Survival Analysis Modeling of Colorectal Cancer Patients with Competing Risks in Golestan Province

Marzieh  Gharanjiki; Abdolhalim  Rajabi; Taghi  Amiriani; Gholamreza  Roshandel; Mohammadali  Vakili

doi:10.18502/ijre.v21i2.19913

Marzieh Gharanjiki Master's student in Biostatistics, Department of Biostatistics and Epidemiology, Faculty of Health, Golestan University of Medical Sciences, Gorgan, Iran
Abdolhalim Rajabi Health Management and Social Development Research Center, Golestan University of Medical Sciences, Gorgan, Iran
Taghi Amiriani Gastroenterology and Hepatology Research Center, Golestan University of Medical Sciences, Gorgan, Iran
Gholamreza Roshandel Gastroenterology and Hepatology Research Center, Golestan University of Medical Sciences, Gorgan, Iran
Mohammadali Vakili Health Management and Social Development Research Center, Golestan University of Medical Sciences, Gorgan, Iran

DOI: https://doi.org/10.18502/ijre.v21i2.19913

Keywords: Colorectal cancer, Survival analysis, Competing risks, Cumulative incidence function

Abstract

Background and Objectives: Colorectal cancer is the most common gastrointestinal malignancy worldwide, and its incidence is increasing in Iran. Competing risk analysis offers a refined approach to identify factors influencing Colorectal cancer-specific mortality. Therefore, This study was designed and conducted to determine the survival of patients with colorectal cancer and its associated factors.

Methods: The study was a historical cohort. Data of patients diagnosed with colorectal cancer between 2013 and 2019 at Golestan University of Medical Sciences were collected, and patients were followed up until May 4, 2024. Patient survival was estimated, and the cumulative incidence function, as well as competing risk models of cause-specific hazards and subdistribution hazards, were applied for competing risk analysis. Model adequacy was assessed using the Akaike Information Criterion. Analyses were performed in STATA version 17 (α = 0.05).

Results: Of the 811 patients, 366 (45.13%) were women and the rest were men. The mean age and standard deviation were calculated as 58.54 ± 13.87 years. In the multivariable analysis of factors associated with colorectal cancer mortality in the presence of competing risks, variables including gender, age, literacy, ethnicity, and metastasis were statistically significant in both models. Findings indicated that the cause-specific hazard model provided a better fit for analyzing survival outcomes in colorectal cancer.

Conclusion: Competing risk analysis based on cause-specific hazards is recommended in the multivariable analysis of factors associated with colorectal cancer mortality, particularly in the presence of competing risks of death from other causes.