Analytical trend troubles scientists
In 2010 two research teams, one from Queen’s University Belfast, the other from the University of Oxford, separately analysed data from the same UK patient database to see if widely prescribed osteoporosis drugs increased the risk of oesophageal cancer. But whilst the QUB analysis, led by Liam Murray and reported in the Journal of the American Medical Association, found that the increase in oesophageal or gastric cancer was 7% higher in those who took the drugs than those who didn’t, a team led by Jane Green at the University of Oxford reported in the BMJ that the oesophageal cancer risk was 30% higher for those on the drugs. The article goes on to explore some of the issues around these and other observational studies, in which scientists use statistical software and large datasets to analyse information collected previously by others and look for correlations.
Originally published on Wall Street Journal (USA) on 3rd May 2012 by Gautam Naik (link to original)