Improved 30 day heart failure rehospitalization prediction through the addition of device‐measured parameters

Abstract

Aims This study aimed to improve in‐person clinical evaluation on the day of heart failure (HF) hospitalization discharge by adding device‐measured parameters to predict 30 day HF rehospitalization risk in cardiac resynchronization therapy‐defibrillator (CRT‐D) patients. Methods and results In a cohort of Medicare patients with CRT‐Ds, the independent prognostic value of four device‐measured parameters was assessed relative to typical clinical parameters associated with rehospitalization risk. Medicare registry, claims, and Medtronic CareLink® Network data for these patients were analysed using logistic regression modelling and net reclassification methods. Among 1563 CRT‐D patients, 411 patients had 607 HF hospitalization events during a median 6.3 years of follow‐up. Compared with clinical variables alone, impedance measurements resulted in a 28% improvement between the predicted probabilities of having vs. not having a 30 day HF rehospitalization (relative integrated discrimination improvement = 0.28) and a net 42% improvement in the classification of 30 day HF rehospitalization events and non‐events after an index HF hospitalization (net reclassification index = 0.42; 95% CI: 0.10, 0.74). Conclusions In CRT patients, intrathoracic impedance measurements improve prediction of 30 day HF rehospitalization over clinical characteristics alone. The present study provides supportive data for the routine evaluation of intrathoracic impedance prior to discharge in patient with CRT devices. Furthermore, the models developed in this study could be used to design interventions to improve compliance with Medicare reimbursement guidelines regarding 30 day HF rehospitalization.

Current Trends in Readmission Prediction: An Overview of Approaches

“…Among all diagnosis-specific studies, half of them (50%) built models to predict readmission among patients with heart-specific conditions. Among the 18 studies focused on heart conditions, 13 papers predicted readmissions among the HF population [ 56 – 60 , 63 , 65 – 71 ], 2 worked on AMI cohorts [ 61 , 62 ], 2 developed models on general cardiovascular disease patients [ 55 , 72 ], and 1 worked on the stroke population [ 64 ]. As shown in Fig.…”

“…Even with the emergence of the ML algorithm, 29 out of 36 articles adopted traditional statistical methods. Among these studies, ~ 90% used LR either as a baseline [ 56 , 58 , 60 , 62 – 64 , 68 , 73 , 74 , 76 – 78 , 83 , 85 – 87 ] or the main model in prediction [ 60 , 69 , 71 , 82 , 88 – 90 ], and 3 studies derived their own risk scores on the basis of LR variable coefficients [ 61 , 66 , 84 ]. In the remaining 3 papers, the prognosis of readmission was carried out with Cox regression survival analysis.…”

An evidence-based data science perspective on the prediction of heart failure readmissions

“…2-4 These HFR studies have varied greatly in their readmission time frames (30d – 1yr) and methodologies (risk indices, statistical models, machine learning), as well as in the data underpinning their predictions (clinical, administrative, psychosocial). 4-8 However, the degrees of performance and applicability needed to make transformative progress in predicting HFR have been notoriously difficult to achieve 4,9 , begging the question of why.…”

“…Surprisingly, many HFR studies never mention or measure precision. 2-4,7,8,13-16,18,20,23-27 However, without precision there is little way to establish confidence in whether or not a prediction for readmission can be trusted. Among the comparatively few studies that report it, precision typically ranges from 0.09 to 0.44, meaning that 56 – 91% of predictions for readmission are often incorrect.…”

Prevalence and impact of malnutrition on readmission among hospitalized patients with heart failure in China

“…This puts medical and economic pressure not only on patients and families but also on society and the country. A study found that readmission after discharge in 25% of patients could be avoided by early risk identification and timely intervention 9 . Nevertheless, readmission has not been extensively studied in the Chinese population.…”