The decisions you make are important, but maybe not as important as the ones made for you by other people (following Mitt, we confer people status on corporations). If a pharmaceutical company decides to market a new drug and the new, and expensive, drug supposedly treats a disease you have, then how do you decide whether to spend money for it or go for an alternative? How do you make the best decision for yourself?

Typically a person would defer to a trained expert to make that decision (assuming you have decided that person is an expert, another decision). But how does a physician decide which drug is best for you? Let us assume that this is a motivated and intelligent physician. Then perhaps that decision is based on reading the test results for this new drug in the literature. That would good, but only if the test results are meaningful and not edited or manipulated in ways unfavorable to you.

Decisions Based on Test Results?Recently I saw an excellent TED video by Ben Goldacre. He points out several ways that pharmaceutical companies present test results such that making an informed decision is difficult or impossible. For instance, positive results are usually reported when sponsored by a company, but negative results are lost somewhere. Negative results are much more often reported by independent researchers trying to confirm previously reported results by a company. The video indicates negative results are four times more likely to be reported by independents. This does not mean the companies are inherently dishonest. They simply have economic pressures to publish results favorable to them and no economic pressure to publish negative results.

Testing for missing negative data is straightforward and only requires some elementary statistics. Ben demonstrates how to do this in the video.

Another interesting point of the video is this question raised: what is the basis of saying a new drug is effective? It turns out that a new drug’s effectiveness can be reported as compared to an equivalent dose of a placebo (typically a sugar pill). That is, the new drug is reported as effective compared to nothing. Ben and I agree that we would rather have the new drug rated compared to the best available alternative on the market rather than comparing it to a sugar pill.

I dwell on the methods of pharmaceutical companies only because that was the major topic of the TED video. Other than being fantastically profitable, there is probably not a lot of difference between them and other major industries. (We know they must be profitable because we are nightly bombarded with television commercials for various pharmaceutical products. I consider that a better measure than wading through massaged financial statements.) The lesson to be learned here is not that pharmaceutical companies are evil creations (or people), but that making the right decision based on corporate data is more difficult than necessary because of the economic pressures in play. Take a look at the video and see if you agree.