Abstract
Background
It is established that comorbidities negatively influence colorectal cancer (CRC)-specific survival. Only few studies have used the relative survival (RS) setting to estimate this association, although RS has been proven particularly useful considering the inaccuracy in death certification. This study aimed to investigate the impact of non-cancer comorbidities at CRC diagnosis on net survival, using cancer registry data.
Methods
We included 1076 CRC patients diagnosed between 2000 and 2001 in the canton of Zurich. The number and severity of comorbidities was expressed using the Charlson Comorbidity Index (CCI). Multiple imputation was performed to account for missing information and 10-year net survival was estimated by modeling the excess hazards of death due to CRC, using flexible parametric models.
Results
After imputation, approximately 35 % of the patients were affected by comorbidities. These appeared to decrease the 10-year net survival; the estimated excess hazard ratio for patients with one mild comorbidity was 2.14 (95 % CI 1.60−2.86), and for patients with one more severe or more than one comorbidity was 2.43 (95 % CI 1.77−3.34), compared to patients without comorbidities.
Conclusions
Our analysis suggested that non-cancer comorbidities at CRC diagnosis significantly decrease the 10-year net survival. Future studies should estimate net survival of CRC including comorbidities as prognostic factor and use a RS framework to overcome the uncertainty in death certification.
Michalopoulou, Eleftheria