Aims and Objective: To investigate the incidence, characteristics and risk factors for hypoactive delirium in patients with non-traumatic ARDS.Background Hypoactive delirium is a known serious complication in ARDS patients and currently there are no effective early detection models and clinical prediction tools, lack of clinical treatment.Methods The study was conducted with non-traumatic ARDS patients who stayed in the ICU for more than 24 hours and were older than 18 years. 205 ARDS patients admitted to ICU in Gansu Provincial People's Hospital from December 2021 to February 2023 were selected. Demographic, clinical characteristics and laboratory test indicators were collected within 24 h after the patient entered the ICU. A multivariate logistic regression analysis was used to investigate risk factors, at the same times evaluate the clinical prediction effect of the model and draw a Nomogram for visual display.Results The incidence of hypoactive delirium among patients included in the study was 41%. Patients with hypoactive delirium were hypertension, diabetes mellitus, APACHE II score ≥ 15, higher procalcitonin, higher CRP, higher lactic dehydrogenase and higher IL-6 than those without hypoactive delirium. The logistic regression analysis results revealed that diabetes mellitus (OR 3.305, 95% CI: 1.866–12.616; p = 0.047), CRP (OR 1.002, 95% CI: 1.001–1.023; p = 0.044), IL-6 (OR 1.045, 95% CI: 1.017–1.063; p = 0.001) were independent risk factors of hypoactive delirium. After ROC, Calibration plot and Decision Curve Analysis (DCA) confirmed that the clinical prediction of this study model was satisfactory, and a nomogram were drawn for visual display.Conclusions Hypoactive delirium is a common serious complication in non-traumatic ARDS patients. Our logistic regression model not only effectively predicts hypoactive delirium early, but also excavates potential clinical therapeutic targets.