For a study, the researchers sought to use the hazard function to evaluate transitions in recurrence hazard and peak recurrence duration in patients with nonmetastatic CRC. Major guidelines recommend a 5-year postoperative observation term for patients with nonmetastatic CRC, but surveillance intervals vary. Instantaneous conditional recurrence rate estimates could aid in the selection of appropriate intervals. Between January 2000 and December 2013, 4,330 patients with stage I to III CRC who received curative resection at the National Cancer Center Hospital were included in the study. The hazard function was used to compute the recurrence rates. In patients with stage I, II, and III CRC, the recurrence rates were 4% (50/1432), 11% (136/1231), and 25% (424/1667), respectively. For 5 years after surgery, the hazard curve for stage I was reasonably flat, and the hazard rates were consistently modest (<0.0015). The stage II hazard curve featured a high hazard rate of 0.0046 at 13.7 months, followed by a long drop to the right. Stage III’s danger curve had an earlier and higher peak (0.0105 at 11.6 months) than stage II’s, with a long hem to the right. The risk of recurrence in CRC patients changed dramatically as they progressed through the stages. The outcomes imply that for the first 3 years after surgery, short-interval surveillance for stage I patients may be unnecessary; for stage III patients, short-interval administration should be explored.