewers, Allan and his colleagues write in The BMJ.

For each show, they analyzed 40 episodes from early 2013, focusing on the nature and content of recommendations by the shows’ physicians and guests. They also noted whether the TV doctors provided additional information, such as the cost of a recommended product, or if there was a conflict of interest, such as a financial benefit to be gained by the person giving the advice.

The researchers then focused on 80 of the strongest recommendations to see if the advice was supported by medical evidence.

They searched several medical databases, and Google, for research to support the recommendations. A major analysis of all existing research on a given topic, called a meta-analysis, was considered to be the strongest type of evidence.

The team also took into account smaller studies, case reports and reviews of past research.

On average, there were about 12 recommendations per episode on “The Dr. Oz Show” and 11 on “The Doctors.” Overall, roughly half (54 percent) the recommendations from both shows were supported by at least one piece of evidence.

For “The Dr. Oz Show,” 46 percent of the recommendations were supported by evidence. But 15 percent of the time, the medical evidence contradicted the show's advice. And for 39 percent of the recommendations, there was no evidence.

Results for “The Doctors” were similar. Evidence supported 63 percent of recommendations, contradicted the advice 14 percent of the time, and was missing for 24 percent of recommendations.

The most common advice on "The Dr. Oz Show" was dietary, while on “The Doctors” the most common recommendation was to consult a health-care professional.

About 40 percent of the time on both programs, specific benefits were mentioned along with the recommendations.

Less than 10 percent of the time, the shows mentioned possible “harms” associated with the recommendations.

"’The Dr. Oz Show’ has always endeavored to challenge the so-called conventional wisdom, reveal multiple points of view and question the status quo," Tim Sullivan, director of publicity for “The Dr. Oz Show,” told Reuters Health.

"The observation that some of the topics discussed on the show may differ from popular opinion or various academic analyses affirms that we are furthering a constructive dialogue about health and wellness," Sullivan said.

In response to a request for comment, “The Doctors” provided a statement that takes issue with the study itself, and says the show was never contacted by The BMJ about the study.

"We feel strongly this study was designed to provide a provocative headline versus inform the public in an unbiased way," the statement reads. "Our production is extremely proud of the credible information we deliver to our viewers on a daily basis and take very seriously our role in how we deliver it."

There are often multiple options for treating most medical conditions, the statement notes. "Our goal is to educate and inform by consistently offering a variety of medical points of view from multiple experts across different specialties; we will always advise that people discuss treatment options with their own physician."

In an online response to critics, Allan's team acknowledges that standard medical practice and doctors' advice to patients aren't always backed by evidence.

“As we note in the paper, in reviews of practicing doctors, only about three quarters of what is recommended is supported by evidence,” they write on The BMJ’s website. “In fact some research suggests that even for guidelines, moderate or good evidence is present for only about 50% of their recommendations. The highest level of evidence, randomised controlled trials, is present for only 10-15% of the recommendations.”

Nonetheless, Allan’s team says, their findings should be considered a call for more research into how health information is communicated.

“Going forward,” they write, “we would encourage healthcare talk shows providing recommendations to be (more) specific regarding anticipated benefit, the magnitude of the benefit, the potential harms (including costs), the presence/absence of supporting evidence and conflict of interest statements for their recommendations.”