Figure 2.

Forest plots in the upper panel display the association between dietary flavonoid intake and cancer-related mortality using univariate (a) and multivariate (b) analyses adjusted for age, ethnicity, gender, PIR, educational status, marital status, daily energy intake, alcohol consumption, smoking status, cancer history, a total score of HEI, DII, and a total time of PA. (c) Forest plot showing the association between cancer-related mortality and total dietary flavonol intake using unadjusted Cox analysis stratified by age, gender, PIR, ethnicity, smoking status, alcohol usage, and disease history. Forest plot showing the association between cancer-related mortality and the dietary intake of peonidin (d), naringenin (e), and catechin (f) using unadjusted Cox analysis stratified by age, gender, smoking status, alcohol usage, and disease history. HR: hazard ratio; PIR: poverty income ratio; HEI: healthy eating index, 2015 version; DII: dietary inflammatory index; PA: physical activity.