This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

CAMHS won't see you now

The two-week rule – effective oncology referrals

The two-week referral rule for suspected cancer is one of the

most controversial and least evidence-based targets –

Dr Stephen Whitaker looks

at the arguments and offers

practical advice

Why this topic is in the news

lOne in three cancer cases is missed by GPs – and referred non-urgently – despite the Government's high-profile two-week rule, according to a

National Audit Office report in March.

lMany GPs admitted they had never even read the referral guidelines –

and others who had seen them said they were flawed.

lPatients in England are still diagnosed with cancer at a more advanced stage than the rest of Europe.

Why is there a two-week rule?

It is now more than five years since the Government white paper 'The New NHS – Modern, Dependable' guaranteed that anyone with suspected cancer would be seen by a specialist within two weeks of urgent referral by their GP.

The referral guidelines1 for suspected cancer were published after consultation in 2000 and are beginning to be reviewed with the experience gained, such as in colorectal cancer.

There has been great effort expended in implementing the process by both primary and secondary care. Why so much effort has been focused on this element of a patient's 'cancer journey', rather than the other elements of time from development of cancer to effective treatment, is not clear.

It was felt to be an extremely anxious time for patients, but there is little scientific evidence that reducing the time from referral to being seen would either reduce overall waiting time or improve cancer treatment outcomes.

The original Department of Health guidance was based on the professional opinions of tumour-specific working parties and suggested the evidence would come from reviewing the two-week rule (TWR) implementation itself, although not in a scientific, randomised trial.

Why is the rule controversial?

No one would argue that a patient with suspected cancer – or any potentially serious condition needing secondary care – should not be seen urgently.

The controversy arises from the perceived inaccuracy of patient selection and therefore GP referral. The receiving specialist expects a high proportion of cancers in the TWR referrals, while most audits show a very low proportion – around 1-2 per cent for head and neck cancers, though higher for breast and colorectal cancer.

More important is the appropriateness of TWR referral, which also varies from 32 per cent of skin TWR referrals to better than 80 per cent for upper GI and urological referrals2,3.

The majority of new cancers are therefore seen outside the TWR system. Patients with head and neck cancer are often seen urgently through a more traditional method of receiving specialist's prioritisation, with an 80 per cent pick-up rate of cancer4.

However, if the purpose of the TWR is to improve the survival and morbidity from cancer – not just alleviate anxiety in the worried well and achieve a political target – then patients must be referred as early as possible, hence with minimal symptoms.

Of course, if the policy had been implemented with extra resources (for TWR clinics in each specialty) the inevitable knock-on effect of increasing waiting times for non-urgent referrals, and delaying cancer diagnosis in those patients without classical symptoms, would not have been seen.

This prolongation to diagnosis for non-urgent patients who make up the majority of new cancers, except breast cancer, may explain the National Audit Office recent findings that patients have more advanced cancer at diagnosis in Britain compared with the rest of Europe5.

More effective TWR referrals

This will come from audit and public education. Comparison of patients diagnosed with cancer as TWR

referrals and non-TWR referrals will

reveal which symptoms are most predictive of cancer and therefore warrant TWR referral.

It is likely that combination of factors is most important – for example, rectal bleeding in under-40s is unlikely to be cancer (1:500), compared with asymptomatic over-60s where the incidence of rectal cancer is 1:400.

Upper GI cancer

The Department of Health referral guidance is shown in table 1.

The incidence of oesophageal cancer has increased more than 50 per cent for males in the last two decades to around 14:100,000 in 2001, with a smaller increase for women6.

It is worth remembering that 99 per cent of upper GI cancer occurs over 40 years. The chance of a dyspeptic patient under the age of 55 having gastric cancer is extremely rare.

In some Cancer Networks there may be local agreement that urgent referral for endoscopy may satisfy the TWR referral and will also achieve a diagnosis.

Local audits show that as little as 17 per cent of upper GI cancers are seen via TWR referral.

Also, many of the trigger features in the guidance are symptoms of advanced disease, where urgent diagnosis is unlikely to benefit chance of survival.

Urological cancer

The referral guidance is shown in table 2. Haematuria in the presence of bladder or renal malignancy may be intermittent and repeat testing for microscopic haematuria may be negative.

Some further interpretation of the guidance may avoid some non-malignant referrals.

Urothelial cancer is more likely in males over 50 who are smokers.

The incidence of prostate cancer has trebled since 1971 due to PSA screening5.

While there is no national screening policy for prostate cancer, the role of PSA screening is covered by 'The Prostate Cancer Risk Management Programme' circulated to GPs.

Uncertain testicular and renal abnormalities may be investigated initially by locally available urgent ultrasound that can also select appropriate cancer referrals.

Skin cancer

The incidence of malignant melanoma has risen five-fold since 1971 and is now around 10:100,0006.

It is probably more important for early referral and diagnosis in melanoma than most other cancer types.

Survival at 10 years drops from 90 per cent for <1mm invasion="" to=""><40 per="" cent="" for="" melanomas="">4mm depth invasion.

Table 3 lists the clinical features that should trigger TWR referral for suspected melanomas. Some interpretation of these guidelines is useful, however. For example, if inflammatory changes such as itching are very recent, of a few days, or follows clear trauma, this is most unlikely to be significant. It would be appropriate to review the mole after two-three weeks.

A large number of malignant melanomas are referred through a non-urgent route. While ulcerated nodular melanomas are fairly obvious (figure 1) more subtle features such as colour change (darkening or depigmentation or mixed – figure 2) are more difficult to recognise.

Having access to photographs showing the variations in appearance can help with confidence in urgent or non-urgent referral.

However, sending a photo with the referral is of variable use – being very dependent on image quality and photographic technique.

Patients with non-healing ulcerated lesions, enlarging over months and where there is significant induration may have squamous carcinomas (figure 3).

More superficial crusting may be just keratosis or basal cell carcinomas (the latter are not subject or require urgent TWR referral).

Breast cancer

The incidence of breast cancer continues to rise from 65:100,000 in 1971 to 114:100,000 in 20015. While the value of breast self-examination remains controversial in terms of improving

survival from breast cancer, women should be encouraged to continue, as early detection allows more conservative treatment.

Typical results from symptomatic referrals to a breast clinic suggest around half of breast cancers are referred as TWR, and half through routine referral. Around 13 per cent of TWR referrals are cancer while only 3 per cent of non-TWR are cancer, suggesting reasonably accurate GP selection.

Table 1 Upper GI cancer – urgent referral

lDysphagia – food sticking on swallowing (any age)

lDyspepsia at any age combined with one of the following 'alarm symptoms':

•Weight loss •Proven anaemia •Vomiting

lDyspepsia in a patient aged 55* or more with at least one of the following high-risk features:

•Onset of dyspepsia less than one year ago

•Continuous symptoms since onset

lDyspepsia combined with at least one of the following known risk factors:

•Family history or upper GI cancer in at least two first-degree relatives

•Barrett's oesophagus

•Pernicious anaemia

•Peptic ulcer surgery more than 20 years ago

•Known dysplasia, atrophic gastritis, intestinal metaplasia


lUpper abdominal mass

lAge 55 is considered to be the maximum age threshold. Local Cancer Networks

may elect to set a lower age threshold, such as 50 or 45

Table 2 Urological cancer – urgent referral

lMacroscopic haematuria in adults

lMicroscopic haematuria in adults over 50

lSwellings in the body of the testis

lPalpable renal masses

lSolid renal masses found on imaging

lAn elevated age-specific PSA in men with a 10-year life expectancy

lA high PSA (>20ng/ml) in men with a clinically malignant prostate or bone pain

lAny suspected penile cancer

NB: PSA testing of symptomatic men or screening for prostate cancer is not national policy. It is recommended that a PSA test, except in men clinically suspicious of prostate cancer, should only be performed after full counselling and provision of written material.

Table 3 Skin cancer – urgent referral


lPigmented lesions on any part of the body which have one of more of the following features:

•Growing in size

•Changing in shape

•Irregular outline

•Changing colour

•Mixed colour



NB: melanomas are usually 5mm or greater at the time of diagnosis,

but a small number of patients with very early melanoma may have lesions of a smaller diameter

Squamous cell carcinoma

lSlowly growing, non-healing lesions with a significant induration

on palpation (commonly on face, scalp, back of hand) – with documented expansion over a period of one-two months

lPatients in whom squamous cell carcinoma has been diagnosed

from a biopsy undertaken in general practice

lPatients who are therapeutically immunosuppressed after an organ transplant have a high incidence of skin cancers, mainly squamous cell carcinoma; these tumours can be unusually aggressive and metastasise; it is therefore recommended that transplant patients

who develop new or growing cutaneous lesions should be referred

under the two-week rule

Table 4 Breast cancer – urgent referral

Urgent referral:

lPatients with a discrete lump in the appropriate age group (for example age >30)

lSigns highly suggestive of cancer such as:


•Skin nodule

•Skin distortion

•Nipple eczema

•Recent nipple retraction or distortion (<3>

Conditions that require referral – but not necessarily urgently:


lDiscrete lump in a younger women

(for example age <>

lAsymmetrical nodularity that persists at review after menstruation


lPersistently refilling or recurrent cyst


lIntractable pain not responding to reassurance, simple measures such as wearing a well-supporting bra and common drugs

Nipple discharge

lAge <50 with="" bilateral="" discharge="" sufficient="" to="" stain="">

lAge <50 with="" bloodstained="">

lAge >50 with any nipple discharge


1 Referral Guidelines for Suspected Cancer. DoH 2000.

Available at

2 Jones R et al, 'Is the Two-Week Rule for Cancer Referrals Working?' BMJ 2001;322:1555-1556 and related correspondence

3 Mearing-Smith T BMJ 2003

4 R Sudderick, personal communication

5 National Audit Office, 'Tackling Cancer in

England: Saving More Lives'. The Stationery Office,

March 16, 2004. Available at

6 Cancer Research UK. Statistics 9/2/2004.

Available at

Stephen Whitaker is consultant clinical oncologist at St Luke's Cancer Centre, Guildford

A GP's perspective

GP comment

Bottom line, as usual, is more resources

Is the two-week rule little more than window dressing by the Government? asks Dr Robert Bailey

Five years on from the implementation of the two-week rule for suspected cancer referrals – and where are we? Perhaps we thought this would be an end to delayed diagnosis. And yet we still lag behind most of Europe on the early diagnosis of cancer. Why hasn't it worked better?

Like most GPs, I welcomed the prospect of being assured that a patient with suspected cancer would always be seen expeditiously. Well, OK, there are referral guidelines to be followed, but if we take care and keep to them, surely problem sorted? So why are we still failing to refer urgently many patients who eventually prove to have cancer?1

Some guidelines are straightforward. The guidelines for suspected upper GI malignancy have always seemed to me to be clear, even if the goalposts do shift from time to time. But others are less so. It is relatively simple to detect a discrete lump in an 80-year-old but not so easy in a 35-year-old who's always had lumpy breasts. In primary care we often have on the agenda our role as gatekeepers, holding back the tide of potential referrals to our colleagues in secondary care. Perhaps we try too hard, and in consequence may not always refer when appropriate. That's the inevitable consequence of trying to ration and refer appropriately at the same time. Then there is that irrational factor of patient pressure.

The dynamics, and therefore the outcome, of a 10-minute consultation are always prone to this. A consultant once said to me the only way to prevent delay in breast cancer diagnosis is for all referrals to be seen within two weeks; and there's the rub. Two-week waits have been implemented at the expense of seeing non-urgent referrals further delayed2.

We must accept that even with guidelines we are not always picking up suspected cancers with the resources and skills we have in primary care. So the answer must be either to improve the skills and resources we have, or alternatively to reduce the wait for all referrals – which is I guess how our European colleagues manage it. We must keep trying harder, but I suspect this is not enough.

Referral guidelines alone will not pick up all cancers anyway, especially early cancer. The cynics would refer all patients urgently and let secondary care sort out the mess, but most of us try to keep a balance.

The implementation of the TWR has been popular with patients in theory, but has it proved to be a false dawn? Although the NHS Cancer Plan sets out specific targets3 for referral to treatment times, these are not being met. This is inevitably frustrating for patients. So, is the TWR little more than window dressing by the Government? The figures would argue strongly for this viewpoint2.

The bottom line is that, as always, we need more resources; in health education; in primary care to ensure more appropriate referrals; and in secondary care so those referrals can be dealt with more expeditiously.

Robert Bailey is a GP in Peterborough and a hospital practitioner in breast cancer

1 National Audit Office: Tackling cancer in England: Saving More Lives, London, Stationery Office March 2004

2 Cancer Research UK Release, July 2003

3 NHS Cancer Plan. London, Stationery Office

September 2000

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say