A Nomogram Prediction of Length of Hospital Stay in Patients with COVID-19 Pneumonia: A Retrospective Cohort Study

• Main Authors: Kang Li, Chi Zhang, Ling Qin, Chaoran Zang, Ang Li, Jianping Sun, Yan Zhao, Yingmei Feng, Yonghong Zhang
• Format: Article
• Language: English
• Published: Hindawi Limited 2021-01-01
• Series: Disease Markers
• Online Access: http://dx.doi.org/10.1155/2021/5598824

Assessing the length of hospital stay (LOS) in patients with coronavirus disease 2019 (COVID-19) pneumonia is helpful in optimizing the use efficiency of hospital beds and medical resources and relieving medical resource shortages. This retrospective cohort study of 97 patients was conducted at Beijing You’An Hospital between January 21, 2020, and March 21, 2020. A multivariate Cox proportional hazards regression based on the smallest Akaike information criterion value was used to select demographic and clinical variables to construct a nomogram. Discrimination, area under the receiver operating characteristic curve (AUC), calibration, and Kaplan–Meier curves with the log-rank test were used to assess the nomogram model. The median LOS was 13 days (interquartile range [IQR]: 10–18). Age, alanine aminotransferase, pneumonia, platelet count, and PF ratio (PaO2/FiO2) were included in the final model. The C-index of the nomogram was 0.76 (95%confidence interval CI=0.69–0.83), and the AUC was 0.88 (95%CI=0.82–0.95). The adjusted C-index was 0.75 (95%CI=0.67–0.82) and adjusted AUC 0.86 (95%CI=0.73–0.95), both after 1000 bootstrap cross internal validations. A Brier score of 0.11 (95%CI=0.07–0.15) and adjusted Brier score of 0.130 (95%CI=0.07–0.20) for the calibration curve showed good agreement. The AUC values for the nomogram at LOS of 10, 20, and 30 days were 0.79 (95%CI=0.69–0.89), 0.89 (95%CI=0.83–0.96), and 0.96 (95%CI=0.92–1.00), respectively, and the high fit score of the nomogram model indicated a high probability of hospital stay. These results confirmed that the nomogram model accurately predicted the LOS of patients with COVID-19. We developed and validated a nomogram that incorporated five independent predictors of LOS. If validated in a future large cohort study, the model may help to optimize discharge strategies and, thus, shorten LOS in patients with COVID-19.