Figure 1.

Multivariable-adjusted hazard ratio (95% CI) for all-cause, cancer, and cardiovascular disease mortality associated with a 2-fold increase in creatinine-corrected urinary cadmium overall, and by important subgroups among men. Adjustment included age (restricted quadratic spline with knots at the 10th, 50th, and 90th percentile of the overall study sample), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, other), urban residence (urban, rural), annual household income (< $20,000, ≥ $20,000), high school education, smoking category (never, former who quit ≥ 4 years ago, former who quit < 4 years ago, current), tertile of pack-years (< 9.0, 9.0–28.2, ≥ 28.2), physical activity (none, 1–2, ≥ 3 times a week), diabetes, BMI (continuous), alcohol consumption (< 12, ≥ 12 drinks in the past year), CRP (not detectable, 0.3–0.9, ≥ 1.0 mg/dL), total cholesterol (continuous), systolic blood pressure (continuous), blood pressure–lowering medication, blood lead (log-transformed), and eGFR (< 60, 60–89, ≥ 90 mL/min, 1.73 m²). A 2-fold increase in urinary cadmium corresponds approximately to the difference between the 75th and 50th percentiles of the cadmium distribution (0.61 μg/g and 0.32 μg/g, respectively). The sizes of the boxes are inversely related to the variance of the point estimate.