Objective To explore the relationship between long‑term preoperative sleep disturbance and preoperative delirium (POD) in elderly patients undergoing gastrointestinal tumor surgery. Methods Two hundred elderly patients of both sexes, aged 65 to 85 years, American Society of Anesthesiologists (ASA) grade Ⅱ or Ⅲ, undergoing gastrointestinal tumor surgery with general anesthesia, were enrolled in this study. The Pittsburgh Sleep Quality Index (PSQI) questionnaire was used to assess the patient's sleep quality in the past month within 24 h before the operation. The preoperative general conditions, intraoperative blood loss, fluid replacement, urine volume and postoperative rehabilitation of the two groups were recorded. POD was diagnosed daily for continueous seven days after operation by the Confusion Assessment Method. The patients were divided into two groups according to whether POD occurred: POD group (n=35) and Non‑POD (NPOD) group (n=150). The related risk factors of POD in patients were analyzed and multivariate Logistic regression analysis was used to evaluate the relationship between long‑term preoperative sleep disturbance and POD. Results Patients with high preoperative PSQI questionnaire score, alcoholism, cerebral infarction, and long‑term preoperative sleep disturbance were at increased risk of delirium (P<0.05). The first out‑of bed activity time and the postoperative length of hospital stay was significantly longer in POD group (P<0.05). There was no difference among other indicators between the two groups (P>0.05). Multivariate Logistic regression analysis showed that alcoholism and long‑term preoperative sleep disturbance were independent risk factors for postoperative delirium (P<0.05). Conclusions Long‑term preoperative sleep disturbance may induce postoperative delirium in elderly patients undergoing gastrointestinal tumor surgery. Correcting sleep disturbance before surgery is of certain significance for the early prevention of POD.