Application of critical condition scoring scale of traditional Chinese medicine in predicting operative risks for elder patients with orthopedic surgery

Objective: To evaluate the capability of Critical Condition Scoring Scale of Traditional Chinese Medicine (CSSTCM) in predicting complication morbidity and mortality among elder patients who received orthopedic surgery. Methods: Evaluate patients in second orthopedic department of Guangdong Provincial Hospital of Chinese Medicine from September 2012 to March 2013 by CSSTCM and POSSUM. All data were analyzed in SPSS 19.0. Results: Logistic regression analysis yielded a statistically significant equation for complication morbidity: In R/(1 - R) = - 7.821+0.495×FS+0.335×CS+0.625×QS (FS: four examinations and eight principles score; CS: constitution score; QS: qi & blood syndrome differentiation score). The cutting point of the model was 48.32%, sensitivity 78.13%, specificity 91.30%, misclassification rate 8.70%, omission classification rate 21.87%, Youden index 0.69 and coincidence rate 85.1% The data did not produce an effective Logistic regression equation for mortality. CSSTCM model had better C-Index (0.902 to 0.757), sensitivity (0.781 to 0.656), specificity (0.913 to 0.812) and coincidence rate (0.851 to 0.762) than POSSUM. Conclusion: CSSTCM can generate a significant and effective regression model for complication morbidity, but fails to establish a significant equation for mortality due to limited sample volume. CSSTCM model in predicting complication morbidity has better fitness for the observed sample than POSSUM does.