Development and internal evaluation of a nomogram for identifying current sleep disturbance in patients with colorectal cancer undergoing chemotherapy: a cross-sectional study

Abstract Objectives This study aimed to identify factors associated with current sleep disturbance (SD) in patients with colorectal cancer (CRC) undergoing chemotherapy and to develop and internally evaluate a nomogram for estimating the probability of current SD. Methods A total of 630 patients with CRC undergoing chemotherapy were recruited from five hospitals in China. Participants were assessed using the Demographic Information Questionnaire, Pittsburgh Sleep Quality Index (PSQI), and Hospital Anxiety and Depression Scale (HADS), while clinical assessment data and laboratory test results at admission were obtained from electronic medical records. The data were randomly divided into a training set and a validation set at a 7:3 ratio. Univariable analyses and multivariable logistic regression were used to identify factors associated with SD in patients with CRC undergoing chemotherapy. The training set was used for multivariable logistic regression modeling and nomogram development, whereas the validation set was used to evaluate the discrimination and calibration of the fixed model. In addition, 500 bootstrap resamples were performed within the training set for internal validation. Results The overall prevalence of current SD was 68.7%. Gender, disease duration, ECOG performance status, HADS-D score, serum potassium, and albumin were independently associated with current SD and were included in the final nomogram. The model showed an AUC of 0.762 (95% CI, 0.716~0.808) in the training set and 0.712 (95% CI, 0.635~0.789) in the validation set. Bootstrap internal validation in the training set yielded a mean absolute calibration error of 0.022. In the validation set, the mean absolute calibration error (Eavg) was 0.055, with a calibration intercept of 0.318, a calibration slope of 0.667, and a Brier score of 0.188. Conclusions Patients with CRC undergoing chemotherapy had a high burden of current SD. The nomogram may assist in identifying patients with a higher probability of current SD who may warrant further sleep assessment.