Diagnostic value of self-reported symptoms in Danish outpatients referred with symptoms consistent with colorectal cancer

Abstract

Objective: To assess the association between self-reported symptoms and a diagnosis of colorectal cancer (CRC) in symptomatic outpatients without CRC risk factors, defined by a Danish expert group.

Method: A cross-sectional study of patients aged 40 years and older referred by general practitioners to two Danish surgical outpatient clinics for symptoms consistent with CRC during a 16-month period. CRC was diagnosed at endoscopy and through follow up. Before their first appointment, participants completed a questionnaire about symptoms.

Results: The study included 2172 patients. Of these, 122 were diagnosed with CRC (5.6%). Median age was 61 years (range: 40-97) and 44.0% were men. All symptoms had high (93.4-96.8%) negative predictive values for CRC. The highest positive predictive values (PPV) were found for dark rectal bleeding (PPV: 20.6%) and CRC was diagnosed in a first-degree relative over the age of 50 years (PPV: 11.2%). At multiple logistic regression analysis dark rectal blood was the most important predictor of CRC (OR: 7.4). Other predictors were age 60 years or older (OR: 3.0), change in frequency of bowel movements (OR: 2.5), CRC diagnosed in a first-degree relative over the age of 50 years (OR: 2.6), male gender (OR: 2.2) and mono-symptomatic fresh rectal blood (OR: 1.7).

Conclusion: No self-reported symptoms presented by outpatients without CRC risk factors can be used accurately to identify patients with CRC with certainty. For clinical purposes self-reported symptoms seems thus not to be a useful accurate tool to distinguish between symptomatic outpatients with and without CRC.