Table 3.

Multivariable hazard ratios (and 95% confidence intervals, CI) of joint association of pre- and post-diagnosis zinc supplement use in relation to all-cause mortality among men with nonmetastatic prostate cancer.

		Post-diagnosis use
		Non-user	1–25 mg/d	25–74 mg/d	>75 mg/d
*All-cause mortality*
Pre-diagnosis non-user	No. of events	2506	63	99	12
Pre-diagnosis non-user	HR (95% CI)^*	1.0	0.93 (0.71, 1.20)	0.89 (0.72, 1.09)	1.07 (0.60, 1.89)
Pre-diagnosis user	No. of events	195	88	191	44
Pre-diagnosis user	HR (95% CI)^*	1.24 (1.07, 1.44)	0.79 (0.63, 0.99)	1.05 (0.90, 1.24)	1.02 (0.75, 1.40)

Adjusted age at diagnosis, stage (T1, T2, T3a), Gleason score (<7, 7, ≥8, missing), treatment (radical prostatectomy, radiation, hormones, other), body mass index (<25, 25–30, ≥30 kg/m2), vigorous physical activity (<1, 1 to <3, ≥3 MET-h/week), smoking (never, former/quit > 10 years ago, former/quit ≤ 10 years ago, current), selenium supplement use (non-user, current), total number of supplements use (0, 2–3, ≥4), diabetes mellitus (yes or no), elevated cholesterol (yes or no), high blood pressure (yes or no), parental history of myocardial infarction before age 60 years, and comorbid condition (yes or no; conditions included myocardial infarction, coronary artery bypass or angioplasty, stroke, emphysema or chronic obstructive pulmonary disorder, and Parkinson disease).