Letters to the Editor
Evidence, risk and the patient
- David Kault
- Aust Prescr 2007;30:47-50
- 1 August 2007
- DOI: 10.18773/austprescr.2007.054
The Editorial Executive Committee welcomes letters, which should be less than 250 words. Before a decision to publish is made, letters which refer to a published article may be sent to the author for a response. Any letter may be sent to an expert for comment. When letters are published, they are usually accompanied in the same issue by their responses or comments. The Committee screens out discourteous, inaccurate or libellous statements. The letters are sub-edited before publication. Authors are required to declare any conflicts of interest. The Committee's decision on publication is final.
Editor, – The article, 'Evidence, risk and the patient' (Aust Prescr 2007;30:47–50) shows the limitations of statistics in medical decision making. While we would like a p-value to answer the question, 'How likely is it that the results are 'for real' and not just due to chance?' this is not the question that the p-value answers. Instead, it answers the question 'If we wanted to blame chance for the results, what sort of chance would we be blaming?'
Consider a trial of the power of anonymous prayer to improve the recovery of patients in coronary care units.1This was summarised in the Australian medical press as concluding that prayer works, but '[t]here was a one in 25 chance that the better outcomes had arisen by chance'.2This misinterpretation of a p-value of 4% implies that there is precisely a 96% chance that there is a God responsive to prayers. What has actually been discovered is that the prayed-for group recovered a little faster to an extent which would be explained by atheists as the outcome of a 4% chance and which would be regarded as anything but chance by the religious.
The calculation of the 'number needed to treat' also has its limitations. In chronic conditions, people who receive the additional treatment may all have an event delayed by a few months, but if the data are arbitrarily presented so that we are told that an extra 10% survive for five years, this implies only 1 in 10 has benefited.
Evidence-based medicine has generated a lot of suspicion amongst 'rank and file' doctors. This is understandable, because if statistics are misunderstood and the clinical context is ignored, bizarre assertions can result. For example, the pronouncements that there is no evidence to support cleaning the skin before administering injections.3
Adjunct lecturer in mathematics
Casual lecturer in epidemiology
James Cook University
Adjunct lecturer in mathematics, Casual lecturer in epidemiology, James Cook University Townsville, Qld