Dataset Information

Data-driven decisions for reducing readmissions for heart failure: general methodology and case study.

ABSTRACT:

Background

Several studies have focused on stratifying patients according to their level of readmission risk, fueled in part by incentive programs in the U.S. that link readmission rates to the annual payment update by Medicare. Patient-specific predictions about readmission have not seen widespread use because of their limited accuracy and questions about the efficacy of using measures of risk to guide clinical decisions. We construct a predictive model for readmissions for congestive heart failure (CHF) and study how its predictions can be used to perform patient-specific interventions. We assess the cost-effectiveness of a methodology that combines prediction and decision making to allocate interventions. The results highlight the importance of combining predictions with decision analysis.

Methods

We construct a statistical classifier from a retrospective database of 793 hospital visits for heart failure that predicts the likelihood that patients will be rehospitalized within 30 days of discharge. We introduce a decision analysis that uses the predictions to guide decisions about post-discharge interventions. We perform a cost-effectiveness analysis of 379 additional hospital visits that were not included in either the formulation of the classifiers or the decision analysis. We report the performance of the methodology and show the overall expected value of employing a real-time decision system.

Findings

For the cohort studied, readmissions are associated with a mean cost of $13,679 with a standard error of $1,214. Given a post-discharge plan that costs $1,300 and that reduces 30-day rehospitalizations by 35%, use of the proposed methods would provide an 18.2% reduction in rehospitalizations and save 3.8% of costs.

Conclusions

Classifiers learned automatically from patient data can be joined with decision analysis to guide the allocation of post-discharge support to CHF patients. Such analyses are especially valuable in the common situation where it is not economically feasible to provide programs to all patients.

SUBMITTER: Bayati M

PROVIDER: S-EPMC4190088 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Data-driven decisions for reducing readmissions for heart failure: general methodology and case study.

Bayati Mohsen M Braverman Mark M Gillam Michael M Mack Karen M KM Ruiz George G Smith Mark S MS Horvitz Eric E

PloS one 20141008 10

<h4>Background</h4>Several studies have focused on stratifying patients according to their level of readmission risk, fueled in part by incentive programs in the U.S. that link readmission rates to the annual payment update by Medicare. Patient-specific predictions about readmission have not seen widespread use because of their limited accuracy and questions about the efficacy of using measures of risk to guide clinical decisions. We construct a predictive model for readmissions for congestive h ...[more]

PMID: 25295524

Similar Datasets

Project description:BackgroundReduction of 30-day all-cause readmissions for heart failure (HF) has become an important quality-of-care metric for health care systems. Many hospitals have implemented quality improvement programs designed to reduce 30-day all-cause readmissions for HF. Electronic medical record (EMR)-based measures have been employed to aid in these efforts, but their use has been largely adjunctive to, rather than integrated with, the overall effort.ObjectivesWe hypothesized that a comprehensive EMR-based approach utilizing an HF dashboard in addition to an established HF readmission reduction program would further reduce 30-day all-cause index hospital readmission rates for HF.MethodsAfter establishing a quality improvement program to reduce 30-day HF readmission rates, we instituted EMR-based measures designed to improve cohort identification, intervention tracking, and readmission analysis, the latter 2 supported by an electronic HF dashboard. Our primary outcome measure was the 30-day index hospital readmission rate for HF, with secondary measures including the accuracy of identification of patients with HF and the percentage of patients receiving interventions designed to reduce all-cause readmissions for HF.ResultsThe HF dashboard facilitated improved penetration of our interventions and reduced readmission rates by allowing the clinical team to easily identify cohorts with high readmission rates and/or low intervention rates. We significantly reduced 30-day index hospital all-cause HF readmission rates from 18.2% at baseline to 14% after implementation of our quality improvement program ( P = .045). Implementation of our EMR-based approach further significantly reduced 30-day index hospital readmission rates for HF to 10.1% ( P for trend = .0001). Daily time to screen patients decreased from 1 hour to 15 minutes, accuracy of cohort identification improved from 83% to 94.6% ( P = .0001), and the percentage of patients receiving our interventions, such as patient education, also improved significantly from 22% to 100% over time ( P < .0001).ConclusionsIn an institution with a quality improvement program already in place to reduce 30-day readmission rates for HF, an EMR-based approach further significantly reduced 30-day index hospital readmission rates.

Project description:IntroductionNew financial incentives, such as reduced Medicare reimbursements, have led hospitals to closely monitor their readmission rates and initiate efforts aimed at reducing them. In this context, many surgical departments participate in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP), which collects detailed demographic, laboratory, clinical, procedure and perioperative occurrence data. The availability of such data enables the development of data science methods which predict readmissions and, as done in this paper, offer specific recommendations aimed at preventing readmissions.Materials and methodsThis study leverages NSQIP data for 722,101 surgeries to develop predictive and prescriptive models, predicting readmissions and offering real-time, personalized treatment recommendations for surgical patients during their hospital stay, aimed at reducing the risk of a 30-day readmission. We applied a variety of classification methods to predict 30-day readmissions and developed two prescriptive methods to recommend pre-operative blood transfusions to increase the patient's hematocrit with the objective of preventing readmissions. The effect of these interventions was evaluated using several predictive models.ResultsPredictions of 30-day readmissions based on the entire collection of NSQIP variables achieve an out-of-sample accuracy of 87% (Area Under the Curve-AUC). Predictions based only on pre-operative variables have an accuracy of 74% AUC, out-of-sample. Personalized interventions, in the form of pre-operative blood transfusions identified by the prescriptive methods, reduce readmissions by 12%, on average, for patients considered as candidates for pre-operative transfusion (pre-operative hematoctic <30). The prediction accuracy of the proposed models exceeds results in the literature.ConclusionsThis study is among the first to develop a methodology for making specific, data-driven, personalized treatment recommendations to reduce the 30-day readmission rate. The reported predicted reduction in readmissions can lead to more than $20 million in savings in the U.S. annually.

Project description:AimsHeart failure (HF) readmission commonly arises owing to insufficient patient knowledge and failure of recognition of the early stages of recurrent fluid congestion. In previous work, we developed a score to predict short-term hospital readmission and showed that higher-risk patients benefit most from a disease management programme (DMP) that included enhancing knowledge and education by a nurse. We aim to evaluate the effectiveness of a novel, nurse-led HF DMP in selected patients at high risk of short-term hospital readmission, using ultrasound-guided diuretic management and artificial intelligence to enhance HF knowledge in an outpatient setting.Methods and resultsRisk-HF is a prospective multisite randomized controlled trial that will allocate 404 patients hospitalized with acute decompensated HF, and ≥33% risk of readmission and/or death at 30 days, into risk-guided nurse intervention (DMP-Plus group) compared with usual care. Intervention elements include (i) fluid management with a handheld ultrasound (HHU) device at point of care; (ii) post-discharge follow-up; (iii) optimal programmed drug titration; (iv) better transition of care; (v) intensive self-care education via an avatar-based 'digital health coach'; and (vi) exercise guidance through the digital coach. Usual care involves standard post-discharge hospital care. The primary outcome is reduced death and/or hospital readmissions at 30 days post-discharge, and secondary outcomes include quality of life, fluid management efficacy, and feasibility and patient engagement. Assuming that our intervention will reduce readmissions and/or deaths by 50%, with a 1:1 ratio of intervention vs. usual care, we plan to randomize 404 patients to show a difference at a statistical power of 80%, using a two-sided alpha of 0.05. We anticipate this recruitment will be achieved by screening 2020 hospitalized HF patients for eligibility. An 8 week pilot programme of our digital health coach in 21 HF patients, age > 75 years, showed overall improvements in quality of life (13 of 21), self-care (12 of 21), and HF knowledge (13 of 21). A pilot of the use of HHU by nurses showed that it was feasible and accurate.ConclusionsThe Risk-HF trial will evaluate the effectiveness of a risk-guided intervention to improve HF outcomes and will evaluate the efficacy of trained HF nurses delivering a fluid management protocol that is guided by lung ultrasound with an HHU at point of care.

Dataset Information

Data-driven decisions for reducing readmissions for heart failure: general methodology and case study.

Background

Methods

Findings

Conclusions

Publications

Data-driven decisions for reducing readmissions for heart failure: general methodology and case study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets