Objective：To explore the risk factors of acute respiratory distress syndrome（ARDS） occurrence in patients with coronavirus disease 2019（COVID-19） in Chongqing，and establish a risk-prediction model for the early diagnosis of ARDS. Methods：Clinical data of 225 laboratory-confirmed cases with COVID-19 admitted to 4 designated hospitals in Chongqing from January 16 to March 30，2020 were analyzed retrospectively. According to the Berlin definition，the patients were divided into ARDS group and non-ARDS group. The differences of initial baseline features，clinical manifestations，laboratory findings，imaging features and prognosis between the two groups were compared. The values were assigned accord－ing to the odds ratios（OR） of each factor to establish the risk-prediction model. The receiver operating characteristic（ROC） curve was used to evaluate the predictive effect of the model. Results：Among 225 patients，there were 141 males and 84 fe－males，and the median age was 56 years old. ARDS occurred in 62 cases（27.6%）. In the logistic analysis，4 variables（age≥60 years old，lymphocyte≤1.0×109/L，D-Dimer≥0.5 mg/L，com－plicated with diabetes） were determined as independent risk factors associated with ARDS in COVID-19 patients，and the odds ratios were 5.849（95%CI=2.716-12.593，P=0.000），4.318（95%CI=2.001-9.316，P=0.000），3.049（95%CI=1.300-7.152，P=0.010） and 2.491（95%CI=1.102-5.632，P=0.028） respectively. The area under the ROC curve of the risk-prediction model was 0.848（95%CI=0.794-0.892，P=0.000）. Conclusion：Age≥60 years old，D-dimer≥0.5 mg/L，lymphocyte≤1.0×109/L and complicated with diabetes are independent risk factors to predict ARDS in COVID-19 patients. The risk-prediction model based on the identified inde－pendent risk factors may have a good early warning value for COVID-19 patients complicated with ARDS.