0 2 M.R. Thompson et al.

surprising, therefore, that advice on the management of rectal bleeding
stresses the importance of its detection [17,18] and prompt investigation
[4,5,19].
This advice has been supported by reports that rectal bleeding has a high
predictive value for cancer in primary care [4,5,19], and that it is impossible
to differentiate between rectal bleeding from benign and malignant
disease [4,5,19,20–24]. It is further supported by the unproven assumption
that early referral of all patients with rectal bleeding will improve the survival
of those with cancer [25].
These ideas have formed the basis for the current paradigm governing
the approach to the management of rectal bleeding, which advises an
aggressive policy of full colonic examination in all patients over the age
of 40 [4,5,19]. This is partly the reason for the serious mismatch between
demand and the resources for investigation—some cancer patients have
long waits to be seen due to the unnecessary investigation of patients at
very low risk of cancer.
We question these assumptions, and propose that it is possible to classify
patients on the basis of their cancer risk for different investigation
strategies.
The British government has recently introduced the “Two-Week Standard”
[26], which promises that all patients suspected by their GPs of having
bowel cancer will be seen within two weeks. This has focused attention on
the problem, which may be partly addressed by a reconfiguration of referrals
by identifying precisely which patients should qualify for urgent referral
and investigation [27–30]. However, it is yet to be seen whether this will
cause a greater delay in patients who do not fulfil these criteria, which in
turn might exacerbate the problem with no overall benefit to all cancer
patients.The introduction of referral guidelines [27–29] must not deflect the
government from the long-term solution, which is for a substantial increase
in hospital resources for all patients requesting and needing investigation,
not just those patients at higher risk.
1. The High Prevalence of Rectal Bleeding in
the Community
The high prevalence of all symptoms in the community, regardless of their
nature, was first described in the Peckham experiment in 1946 [31] and subsequently
by Wadsworth [32] and Hannay [33].These studies demonstrated
that most people have various symptoms most of the time, which they either
self-treat or which resolve spontaneously without medical consultation.
Only a small proportion of patients who eventually consult their doctors
are referred to hospital for investigation [31–33]. This observation is also
true for rectal bleeding [34–36].
2 M.R. Thompson et al.