Detecting Important Risk Factors of Survival Time of Lung Cancer Patients Using Censored Quintile Regression

Abstract

Objective: Lung cancer is the most frequently diagnosed cancer and the most leading cause of death due to cancer<br />worldwide. This study aimed to use censored quintile regression model to estimate the effect of potential risk factors<br />on survival of lung cancer patients. Methods: In this study we used a dataset of a retrospective cohort study conducted<br />in West Azerbaijan (during 2007 and 2014). Demographic variables included age, and gender and biological variables<br />included Eastern Cooperative Oncology Group, smoking, tumor type, stage, metastasis, and treatment were investigated<br />as risk factors of survival of lung cancer patients by using censored quintile regression. Result: The mean (± standard<br />deviation) age of the 347 lung cancer patients was 63.48 (± 13.55) years. The survival time ranges from 11 to 91. A<br />total of 240 (67.6%) experienced death by the end of the study. The impact of Eastern Cooperative Oncology Group<br />(IV), smoking and treatment were significant for most of quintiles (p<0.05). Stage of cancer and metastasis are the other<br />affective risk factors on the survival of lung cancer patients(p<0.05). It was shown that Eastern Cooperative Oncology,<br />smoking habit and treatment were significantly associated with a shorter time-to-death progression. Conclusion: As<br />censored quintile regression could consider time-varying effects and the interpretation of its regression coefficients are<br />easy it could be an adequate choice for analyzing survival data.