A Public-Private Partnership Develops and Externally Validates a 30-Day Hospital Readmission Risk Prediction Model

Abstract

Introduction: Preventing the occurrence of hospital readmissions is needed to improve quality of care and foster population health across the care continuum. Hospitals are being held accountable for improving transitions of care to avert unnecessary readmissions. Advocate Health Care in Chicago and Cerner (ACC) collaborated to develop all-cause, 30-day hospital readmission risk prediction models to identify patients that need interventional resources. Ideally, prediction models should encompass several qualities: they should have high predictive ability; use reliable and clinically relevant data; use vigorous performance metrics to assess the models; be validated in populations where they are applied; and be scalable in heterogeneous populations. However, a systematic review of prediction models for hospital readmission risk determined that most performed poorly (average C-statistic of 0.66) and efforts to improve their performance are needed for widespread usage.Methods: The ACC team incorporated electronic health record data, utilized a mixed-method approach to evaluate risk factors, and externally validated their prediction models for generalizability. Inclusion and exclusion criteria were applied on the patient cohort and then split for derivation and internal validation. Stepwise logistic regression was performed to develop two predictive models: one for admission and one for discharge. The prediction models were assessed for discrimination ability, calibration, overall performance, and then externally validated.Results: The ACC Admission and Discharge Models demonstrated modest discrimination ability during derivation, internal and external validation post-recalibration (C-statistic of 0.76 and 0.78, respectively), and reasonable model fit during external validation for utility in heterogeneous populations.Conclusions: The ACC Admission and Discharge Models embody the design qualities of ideal prediction models. The ACC plans to continue its partnership to further improve and develop valuable clinical models.

A comparison of models for predicting early hospital readmissions

“…Some methods, such as in [4], leverage a variety of data sources, including patient demographic and social characteristics, medications, procedures, conditions, and lab tests. based on only a single source of data, for instance, solely on administrative claims data, as in [5].…”

Predicting 30-Day Readmissions in an Asian Population: Building a Predictive Model by Incorporating Markers of Hospitalization Severity

“…The more notable prediction scores include PARR-30 developed in the United Kingdom [6], HOSPITAL score [7] in the United States for potentially avoidable readmissions and the LACE index (Length of stay, Acuity of admission, Charlson comorbidity index, Emergency department visits in past 6 months) developed in Ontario, Canada [8]. With the exception of the LACE index, most of these models have limited generalizability to other health systems due to the unique socio-demographic variables [6,9] or have limited clinical utility due to the complexity of the model [5,9,10]. Moreover, the LACE index had moderate discriminative ability c -statistic 0.7 despite its simplicity while only four out of 25 other predictive models reviewed by Kansagara et al performed better [5].…”

A predictive framework for modeling healthcare data with evolving clinical interventions

“…We compare the prediction performance (AUC) of our models (HDP‐LR, and dHDP‐LR) with the following baselines: A clinical baseline using the HOSPITAL score , A clinical baseline using the LACE score , A clinical baseline using Stepwise Logistic Regression (Stepwise LR) , An intervention‐driven predictive model DP‐LR , Bayesian Logistic Regression (B‐LR), A dynamic hierarchical Bayesian model, Evo‐HDP , SVM with linear kernel (Linear‐SVM), SVM with 3rd order polynomial kernel (Poly3‐SVM), Naive Bayes, and Random Forest . …”