Balancing probabilitiesBMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d3048 (Published 17 May 2011) Cite this as: BMJ 2011;342:d3048
All rapid responses
Reading the correspondence[1-4] arising from Schuster's argument,
"the real problem is the biomedical ignorance of statisticians", I am
provoked into repeating the comment I made last year.
'I was told in 1989 by the late Victor Chang, after my coronary
bypass surgery, that I had "a 70% chance of living for 10 years." What did
that mean? It really meant that I was in one or other category of survivor
- those 30 who would die within 10 years or those 70 who would live longer
than a decade. The truth was that Victor did not know which group I was in
- but he didn't say so.'
'Is it not time to give up our antediluvian pretence at omniscience
and learn to tell our patients the truth - "I don't know?" Is it not time
that we conditioned ourselves, and our patients, to learn that life is a
lottery, that we are all compelled to take chances, that no human action
is risk-free and that we are, quite simply and honestly, unable to
truthfully answer many of their questions?'
No amount or detail of statistical theory nor of statistical
understanding can tell the individual patient what will or will not happen
to them. It is that which is their concern and none of us knows the
 McNulty SJ and Williams P. Skill of interpreting imperfect
investigations. BMJ 2011;342:d3026 (21 May)
 Let's work together. Hemming K. BMJ 2011;342:d3030 (21 May)
 Barraclough KP. Come clean if you don't know. BMJ 2011;342:d3044
 Rathbone P. Role of cognitive bias. BMJ 2011;342:d3047 (21 May)
 Shuster S. The real problem is the biomedical ignorance of
statisticians. BMJ 2011;342:d2579. (21 April.)
 Arnold P. I don't know. BMJ blog. 14 June 2010.
Dr Peter Arnold (retired Sydney GP)
PO Box 280,
NSW, Australia firstname.lastname@example.org
Competing interests: No competing interests