The Annals of Applied Statistics
- Ann. Appl. Stat.
- Volume 13, Number 4 (2019), 2637-2661.
Hierarchical infinite factor models for improving the prediction of surgical complications for geriatric patients
Nearly a third of all surgeries performed in the United States occur for patients over the age of 65; these older adults experience a higher rate of postoperative morbidity and mortality. To improve the care for these patients, we aim to identify and characterize high risk geriatric patients to send to a specialized perioperative clinic while leveraging the overall surgical population to improve learning. To this end, we develop a hierarchical infinite latent factor model (HIFM) to appropriately account for the covariance structure across subpopulations in data. We propose a novel Hierarchical Dirichlet Process shrinkage prior on the loadings matrix that flexibly captures the underlying structure of our data while sharing information across subpopulations to improve inference and prediction. The stick-breaking construction of the prior assumes an infinite number of factors and allows for each subpopulation to utilize different subsets of the factor space and select the number of factors needed to best explain the variation. We develop the model into a latent factor regression method that excels at prediction and inference of regression coefficients. Simulations validate this strong performance compared to baseline methods. We apply this work to the problem of predicting surgical complications using electronic health record data for geriatric patients and all surgical patients at Duke University Health System (DUHS). The motivating application demonstrates the improved predictive performance when using HIFM in both area under the ROC curve and area under the PR Curve while providing interpretable coefficients that may lead to actionable interventions.
Ann. Appl. Stat., Volume 13, Number 4 (2019), 2637-2661.
Received: July 2018
Revised: May 2019
First available in Project Euclid: 28 November 2019
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Lorenzi, Elizabeth; Henao, Ricardo; Heller, Katherine. Hierarchical infinite factor models for improving the prediction of surgical complications for geriatric patients. Ann. Appl. Stat. 13 (2019), no. 4, 2637--2661. doi:10.1214/19-AOAS1292. https://projecteuclid.org/euclid.aoas/1574910058
- A. Proofs of HIFM properties. Properties of hierarchical infinite factor model prior on loadings matrix.
- B. Inference for full model. All steps needed to sample the model.
- C. Variable definitions shown in Figure 4. Description of variable names shown in Figure 4.