Colorectal cancer diagnostic pathways in Ontario

Abstract

Purpose: The purpose of this study was to differentiate diagnostic pathways in colorectal cancer in Ontario through the development of a pathway categorization scheme, and evaluate the patient-, disease-, system-related characteristics of patients and the length of the diagnostic interval across the diagnostic pathway categories. Methods: This was a cross-sectional study using an existing cohort at ICES. The study population included patients who were diagnosed with colorectal cancer in Ontario between 2009 and 2012. Cluster analysis used eleven variables that were related to patient presentation, patient visit pattern, patient symptom pattern and referral process to characterize and categorize diagnostic pathways. Chi-square test and One-way ANOVA were used to assess the association between the examined factors (age, sex, material deprivation quintile, comorbidities, and stage) and the diagnostic pathway categorization scheme. Unadjusted quantile regression was used to assess the association between the diagnostic interval length and the diagnostic pathway categorization scheme. Results: Six distinct diagnostic pathways were identified: asymptomatic pathway (N=4,494), colonoscopy pathway (N=10,066), the imaging and colonoscopy pathway (N=3,427), imaging alone pathway (N=2,238), the imaging and emergency presentation pathway (N=2,849) and no pre-diagnostic workup pathway (N=887). Patients who went through a pathway that was more adherent to diagnostic pathway guidelines (eg. asymptomatic pathway) were more likely to be younger, healthier and living in less deprived areas, and they tended to be diagnosed at an early stage with a short diagnostic interval. Patients who were female, older, living in more deprived areas and with more comorbid disease were more likely to go through pathways that were divergent from those guidelines. The length of the diagnostic interval was correlated to the number of colorectal cancer diagnosis-related visits occurring during the interval. All examined factors and the diagnostic interval were significantly associated with the pathway categorization (p<0.0001). Conclusions: This study demonstrated substantial variations in colorectal cancer diagnostic pathways in Ontario. Interventions should be designed to provide individualized and more effective diagnostic services to patients.