Objective:To evaluate whether individualized dialectical treatment decoction taken for≥6 months after radical surgery for stageⅡ-Ⅲ colorectal cancer(CRC)is superior to surgery alone at preventing early(within 2 years)recurrence and metastasis.Methods:Patients with CRC who underwent radical surgery were enrolled consecutively in The First Affiliated Hospital of Henan University of CM and The First Affiliated Hospital of Henan University.The primary outcome was defined as early recurrence and metastasis for patients who underwent≥6 months of traditional Chinese medicine(TCM)treatment versus those who did not after surgery(Non-TCM).Multivariable modified Pois-son regression was used to model the primary outcome,and propensity score matching(PSM)was included in the regression-based sensit-ivity analysis.Results:A total of 103 patients received TCM treatment and 384 did not.In the entire cohort,the multivariable modified Pois-son regression model showed that the adjusted risk for early recurrence and metastasis was lower in the TCM group10.7%(12/103)vs.21.4%(82/384),relative risk(RR):0.448,95%confidence interval[CI]:0.261-0.769).PSM analysis was used at a 1:2 ratio,resulting in 99 pa-tients in the TCM group and 198 patients in the Non-TCM group.The TCM group still showed a lower risk for early recurrence and metastas-is 11.1%(11/99)vs.21.2%(42/198),RR:0.464,95%CI:0.227-0.948.The results of sensitivity analysis were consistent with multivariable ana-lysis.In this prespecified subgroup analysis,the difference in the rate of the primary outcome between the TCM and the Non-TCM groups was greater among males and those with N2 and moderately-highly differentiated tumors.Kaplan-Meier survival analysis showed that the incidence of early-recurrence-and metastasis-free survival was significantly higher in the TCM group than the Non-TCM groups.Conclusions:For patients with stage Ⅱ-Ⅲ colorectal cancer after radical surgery,individualized dialectical treatment with TCM was associated with a lower risk of early recurrence and metastasis.