Using estimated probability of pre-diagnosis behavior as a predictor of cancer survival time: an example in esophageal cancer

Table 1 Association Between Survival Time and Probability of Pre-Diagnosis Health Behavior; All Esophageal Cancers

	1 year lag, unadjusted			1 year lag, adjusted^b			10 year lag, adjusted^b
Health behavior	HR^a	95% CI	P value	HR^a	95% CI	P value	HR^a	95% CI	P value
Current smoker	0.99	0.99,1.01	0.252	1.20	1.18,1.22	< 0.001	1.18	1.15,1.21	< 0.001
Alcohol - Heavy drinking	0.53	0.50,0.56	< 0.001	0.82	0.76,0.88	< 0.001	1.16	1.04,1.30	0.011
Alcohol - Binge drinking	0.76	0.74,0.77	< 0.001	0.96	0.93,0.99	0.012	1.04	0.99,1.08	0.093
Current smoking and alcohol consumption	0.87	0.83,0.92	< 0.001	1.93	1.79,2.07	< 0.001	1.69	1.56,1.84	< 0.001
Undertook exercise in past 30 days other than regular job	0.78	0.77,0.79	< 0.001	0.82	0.81,0.84	< 0.001	0.80	0.78,0.83	< 0.001
Obese	0.95	0.94,0.97	< 0.001	1.04	1.03,1.06	< 0.001	1.10	1.07,1.14	< 0.001

Abbreviations: CI Confidence interval, HR Hazard ratio
^aThe hazard ratio describes the impact of a 0.1 increase in the probability of having the specified health behavior
^bAdjusted for age and cancer stage at diagnosis

ISSN: 1471-2288