ABSTRACT
PURPOSE: The UK government target expects all suspected colorectal cancer (CRC) patients to be seen within the Two-Week Referral (TWR) system made by general practitioners. These guidelines originally derived from only level 5 evidence. However, this has significant impact on the workload for colorectal surgeons. The aim of the study is to investigate the effectiveness of this colorectal service and whether the referral criteria are predictive of CRC. METHODS: A retrospective study of all patients referred under the TWR guidance in 2010 was assessed. The first 573 TWRs were piloted for analysis. Clinical information from each patient was collected regarding TWR criteria and additional colorectal symptoms or risk factors. Multiple regression analysis was performed to determine which symptoms independently correlated with CRC. RESULTS: One hundred twenty-six CRCs were diagnosed via all methods of referral in 2010. There were 940 patients referred under the TWR guidelines in that year, when 50 CRC patients were identified. Amongst the 573 patients, 32 CRCs were diagnosed. Multiple regression analysis revealed tenesmus to be independently associated with CRC (p = 0.003, Pearson's r = 0.09185). None of the individual TWR criteria confidently predicted CRC. CONCLUSION: Our preliminary results suggest that the current TWR guidelines cannot effectively predict CRC. There is an urgent need for an evidence-based approach to referral criteria for suspected CRC.
