One important safeguard against imperfect or flawed scientific reporting is peer review; i.e., scientists scrutinize each other's work in advance. Almost all well-respected scientific journals rely on peer review to select papers for publication. Any study that has not undergone peer review should be regarded with the utmost scepticism. For example, one should be wary of findings announced at a press conference that are not accompanied by publication in a journal or by a presentation at a scientific forum. At the same time, it's also true that peer review is no guarantee by itself that a study is reliable. For example, expert reviewers have no way of knowing if an investigator has falsified the data in an article. And even if a study is well-designed and scientifically valid, it may have absolutely no relevance to most people.
Thursday, August 12, 2010
Read editorials in a medical journal
One important safeguard against imperfect or flawed scientific reporting is peer review; i.e., scientists scrutinize each other's work in advance. Almost all well-respected scientific journals rely on peer review to select papers for publication. Any study that has not undergone peer review should be regarded with the utmost scepticism. For example, one should be wary of findings announced at a press conference that are not accompanied by publication in a journal or by a presentation at a scientific forum. At the same time, it's also true that peer review is no guarantee by itself that a study is reliable. For example, expert reviewers have no way of knowing if an investigator has falsified the data in an article. And even if a study is well-designed and scientifically valid, it may have absolutely no relevance to most people.
Tuesday, August 10, 2010
Scrutinize the results - results of one study usually don’t provide enough proof
Third, scrutinize the results. Does this information reveal a direct cause-and-effect relation between two factors? Or is it merely an association? For example, someone could argue there’s an association between matches and lung cancer because matches light the tobacco that causes lung cancer. But common sense would tell you that lighted matches don’t cause lung cancer. Typically, years of consecutive studies are required to prove a cause-and-effect relation and the results of one study usually don’t provide enough proof. If just one medical study has documented an unusual or peculiar finding, and if the results have never been replicated by any other study, then this situation suggests that the study is not reliable!
Monday, August 9, 2010
Look beyond the statistics
Second, look beyond the statistics. When reports hurl at you statistics like ‘a 30 per cent increase’ or phrases such as ‘a 50 per cent higher risk,’ take a closer look at the exact numbers. Many of us get ‘turned off’ by numbers, but this attitude can prove dangerous: you need to ask yourself what the numbers really mean and how they apply to you? Benjamin Disraeli once remarked that there are three kinds of lies: lies, damned lies and statistics. Remember that statistical methods are simply tools, and they can produce blatantly wrong conclusions unless sensibly used. One common way in which statistics can be misleading pertains to reporting of relative and absolute risk. A headline that screams ‘X Doubles the Risk of Y’ is way off target if your chance of contracting ‘Y’ is one in a million to begin with. Doubling the risk of ‘Y’ only makes it two in a million! The relative risk (doubled) is nothing to worry about if the absolute risk (an increase from one to two in a million) is tiny.
Friday, August 6, 2010
How does one interpret or understand the medical stories?
We are now awash in a flood of health information, and barely a day goes by without a report of a spectacular new cure for a formerly incurable illness. However, many people find themselves increasingly frustrated in the face of the media barrage of confusing and contradictory health advice. One day, drinking alcohol is bad for your health; the next day it is reported to help prevent heart disease. One day, margarine is healthier than butter; the next day it's not. One day fish prevents heart disease, then it doesn't! You may be exasperated enough to ask: Why can't researchers get their facts right the first time? And how are you supposed to make sense of what you read, if the experts themselves can't make up their minds?
In order to maintain a balanced perspective, it's important to remember that news, by its very definition, implies something new and unusual. This is why medical stories in the media often seem to be at loggerheads with what common sense tells us. After all, the hundredth study showing a relationship between high cholesterol and heart disease is hardly news, but the one study that shows that eating fat helps prevent heart disease is likely to become a headline — no matter how flawed it may be! The media is often guilty of oversimplifying or exaggerating results. Moreover, headline writers may focus on an angle that gives a distorted impression, which often means that facts are sacrificed at the altar of readability or circulation figures.
Editors crave for stuff which is ‘new’ and doctors and hospitals are only to happy to tom-tom their latest gadgets and gizmos. Reporters are often not specialized enough to understand the medical technical background. Often, they do not do their homework properly, which results in misreporting, which is, unfortunately, a common occurrence in