Risk, statistics, and the individual
BMJ 2003; 327 doi: https://doi.org/10.1136/bmj.327.7417.757 (Published 25 September 2003) Cite this as: BMJ 2003;327:757All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I remain deeply frustrated by the claims of some doctors to treat
patients as individuals and that by implication those of us who utilise
the results of published information do not. Of course we are all
individual human beings and treatment and management in the social sense
should tailored to that individual. But assessment of risk and benefit in
the complexity of human physiology can only be made in the presence of
information.
To illustrate let me offer the example of streptococcal sore throat. In
the absence of information to the contrary the risk of anaphylactic shock
resulting from treatment with penicillin is small but real. Would you
warn every patient that you give penicillin to of this possibility? Would
it influence their request for treatment? Would it influence them in a
sensible way? How should we impart that information without being
paternalistic or too statistical? I am treating this patient on a
statistical risk/benefit analysis which says serious adverse side-effects
are rare but am I treating them as some special individual, I think not.
However if we ask the patient whether they are allergic to penicillin or
indeed if we have documented knowledge of a previous adverse reaction then
we have increased information about the risk/benefit in that individual
and would be foolish to treat with penicillin. In neither case is the
patient treated as anything other than an individual but the likelihood of
any particular outcome is determined by our knowledge of the individual
factors which affect that outcome. This is no more than to state the
obvious.
For many conditions these individual factors are simply too complex to
know in precise depth and may well require a detailed elucidation of the
genetic code of any given person at the very least.
What the outcome cannot be determined by is gut feeling because this is
not related to information about the patient - unless it is used to
describe a summative subconscious assessment of all the available
information which might apply. If that is what the best clinician is
acheiving then they will out perform any protocol one can write and
furthermore they would probably be able to retire if they applied their
talent to the Stock Market.
I'm afraid I for one will have to stick with taking a history, making a
diagnosis, listening to the patients fears and aspirations and trying to
keep abreast of what the literature tells me is the likely response of an
individual with given characteristics, to therapy which has a statistical
chance of such and such an outcome when applied to a group of patients
with specific and probably different characteristics!
Competing interests:
None declared
Competing interests: No competing interests
Well indeed, everybody wants the right answer including the treating
physician but can only have an appropirate answer in this world of
uncertainities.
Every effort to remove this uncertainities is the way out, and that is by
appropirate trials.
Competing interests:
None declared
Competing interests: No competing interests
Well done, Dr Shelford, you have reminded us all of our first duty
beautifully. "Statistics make bad cases" and our job is to give the
patient the best investigation and management we can without blinding him
(and ourselves!) with science and dodgy stats. Ultimately the patient's
decisions about treatment for serious diseases will end up as a gut-feel
response after a cosy chat with a trusted doctor and family/friends:
nothing much rational in that.
Dr. Shelford's patients are fortunate indeed!
Competing interests:
None declared
Competing interests: No competing interests
Managing patients with low risk rectal bleeding
Dear Sir
Dr Shelford’s article ‘Risk, statistics, and the individual’ (BMJ
2003; 327:757) emphasises the dilemma that general practitioners have in
managing patients with rectal bleeding; not delaying the very few with
cancer while protecting the majority of patients from the worry and
occasional harm of unnecessary investigation. Although Dr Shelford has
quite rightly identified a major reason for introducing the two-week
standard was a lack of hospital resources, until these are readily
available patients will have to be prioritised so that those at higher
risk are seen more promptly.
In Dr Shelford’s case it is quite possible that he could have been
referred to a two-week standard clinic, as 95% of patients with rectal
bleeding from a bowel cancer have either an associated change in bowel
habit to increased frequency and/or looser stools or a palpable rectal
mass, either of which would have qualified him for referral to the fast-
track clinics (BMJ 2003; 327:263-265). Only 1% of all bowel cancers occur
in patients below the age of 40, and patients in this age group presenting
with low risk rectal bleeding comprise only 0.02% of all bowel cancers.
99.9% of patients with rectal bleeding in the community do not have cancer
(Thompson JA, Pond CL, Ellis BG, Beach A, Thompson MR. Rectal Bleeding in
General and Hospital Practice; 'The tip of the iceberg'. Colorect Dis
2000; 2: 288-293) and it would be difficult to advocate the prompt
investigation of all these patients.
However in Dr Shelford’s case because of the very understandable
anxiety arising from his positive family history, he could have been
referred in the ‘third way’ of referral, that is an urgent appointment in
a routine clinic (BMJ 2003; 327: 757).
Finally, as our article emphasises, ‘treat, watch-and-wait’ policies
should only be with the consent of the patient, and if a patient is
concerned for whatever reason they should be referred promptly, but to a
routine clinic if they have low risk symptoms.
Yours faithfully
M R Thompson, I Heath, B G Ellis, E Swarbrick, L Faulds Wood and W S
Atkin
Competing interests:
None declared
Competing interests: No competing interests