DURATION OF COLORECTAL-CANCER SYMPTOMS AND SURVIVAL - THE EFFECT OF CONFOUNDING CLINICAL AND PATHOLOGICAL VARIABLES

The relationship between symptom duration and long-term survival follo wing colorectal cancer is complex, and a number of factors may influen ce the length of time from onset of symptoms to cancer diagnosis. We p rospectively studied 777 consecutive colorectal cancer patients to det ermine the association between symptom duration and survival independe nt of other clinical and pathological features. We used survival curve s, the logrank test and Cox's proportional hazards model to assess pos sible changes in relative risk of death with increasing symptom durati on, without making any a priori assumptions. We found that symptom dur ation shortened with advanced tumour stage (P < 0.0006) and was also s horter for patients presenting with bowel obstruction (P < 0.0001). Un ivariate survival analysis showed that long-term survival increased co nsistently with symptom duration (P < 0.001). However, when the effect of tumour stage and bowel obstruction were accounted for in a multiva riate analysis, no decrease in the relative risk of death was seen as symptom duration increased. The addition of other variables to the pro portional hazards model such as age, sex or tumour site did not furthe r influence the risk function form of symptom duration. Our results su ggest that early diagnosis of colorectal cancer should remain our goal when assessing patients with suggestive gastrointestinal symptoms. (C ) 1997 Elsevier Science Ltd.