Intended for healthcare professionals

Letters Risk illiteracy again

Put statistics back in its bottle

BMJ 2011; 343 doi: (Published 12 July 2011) Cite this as: BMJ 2011;343:d3975

Convoluted statistics and confused clinicians

While clinicians are being urged to gain more statistical knowledge,
I do wonder why 'statistical methods and results' are not more user
friendly (1, 2). It is a shortcoming of current statistical methods that
results are NOT easily understandable by many ordinary clinicians. (Some
notable exceptions eg: NNT number needed to treat) (3).

One need not look further than the widely used 'p' values to
recognise how convoluted most statistical results are. I quote verbatim
from the Wikipedia article on P value (4): 'In statistical significance
testing, the p-value is the probability of obtaining a test statistic at
least as extreme as the one that was actually observed, assuming that the
null hypothesis is true. One often "rejects the null hypothesis" when the
p-value is less than the significance level ? (Greek alpha), which is
often 0.05 or 0.01. When the null hypothesis is rejected, the result is
said to be statistically significant'. (Yes, Clinicians are more likely to
read a Wiki article rather than a Statistics text book !)

In trying to educate clinicians about complex statistics, the analogy
from computing world is perhaps relevant here. Steve Jobs (Apple Inc) did
not design products which are more complex than existing products (5).
Rather the recent success of Apple products is more due to its user
friendly applications and apparent simplicity. Perhaps the field of
medical statistics needs someone as inspiring as Steve Jobs to devise
statistical tests whose results are easily understandable, generalizable
and clinically applicable.

If patients cannot grasp any information given by clinicians, it is
certainly not the fault of the patients. We do not blame the patients for
their ignorance and exhort them to learn more about their disease. The
same principle applies to clinicians; perhaps the statisticians should try
a bit harder to educate the ignoramuses.

Finally, if clinicians find it difficult to understand statistical
concepts and results, what hope is there for the lay people?

I would prefer 'medical statistics' to be more like my laptop. I
would prefer to concentrate on my 'sound bites' rather than 'sound cards
and mega bytes'


1. Heath I. Dare to know: risk illiteracy and shared decision making. BMJ
2011;342:d2075. (6 April.)

2. Shuster S. Risk illiteracy again. Put statistics back in its
bottle. BMJ 2011;343: doi:10.1136/bmj.d3975 (12 July 2011)

3. Cook RJ, Sackett DL. The number needed to treat: a clinically
useful measure of treatment effect.BMJ 1995; 310:452-454.

4. P value. (accessed 17 June

5. Fortune Magazine. Steve Jobs: CEO of the decade.
(accessed 17 June 2011)

Competing interests: (1).Biological sciences interested me more than mathematics at school. (2).iPad and BMJ app used to read the relevant BMJ articles but this letter was drafted using a Windows 7 laptop.

18 July 2011
S Sundar
Consultant Oncologist
Nottingham University Hospitals NHS trust