Stephen Strauss wrote articles, columns and editorials about science and technology for the Globe and Mail for more than 20 years. He has also authored three books, several book chapters, and for his efforts received numerous awards. Through all his time in journalism, he still remains smitten by the enduring wisdom of the motto of Austrian writer Karl Kraus. Say what is.

So there you are, pill in hand, about to swallow the drug your doctor has said will make you all better.

Ah, you think as the water flushes the supposedly healing pharmaceutical down your throat, good to know that there must be unbiased evidence that taking it is good for me.

Nice belief. But if you had been at a recent lecture at the University of Toronto you might well have also wished you had mastered the art of drug unswallowing. The speaker was Aubrey Blumsohn, a South Africa-born scientist who recently pinned American pharmaceutical giant Proctor & Gamble's to the public relations mat.

In 2002, Blumsohn was a researcher at the University of Sheffield in England. There, he and others were contracted by P & G to study the effectiveness of Actonel — a drug designed to prevent the fractures and bone weakening associated with osteoporosis.

The medication was locked in a market-share dogfight with Merck & Co's drug Fosamax, and there was widespread evidence that Fosamax was better at adding bone and stopping bone loss than Actonel. What the company hoped to find was that at a certain point more bone didn't translate into fewer fractures. That is to say, a biological plateau was reached in the body that rendered the two competing drugs equally effective.

Blumsohn and his compatriots collected three years worth of data from thousands of women in more than 100 international research centres. The idea was to derive from them a database that correlated women taking Actonel with women who had suffered fractures. This in turn would be correlated with changes in bone density and bone loss.

Having gathered the numbers, the Sheffield researchers shipped their raw data to P & G statisticians, who then relayed their interpretations of the data back to them. However, when it came time to publish the results — the analysis seemed to demonstrate a plateau just where the P & G wanted it to be — Blumsohn rather naively asked to see the raw data.

Since his name was going to be on the scientific paper, Blumsohn reasoned that it made sense to double-check the correlation himself. Sounds reasonable, but the company's response was an unyielding: "No way, O-bray."

"We do not intend for someone else to do the analysis. Resource-wise it wouldn't make sense … to spend a lot of money training someone else to do the analysis," they wrote in an e-mail.

Blumsohn — who was not just a someone, but the scientist whose name would be forever linked to the result — wouldn't take "buzz off" for an answer. Show me the data, he kept saying to the company, feeling he could understand what they had done without a lot of special training.

His recalcitrance made him decidedly unpopular at a university where P & G had given millions in research funding. Then he was slipped a partial set of the data by an insider and discovered his worst fears were probably true: The company had conveniently forgotten to include mention of the 40 per cent of the statistics that largely disproved their plateau thesis.

Blumsohn tried to get the university to back him up in his demand to see the whole data set, but when that failed he went to the press, and — surprise, surprise — the resulting news stories were devastating to P & G. Last month they finally agreed to let Blumsohn see all the data, which verified his original suspicions.

When the lecture ended I was thinking, it's all so Erin-Brockovich-comes-to-Hollywood lovely — save that Blumsohn has left Sheffield.

But then a member of audience member piped up with another view of what it all meant. The real issue was not whether an individual had triumphed, but could a system be put in place where drug companies with billions at stake can't cover up inconvenient truths. Wouldn't it make sense for an independent organization to be fed medical data and analyze them outside of the obviously self-interested framework of the drug companies?

Hmm, I thought, that sounds so interesting but not just for drug companies.

Quite pointedly, Blumsohn was introduced at his talk by Dr. Nancy Olivieri, a University of Toronto researcher who, several years ago, was embroiled in a data dispute with the pharmaceutical company Apotex over the effectiveness of a drug that counteracted the effects of rare blood disease.

Her rationale was that she was trying to bring out possible side effects the company was loath to talk about. However, according to a recent award-winning book by Dr. Miriam Shuchman, there was a devilish counter-spin to the Big Bad Drug Company/Little Honourable Researcher confrontation.

Dr. Olivieri apparently tried to keep contentious data to herself — in part, it seems, because she wanted Apotex to give her more money to do more studies. Moreover, later evidence suggested Olivieri's worries about the safety of the drug were, to say the least, overstated.

It seems that we, the pill-taking public, need an impartial defence against over-zealous researchers as well as drug companies. While this sort of agency has been discussed in the past, none, as far as I know, currently exists anywhere in the world.

So, I call upon our federal government to render a service to mankind by making this country the first to set up just such a truly independent drug-data analysis facility — make drug approval and drug claims contingent on having the data independently correlated.

Such a facility would protect us from scientific egos and drug company greed all at once. And, in the spirit of my last column, which honoured caution as our most universal national trait, I even have a name for the new organization: Cautious As A Canadian Inc.