Development and internal validation of the multivariable CIPHER (Collaborative Integrated Pregnancy High-dependency Estimate of Risk) clinical risk prediction model.

BACKGROUND: Intensive care unit (ICU) outcome prediction models, such as Acute Physiology And Chronic Health Evaluation (APACHE), were designed in general critical care populations and their use in obstetric populations is contentious. The aim of the CIPHER (Collaborative Integrated Pregnancy High-dependency Estimate of Risk) study was to develop and internally validate a multivariable prognostic model calibrated specifically for pregnant or recently delivered women admitted for critical care. METHODS: A retrospective observational cohort was created for this study from 13 tertiary facilities across five high-income and six low- or middle-income countries. Women admitted to an ICU for more than 24 h during pregnancy or less than 6 weeks post-partum from 2000 to 2012 were included in the cohort. A composite primary outcome was defined as maternal death or need for organ support for more than 7 days or acute life-saving intervention. Model development involved selection of candidate predictor variables based on prior evidence of effect, availability across study sites, and use of LASSO (Least Absolute Shrinkage and Selection Operator) model building after multiple imputation using chained equations to address missing data for variable selection. The final model was estimated using multivariable logistic regression. Internal validation was completed using bootstrapping to correct for optimism in model performance measures of discrimination and calibration. RESULTS: Overall, 127 out of 769 (16.5%) women experienced an adverse outcome. Predictors included in the final CIPHER model were maternal age, surgery in the preceding 24 h, systolic blood pressure, Glasgow Coma Scale score, serum sodium, serum potassium, activated partial thromboplastin time, arterial blood gas (ABG) pH, serum creatinine, and serum bilirubin. After internal validation, the model maintained excellent discrimination (area under the curve of the receiver operating characteristic (AUROC) 0.82, 95% confidence interval (CI) 0.81 to 0.84) and good calibration (slope of 0.92, 95% CI 0.91 to 0.92 and intercept of -0.11, 95% CI -0.13 to -0.08). CONCLUSIONS: The CIPHER model has the potential to be a pragmatic risk prediction tool. CIPHER can identify critically ill pregnant women at highest risk for adverse outcomes, inform counseling of patients about risk, and facilitate bench-marking of outcomes between centers by adjusting for baseline risk.

Machine learning-enabled maternal risk assessment for women with pre-eclampsia (the PIERS-ML model): a modelling study.

BACKGROUND: Affecting 2-4% of pregnancies, pre-eclampsia is a leading cause of maternal death and morbidity worldwide. Using routinely available data, we aimed to develop and validate a novel machine learning-based and clinical setting-responsive time-of-disease model to rule out and rule in adverse maternal outcomes in women presenting with pre-eclampsia. METHODS: We used health system, demographic, and clinical data from the day of first assessment with pre-eclampsia to predict a Delphi-derived composite outcome of maternal mortality or severe morbidity within 2 days. Machine learning methods, multiple imputation, and ten-fold cross-validation were used to fit models on a development dataset (75% of combined published data of 8843 patients from 11 low-income, middle-income, and high-income countries). Validation was undertaken on the unseen 25%, and an additional external validation was performed in 2901 inpatient women admitted with pre-eclampsia to two hospitals in south-east England. Predictive risk accuracy was determined by area-under-the-receiver-operator characteristic (AUROC), and risk categories were data-driven and defined by negative (-LR) and positive (+LR) likelihood ratios. FINDINGS: Of 8843 participants, 590 (6·7%) developed the composite adverse maternal outcome within 2 days, 813 (9·2%) within 7 days, and 1083 (12·2%) at any time. An 18-variable random forest-based prediction model, PIERS-ML, was accurate (AUROC 0·80 [95% CI 0·76-0·84] vs the currently used logistic regression model, fullPIERS: AUROC 0·68 [0·63-0·74]) and categorised women into very low risk (-LR <0·1; eight [0·7%] of 1103 women), low risk (-LR 0·1 to 0·2; 321 [29·1%] women), moderate risk (-LR >0·2 and +LR <5·0; 676 [61·3%] women), high risk (+LR 5·0 to 10·0, 87 [7·9%] women), and very high risk (+LR >10·0; 11 [1·0%] women). Adverse maternal event rates were 0% for very low risk, 2% for low risk, 5% for moderate risk, 26% for high risk, and 91% for very high risk within 48 h. The 2901 women in the external validation dataset were accurately classified as being at very low risk (0% with outcomes), low risk (1%), moderate risk (4%), high risk (33%), or very high risk (67%). INTERPRETATION: The PIERS-ML model improves identification of women with pre-eclampsia who are at lowest and greatest risk of severe adverse maternal outcomes within 2 days of assessment, and can support provision of accurate guidance to women, their families, and their maternity care providers. FUNDING: University of Strathclyde Diversity in Data Linkage Centre for Doctoral Training, the Fetal Medicine Foundation, The Canadian Institutes of Health Research, and the Bill & Melinda Gates Foundation.