Study on Effectiveness of Severity of Illness Scoring Systems in Children with Complex Congenital Heart Disease after Heart Surgery

Lijuan Fu, Tingitng Zhang, Haibo Zhang

Shanghai Childrens Medical Center, Shanghai, China

Date, time and location: 2018.05.25 15:30, Exhibition area, 1st Floor. Zone – B

Abstract

Objective:

To analyze the applicability of PRISM, PRISMⅢ, PIM, PIM2 and PCIS applied on children with complex congenital heart diseases. And construct a new score especially for these children and analyzed its applicability.

Methods:

The general information and clinical data were collected according to the requirements of these scores. Intraclass correlation coefficient (ICC) was used to measure the inter-rater reliability. Logistic regression, method of percentiles and expert opinions were used to construct the new score. T-test, Area under ROC curve, Hosmer-Lemeshow goodness of fit test, and missing value analysis were used to evaluate the performance and applicability of the five scores and the new score.

Results:

First section, 237 objects were included. ICC>0.8, which means the data collection was reliable. The calibration capability of the three scores were tolerable (P>0.05), but the discriminatory power were insufficient. The missing values were related with the different working pattern in China. Second section, 1284 objects were included. The new score included seven indexes. The range of the score was 0-57 and the cut-off point was 10. The calibration capability of the score was tolerable (P<0.05), and the discriminatory power was sufficient. The effectiveness of the new score was better than the five scores as verified with 146 objects.

Conclusion:

It’s necessary to optimize one more specific scoring system. The applicability of new score was well. It will improve the survival rates through assisting the medical workers in making decision, standardizing medical and nursing practice and promote clinical evidence-based practice.