The investigators analyzed deaths according to type using a GCE model that controlled for age, sex, tumor site, and Charlson Comorbidity Index. “The GCE model essentially describes the degree to which cancer is the patient’s problem,” Dr. Katzel explained.
Results showed that both women and men were more likely to die from cancer than from other causes; however, the ratio was 7 for women, compared with just 3.8 for men, a difference translating to a relative hazard ratio of 1.92 (95% CI, 1.07-3.43).
In terms of potential confounding, there were only 19 noncancer deaths among the women studied, suggesting that they may have been more healthy than the men, which could have influenced the calculations, according to Dr. Katzel.
“This GCE model has been validated in head and neck cancer, but also in breast cancer, prostate cancer, and endometrial cancer, so we are using a validated model to do this evaluation,” he noted. “So I would say we are confident in our findings.”
Dr. Katzel disclosed that he had no relevant conflicts of interest. The study received funding from Kaiser Permanente Northern California Graduate Medical Education Department.