Medical journals contain statistical errors, deficiencies: expert

Former Dean Postgraduate Medical Education, Cambridge University, UK, Professor Dr John Biggs has said that almost half of the literature published in medical journals around the world is leading people astray because the standards in the use of statistics in medical research are generally low.
This was stated by him while addressing the inaugural session of a workshop on "Biological Data Analysis with Hands-on Training on Statistical Software", here at the University of Health Sciences (UHS). The 4-day workshop has been organised in collaboration with Higher Education Commission.
According to him, a growing body of literature points to persistent statistical errors, flaws and deficiencies in most medical journals. Careful and accurate use of statistics in medical research was of major importance and, therefore, must be enforced emphatically.
He said: "The use of statistics in medical diagnosis and biomedical research may affect whether individuals live or die, whether their health is protected or jeopardised, and whether medical science advances or gets side-tracked." Professor Biggs further said that problem was a serious one, as the inappropriate use of statistical analysis might lead to incorrect conclusions, artificial research results and a waste of valuable resources. According to a research, serious statistical errors were found in 40 percent articles published in a renowned psychiatry journal, he added.
UHS Vice-Chancellor Professor Dr Malik H. Mubbashar said on the occasion that medical researches need to be encouraged to learn more about statistics. Statisticians should be involved early in study design, as mistake at this point could have major repercussions, negatively affecting all subsequent stages of medical research, he added. He further said that inappropriate use of statistical method and technique caused loss of time and cost and it could be misleading to other scientific researchers.
The facilitator of the workshop Waqas Sami said that inappropriate use of statistical methods was found in every stage of medical research related to data analysis, design of the experiment, data collection and pre-processing, analysis method, implementation and interpretation. He said that Pakistani researchers lagged behind in research publications because of not having enough statistical knowledge.