OBJECTIVES: We aimed to investigate factors associated with colorectal cancer survival in Golestan, Iran. METHODS: We used a population based cancer registry to recruit study subjects. All patients registered since 2004 were contacted and data were collected using structured questionnaires and trained interviewers. All the existing evidences to determine the stage of the cancer were also collected. The time from first diagnosis to death was compared in patients according to their stage of cancer using the Kaplan-Meir method. A Cox proportional hazard model was built to examine their survival experience by taking into account other covariates. RESULTS: Out of a total of 345 subjects, 227 were traced. Median age of the subjects was 54 and more than 42% were under 50 years old. We found 132 deaths among these patients, 5 of which were non-colorectal related deaths. The median survival time for the entire cohort was 3.56 years. A borderline significant difference in survival experience was detected for ethnicity (log rank test, p=0.053). Using Cox proportional hazard modeling, only cancer stage remained significantly associated with time of death in the final model. CONCLUSIONS: Colorectal cancer occurs at a younger age among people living in Golestan province. A very young age at presentation and what appears to be a high proportion of patients presenting with late stage in this area suggest this population might benefit substantially from early diagnoses by introducing age adapted screening programs.
Cohort Studies ; Colorectal Neoplasms ; Humans ; Mass Screening ; Proportional Hazards Models ; Surveys and Questionnaires