A Machine Learning Approach to Management of Heart Failure Populations

• Published in: JACC. Heart failure Vol. 8; no. 7; pp. 578 - 587
• Main Authors: Jing, Linyuan, Ulloa Cerna, Alvaro E., Good, Christopher W., Sauers, Nathan M., Schneider, Gargi, Hartzel, Dustin N., Leader, Joseph B., Kirchner, H. Lester, Hu, Yirui, Riviello, David M., Stough, Joshua V., Gazes, Seth, Haggerty, Allyson, Raghunath, Sushravya, Carry, Brendan J., Haggerty, Christopher M., Fornwalt, Brandon K.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.07.2020
• Subjects: Aged; Aged, 80 and over; data science; Disease Management; electronic health records; Female; Heart Failure - epidemiology; Heart Failure - therapy; Humans; Machine Learning; Male; Middle Aged; Morbidity - trends; population health; Population Surveillance - methods; Retrospective Studies; Risk Assessment - methods; ROC Curve; United States - epidemiology; United States; population health; ACE; CRT; ECG; electronic health records; HFpEF; HFrEF; BP; EHR; AUC; ARB; data science; EBBB; HF
• ISSN: 2213-1779; 2213-1787
• DOI: 10.1016/j.jchf.2020.01.012

Heart failure is a prevalent, costly disease for which new value-based payment models demand optimized population management strategies. This study sought to generate a strategy for managing populations of patients with heart failure by leveraging large clinical datasets and machine learning. Geisinger electronic health record data were used to train machine learning models to predict 1-year all-cause mortality in 26,971 patients with heart failure who underwent 276,819 clinical episodes. There were 26 clinical variables (demographics, laboratory test results, medications), 90 diagnostic codes, 41 electrocardiogram measurements and patterns, 44 echocardiographic measurements, and 8 evidence-based “care gaps”: flu vaccine, blood pressure of <130/80 mm Hg, A1c of <8%, cardiac resynchronization therapy, and active medications (active angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker/angiotensin receptor-neprilysin inhibitor, aldosterone receptor antagonist, hydralazine, and evidence-based beta-blocker) were collected. Care gaps represented actionable variables for which associations with all-cause mortality were modeled from retrospective data and then used to predict the benefit of prospective interventions in 13,238 currently living patients. Machine learning models achieved areas under the receiver-operating characteristic curve (AUCs) of 0.74 to 0.77 in a split-by-year training/test scheme, with the nonlinear XGBoost model (AUC: 0.77) outperforming linear logistic regression (AUC: 0.74). Out of 13,238 currently living patients, 2,844 were predicted to die within a year, and closing all care gaps was predicted to save 231 of these lives. Prioritizing patients for intervention by using the predicted reduction in 1-year mortality risk outperformed all other priority rankings (e.g., random selection or Seattle Heart Failure risk score). Machine learning can be used to priority-rank patients most likely to benefit from interventions to optimize evidence-based therapies. This approach may prove useful for optimizing heart failure population health management teams within value-based payment models. [Display omitted]