On 'Bad Science'

span style="font-style: italic;"Bad Science/span by Ben Goldacre is a educating read. I liked the book as he is good at explaining to non-scientists like me how science is misused and misrepresented. He also has a a href="http://www.badscience.net/" blog /a.br /br /In a modern, technological society it is important that as many people as possible have at least a basic grasp and understanding of the scientific knowledge that makes this society possible.br /br /He illustrates the importance of evidence and clinical trials in medicine. Methodology is key to how reliable a trial is – and to what conclusions can be drawn from it. It is through evidence-based medicine that we have been able to achieve the great improvements in health and in treating and preventing disease over the last couple of centuries.br /br /He verbally dissects the dubious claims from homeopaths, nutritionists and other vendors of “complementary” or “alternative” medical remedies. Why they are ‘alternative’ is because they have not been able to demonstrate in proper, neutral scientific tests that they have much of a positive effect. He is particularly good at verbally skewering some of the claims of Gillian McKeith.br /br /The placebo effect seems to be notable in some studies. This is why some ‘alternative’ treatments may work – since they are acting to reassure the patient psychologically. The same effect exists when people in clinical trials are given placebos.br /br /The misunderstanding of statistics is a major problem. People don’t necessarily understand what the risk of something is and how it compares to other risks. Human cognitive biases make our understanding of probability rather iffy. We overestimate the risk of some (unusual) events and understate the risks of day-to-day life.br /br /An example of where the misunderstanding of probability was key to a court case was the case of Sally Clarke. This was a woman accused of murdering her children, convicted for this and then freed on appeal. The prosecution in her case calculated the chance of a child dying of cot death (less than 1 in 8000) and squared that to give the probability that both children had died of cot death – which was thus given as 1 in 73m. However, the key error – given that the children had already died – was taking account of the general low prevalence of cot death. A better approach was to compare it to the chance that the two children had been killed by their own mother. And, furthermore, cot death events may not be independent – the same environmental or genetic factors could be causing the death of both siblings. Therefore, it may have been more appropriate to compare the fraction of children dying of natural causes to the fraction killed by their parents. In such a statistical comparison, the evidence would not have looked so damning for Ms Clarke.br /br /Goldacre is particularly scathing about the media. They report science stories in a simplistic and sensationalist manner. Even when they have specialist science reporters, when a story gets big they give it to generalists. This can be seen with the MMR furore. The MMR furore – like, in an even more dramatic example, the furore about retrovirals to alleviate AIDS – is a furore that has cost lives. By putting people off using retrovirals and putting people off vaccinating their children, cranks have led to more people dying than would otherwise have been the case. Such is the human cost of ignorance and disregard for scientific trials and evidence.div class="blogger-post-footer"img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/38622711-1945552336712941590?l=vinospoliticalblog.blogspot.com'//div

technorati tags:
political news | news | world news
More at: News 2 Cromley