TORONTO - Drug advertisements aimed at consumers may not be having the effect on sales that opponents and proponents of the practice assume they do, a new study suggests.

The analysis, by researchers from Harvard Medical School and the University of Alberta, looked at Canadian sales data for three drugs that were heavily advertised in the United States, ads which Canadians watching U.S. television would have seen.

The researchers found no evidence of a spike in sales for two of the drugs after the TV ads started to run. There was a marked increase in sales for a third drug but the effect was short-lived.

"I think that we've shown that the effects are pretty unimpressive for the three drugs we've looked at," said Harvard professor Stephen Soumerai, the senior author.

"Two out of three there isn't an ounce of effect."

The study was published Tuesday in the British Medical Journal's online edition. The authors say it is the first to actually test for evidence of an impact of drug ads by using what's called a control group.

In science, evidence generated by comparing a group that received an intervention - in this case, watched TV drug ads - to a group that did not is considered to be a stronger level of proof than evidence gathered by simply observing what happens after an intervention.

For this study, the researchers took advantage of Canada's linguistic divide, comparing sales for the three drugs in most of anglophone Canada to sales in Quebec. The authors assumed Quebecers watch much less U.S. television than other Canadians and therefore would see fewer drug ads.

Direct-to-consumer drug advertising is illegal in most countries around the world, including Canada. In Canada, media giant CanWest Global Communications is fighting a court challenge of the ban, saying it violates the right to free expression enshrined in the Charter of Rights and Freedoms.

Currently, Health Canada allows drug makers to advertise the name of a drug without stipulating what it is for, or advertise about a medical condition, while urging sufferers to seek medical help. Health Canada does not require cable or satellite TV operators to block drug ads on U.S. channels available to Canadians.

Only two countries - New Zealand and the United States - allow direct-to-consumer drug advertising.

In the latter, it's a multibillion-dollar enterprise; it is estimated the U.S. pharmaceutical industry spends about $5 billion a year advertising prescription medications to the public through TV, radio and print ads.

Studies have suggested they do it for a reason, said Barbara Mintzes, a University of British Columbia drug policy analyst who specializes in direct-to-consumer advertising.

Mintzes said one U.S. study estimated drug companies rake in $4 for every $1 they spend on advertising. Another, by IMS Health, found on average pharmaceutical manufacturers saw a return of $2.20 for every $1 spent on consumer advertising, with that figure rising to $3.66 for every $1 spent on advertising of so-called blockbuster drugs.

"There is that kind of commonsense question about it," she said.

"Why would it have grown to be more than US$5 billion a year (if it didn't work)?"

Mintzes, who is with UBC's Therapeutics Initiative, has done a number of studies on the impact of drug ads. One in particular found people who ask their doctors for a prescription for a drug they've seen advertised are substantially more likely to leave with a prescription for that drug than people who do not ask for new medication.

Rather than seeing this new study as showing little impact of drug ads, she focused on the fact sales did increase for one of the drugs, tegaserod, which was sold by the brand name Zelnorm.

The drug, sold for irritable bowel syndrome and constipation, was the subject of award-winning ads in which the words "I feel better" were projected onto the abdomens of actresses.

A few years after being brought to market, the drug was withdrawn in both the United States and Canada amid concerns users were at greater risk of suffering a heart attack or stroke.

The two other drugs studied were etanercept (Enbrel), a rheumatoid arthritis drug administered intravenously and mometasone (Nasonex), a nasal steroid spray for allergy symptoms. There were no differences in prescribing rates for the drugs in Quebec and the rest of Canada, suggesting drug ads didn't drive up sales.

Mintzes questioned whether those two drugs were good examples to use to test the impact of drug advertising. Anyone who wanted a prescription for the arthritis drug would have had to be referred to a specialist by a family doctor - a potentially lengthy process that could deter some would-be users. And there were cheaper alternatives to the allergy drug already on the market, she said.

But Soumerai said that is in fact the study's point - that the steps one has to go through to obtain a prescription for a drug seen in a TV ad are sufficiently time-consuming and onerous that it's unlikely many people would actually follow through.

"From a behavioural science perspective, it's a reach to think that this kind of advertising would have such big effects," he said.

"You have to go through that chain. . .from seeing these ads - many of which are quite silly-seeing all these adverse effects - many of which are quite scary - taking this information to your doctor on your next visit, remembering it, talking to your doctor about it, the doctor then agreeing that they should prescribe that drug and not another, the patient then going to the pharmacy and filling the drug (prescription)."

"Those are just some of the steps that are necessary."