Abstract: Objective To develop an early warning assessment scale for high-risk pregnant women, and to test the reliability and validity and explore its predictive value. Methods With literature review and Delphi expert consultation method, the draft of early warning assessment scale for high-risk pregnant women was formed. The reliability and validity of the scale were tested in 150 high-risk hospitalized pregnant women, and the predictive efficacy of the scale for the occurrence of serious/critical disease in high-risk pregnant women was analyzed. Results The early warning assessment scale for high-risk pregnant women consisted of 4 dimensions and 23 items. The intra-class correlation coefficient (ICC) was 0.905, and the 95% confidence interval (CI) ranged from 0.855 to 0.945. The average content validity ratio (CVR) was 0.96 and the sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and Jordan index were 92.5%, 68.2%, 51.4%, 97.4%, 18.00, 0.53 and 0.61 respectively, and AUROC of the assessment scale was 0.803. AUROC of predicting severe/critical disease was 0.719 and the sensitivity, specificity, positive likelihood ratio and negative likelihood ratio were 85.7%, 58.1%, 10.23, and 0.64 respectively. Conclusion The early warning assessment scale for high-risk pregnant women constructed in this study has good reliability and validity. It can predict the occurrence of acute obstetric events and critical obstetric diseases. The scale can be used as a tool to evaluate the condition of high-risk pregnant women and provide basis for early warning and evaluation of high-risk pregnant women.

Key words: high-risk pregnant women, early warning, assessment, reliability, validity