FIGURE 2.

Association between tea consumption and risk of all-cause mortality among diabetic participants. Values were obtained from a Cox proportional hazards analysis. Multivariable analyses were adjusted for age (years); education (no formal school, primary school, middle school, high school, college, or university or higher); random glucose (millimoles/liter); treatment for diabetes (taking insulin and/or oral hypoglycemic drugs, or no treatment); smoking (never smoker; former smoker who has quit for reasons other than illness; current smoker or former smoker who has quit because of illness: 1–9, 10–19, 20–29, or ≥30 cigarettes or equivalent/d); alcohol intake (nonweekly drinker; former weekly drinker; weekly drinker; daily drinker: <15, 15–29, 30–59, or ≥60 g/d of pure alcohol); level of physical activity (MET-hours/day); intakes of red meat, fresh vegetables, and fruits (days/week; calculated by assigning participants to the midpoint of their consumption category); BMI (kg/m²); waist circumference (centimeters); and baseline prevalence of hypertension, cancer, stroke, and coronary heart disease (yes or no). Solid squares represent the HRs and horizontal lines represent the corresponding 95% CIs. Unadjusted incidence rates are reported per 1000 PYs of follow-up. MET, metabolic equivalent task; PY, person-year; Ref, reference.