Showing 2 results for Cox Proportional Hazards Model
Saman Maroufizadeh, Ebrahim Hajizadeh, Ahmad Reza Baghestani, Seyed Reza Fatemi,
Volume 15, Issue 2 (6-2012)
Abstract
Background: Gastric cancer is a common and lethal disease throughout the world. This study was designed and carried out to determine the five-year survival rate of gastric cancer patients who had undergone surgical treatment at Taleghani Hospital of Tehran, and to assess its associated factors. Materials and Methods: In this historical-cohort study, information obtained from a total of 213 gastric cancer patients who underwent surgery at Taleghani Hospital of Tehran between 2003 and 2008 was included. In the analyses, Kaplan-Meier method, log-rank test, Cox proportional hazards model, and Lin-Ying additive hazards model were used. Results: The five-year survival rate and the median life expectancy in the studied patients were 14.6% and 29.6 months, respectively. Two covariates showed significant impacts on the gastric cancer patients’ data in both models: age at diagnosis and tumor size. We found that pathologic stage was significant under the Lin-Ying model (P=0.043) however, it was not significant under the Cox model (P=0.069). Other clinicopathological characteristics were not statistically significant (P>0.05). Conclusion: Since Cox and Lin-Ying models consider different aspects of the association between risk factors and the study outcome, it seems desirable to use both of them as complementary models but not as alternative models to obtain a more comprehensive understanding of data. This study showed that the early detection of patients in younger ages and in primary stages is important to decrease the risk of death in patients with gastric cancer and to increase the survival rate.
Hoda Noorkojuri, Ebrahim Hajizadeh, Ahmad Reza Baghestani, Mohammad Amin Pourhoseingholi ,
Volume 15, Issue 10 (3-2013)
Abstract
Background: Gastric cancer is the second leading cause of mortality due to cancer worldwide and it is the most common type of cancer in Iran. The aim of this study was to assess the effects of prognostic factors on the survival of patients with gastric cancer using the penalized spline in Cox model and compare it with Cox proportional hazards model. Materials and Methods: In this retrospective study, information obtained from a total of 216 gastric cancer patients who underwent surgery at Taleghani Hospital of Tehran between 2003 and 2008 years was included. Cox proportional hazards model and penalized spline in Cox model were used. R software was used for data analysis. The efficacy of these models was compared according to Akaike information criterion. Results: The five-year survival rate was 30% and the mean follow-up time was 16.60±15.28 months. Survival mean and median were 46 and 30 months, respectively. The analysis of Cox proportional hazards and penalized spline models resulted in age at diagnosis and tumor size as prognostic factors for survival time (P<0.05). Also, Akaike information criterion and survival curve for patients with a tumor size over 35 mm and age at diagnosis over 45 years were equal in both models. Conclusion: Cox proportional hazards and penalized spline models generated similar results thus, it is more appropriate to use Cox proportional hazards model because of its ease of interpretation and capability of modeling both continuous and discrete covariates. This study also showed if gastric cancer is diagnosed early, the relative risk of death will reduce.
