Tuesday, May 8, 2007

Some medical studies may be good for your health

A critical appraisal of Steve Pavlina's negative claim

Steve PavlinaSteve Pavlina is one of the very few successful bloggers who can make a living of it. His tips, mainly about success and self management, have attracted a wide readership. One of his minor but not less important issues is health.

He claims that medical studies are worthless to those who care about health. The main reason for him is money: The agenda of medical research is set by people who want to make money, he says, and the outcome is influenced by the people who spend the money on the studies. Every person is unique and therefore does not fit into any statistic. And, finally, he says that the reputation of medical science is bad because tens of thousands of people are killed by medical errors every year. After having picked medical studies to pieces, he puts forward two reasons why nobody should need them. At first, you do not need a study to tell you that smoking or being fat is bad for your health. And secondly, it is better to feel what is good for you because you (and no study authors) are the expert in your unique case.

Strong points of Steve's claim

  1. Money-driven research agenda setting. I strongly agree.
  2. Money-driven publication bias: Negative results withdrawn from publication by the funders of the study. I strongly agree, and also medical journals have recognized this problem. It has been verified by a number of studies about studies.
  3. Nonsense of blindly believing in study results while not using his own brain for thinking.
  4. Nonsense of using generalized statistics to predict the outcome of an individual case.


Needles in the haystack

As a conclusion, I widely agree with Steve as far as the vast majority of medical studies is concerned. I call them the haystack. In his post he does not mention the possibility that needles may be found there. I am convinced that they must be there, for these reasons:

  1. There are still tens of thousands of studies that are (1) not about products and (2) not initiated out of financial interests. Some of them have to be good, to say it with Technorati.
  2. Without medical studies we would not know that smoking or being fat is unhealthy. Until recently, big tobacco in China has advertised a cigarette brand «long life», some Chinese in the countryside still believe that smoking makes you live longer. Would be good for them to know the scientific facts.
  3. Other issues where medical studies may provide useful facts: Dangers of passive smoking; lacking effect of widely recommended food additives, drugs or treatments; alert of special dangers (e.g. vein thrombosis) in certain health conditions (e.g. tooth infections) - just to mention a few.


My personal rating of medical studies

In every 400 new medline entries that I see on average every day, I rate 4 as interesting and 396 as worthless to those who care about health. Of the remaining four, I select one for a post and put the others aside for later use in a topical context.

No comments: