Intended for healthcare professionals


Referral and diagnostic process in suspected colorectal cancer needs to be improved to achieve two week target

BMJ 2000; 321 doi: (Published 16 December 2000) Cite this as: BMJ 2000;321:1527
  1. Coliette Riesewyk, diagnostic service manager,
  2. Chris Hayward, specialist registrar,
  3. Veda Enser, divisional manager,
  4. John Northover, consultant surgeon (j.northover{at}
  1. St Mark's Hospital, North West London Hospital NHS Trust, Harrow, Middlesex HA1 3UJ

    EDITOR—In response to the government initiative to improve the diagnosis of cancer, and using the recently published referral guidelines for suspected colorectal cancer, we have completed a three month pilot study looking at the introduction of the two week target at the North West London Hospitals NHS Trust.

    Consultation occurred with general practitioners in the pilot area's primary care group. The general practitioners were asked to use a specific two week target form, which contained six symptom categories, when they referred patients. The completed form was faxed to St Mark's Hospital, and an outpatient appointment was then given within the two week target time. Each patient was seen by a consultant surgeon or surgical specialist registrar. The hospital specialist, blinded to the general practitioner's form, completed a similar form, which had an additional category—“patient does not fulfil any of the above criteria.”

    Altogether 364 clinic slots were reserved for patients referred under the two week target scheme, on the basis of calculations for projected appropriate referrals for a population of 500 000. However, just 20 patients were referred during the three month pilot period. General practitioners and specialists categorised five patients identically (two of these patients had a letter attached to the fax, which the specialist read in the clinic despite the protocol). Six patients were categorised by specialists as not fulfilling any of the criteria for urgent referral. All patients were seen within two weeks. Thus far, four cancers have been diagnosed—three adenocarcinomas and one squamous cell carcinoma. During the pilot period a further seven cases of cancer from the pilot area were diagnosed that had not been referred via the two week target process.

    This preliminary experience suggests that, despite close liaison with the primary care group, the referral process will take time to be adopted and implemented. Despite clear referral guidelines there was massive underuse of the facility, considerable variance between the general practitioner's and specialist's assessments of symptoms, and a high rate of apparently inappropriate urgent referral. Unless these outcomes are improved dramatically, the efficiency of the referral and diagnostic process in suspected colorectal cancer would not improve for most patients. The audit is continuing.