Predictors of hospital readmission after discharge in patients with congestive heart failure

doi:10.1016/j.hrtlng.2005.01.001

Heart & Lung

Volume 34, Issue 4, July–August 2005, Pages 231-239

https://doi.org/10.1016/j.hrtlng.2005.01.001 Get rights and content

Purpose

The purposes of this study were to (1) describe the characteristics of the population with congestive heart failure (CHF) who were admitted to a large, southeastern, acute-care hospital and (2) determine which patients are at risk for readmissions within 6 months.

Methods

A descriptive correlational design, using variables maintained in a computerized data bank on patients with CHF (N = 557, 39% were black) who were admitted between October 2000 and March 2002, was used to describe the adult population with CHF and identify variables associated with a likelihood of readmission.

Results

In the 6 months after the index admission, 224 (40%) of the patients were readmitted to the hospital for CHF. Variables significantly associated with readmission included lack of cardiology consult during admission, living status, point of entry of index admission, receiving Medicare, and having pulmonary hypertension. Four models, composed of subsets of variable from the data bank were developed and tested with logistic regression. The model composed of discharge variables was the only model that predicted readmission at a significant level.

Conclusions

There is a need to develop comprehensive data banks to describe patterns of care and their outcomes. Such data should inform plans to manage this vulnerable population.

Section snippets

Review of research literature

Various factors that relate to or predict readmissions in patients with CHF are described in research literature. The review of research for this study focused on the variables maintained in a computerized data bank on all patients with CHF in our institution. These variables can be grouped into 7 broad categories: demographics, comorbidities, clinical parameters, medication regimen, discharge factors, medical management information, and psychosocial status.

Design

A retrospective, descriptive correlational design, with comparative procedures for subgroups, was used to examine relationships among variables of patients with CHF who were admitted to a southeastern regional referral center during the months from October 2000 to March 2002. Subjects were divided into 2 groups: (1) those readmitted for CHF within 6 months of discharge and (2) those who were not readmitted within 6 months of discharge.

Sample and setting

Subjects were all patients with CHF (N = 557) admitted to a

Discussion

The primary finding of this study was that 40% of patients admitted with CHF during the study period were readmitted to the hospital within 6 months of discharge. These findings are consistent with national averages²⁵ in which 44% of 17,448 patients with CHF who received Medicare were readmitted within 6 months of initial hospitalization. The finding that Medicare patients were more likely to be readmitted than those who self-pay or are insured with Medicaid or others emphasizes the need to

References (30)

L.S. Evangelista et al.
Racial differences in treatment seeking delays among heart failure patients
J Card Fail
(2002)
C.A. Sueta et al.
Effects of angiotensin-converting enzyme inhibitor therapy on 30-day outcome in patients 65 years of age or older with chronic congestive heart failure
Am J Cardiol
(2000)
M.P. Kossovsky et al.
Unplanned readmissions of patients with congestive heart failuredo they reflect in-patient quality of care or patient characteristics?
Am J Med
(2000)
J.B. Braunstein et al.
Noncardiac comorbidity increases preventable hospitalizations and mortality among Medicare beneficiaries with chronic heart failure
Am J Cardiol
(2003)
C.A. Polanczyk et al.
Quality of care and hospital readmission in congestive heart failurean explicit review process
J Card Fail
(2001)
S. Stewart et al.
Effects of a multidisciplinary, home-based intervention on unplanned readmissions and survival among patients with chronic congestive heart failure
Lancet
(1999)
E.F. Philbin et al.
Socioeconomic status as independent risk factor for hospital readmission for heart failure
Am J Cardiol
(2001)
E.F. Philbin et al.
Prediction of hospital readmission for heart failuredevelopment of a simple risk score based on administrative data
J Am Coll Cardiol
(1999)
Heart Disease and Stroke Statistics-2004 Update
(2003)
S.S. Rathore et al.
Race, quality of care, and outcomes of elderly patients hospitalized with heart failure
JAMA
(2003)

M.J.L. Alonso et al.

[Hospital readmission in internal medicine]

An Med Interna

(2001)

H.M. Krumholz et al.

Predictors of readmission among elderly survivors of admission with heart failure

Am Heart J

(2000)

M. Tsuchihashi et al.

Medical and socioenvironmental predictors of hospital readmission in patients with CHF

Am Heart J

(2001)

E.F. Philbin et al.

Longer hospital length of stay is not related to better clinical outcomes in congestive heart failure

Am J Manag Care

(1997)

D.P. Holst et al.

Improved outcomes from a comprehensive management system for heart failure

Eur J Heart Fail

(2001)

Cited by (52)

Predictors of early heart failure rehospitalization among older adults with preserved and reduced ejection fraction: A review and derivation of a conceptual model
2023, Heart and Lung
Citation Excerpt :
These models evaluated certain predictors (admission and discharge variables) that helped determine the risk of HF rehospitalizations within six months of discharge after an index HF hospitalization. Hamner and Ellison52 developed four models composed of subsets of variables from a hospital data bank and tested them using logistic regression. The models composed of discharge variables stood out as the only models that predicted rehospitalization at a significant level.52
Heart failure (HF) is prevalent among older adults who suffer with either heart failure preserved ejection fraction (HFpEF) or heart failure reduced ejection fraction (HFrEF) and have a high rate of early HF rehospitalization. Preventing early rehospitalization is complex because of major differences between the two subtypes of HF as well as inadequate predictive models to identify key contributing factors.
To present research addressing relationships between selected clinical, hemodynamic, social factors, and early (≤ 60-day) HF rehospitalization in older adults with HFpEF and HFrEF, derive a conceptual model of predictors of rehospitalization, and understand to what extent the literature addresses these predictors among older women.
Four computerized databases were searched for research addressing clinical, hemodynamic, and social factors relevant to early HF rehospitalization and older adults post index hospitalization for HF.
21 full-text articles were included in the final review and organized thematically. Most studies focused on early (≤ 30-day) HF rehospitalizations, with limited attention given to the 31 to 60-day period. Specific clinical, hemodynamic, and social factors which influenced early HF rehospitalization were identified. The existing literature confirms that risk predictors or their combinations which influence early (≤ 60-day) HF rehospitalization after an index HF hospitalization remains inconsistent. Further, the literature fails to capture the influence of these predictors solely among older women. A conceptual model of risk predictors is proposed for clinical intervention.
Further evaluation to understand risk predictors of early (31 to 60-day) HF rehospitalizations among older women is needed.
Readmission prediction using deep learning on electronic health records
2019, Journal of Biomedical Informatics
Unscheduled 30-day readmissions are a hallmark of Congestive Heart Failure (CHF) patients that pose significant health risks and escalate care cost. In order to reduce readmissions and curb the cost of care, it is important to initiate targeted intervention programs for patients at risk of readmission. This requires identifying high-risk patients at the time of discharge from hospital. Here, using real data from over 7500 CHF patients hospitalized between 2012 and 2016 in Sweden, we built and tested a deep learning framework to predict 30-day unscheduled readmission. We present a cost-sensitive formulation of Long Short-Term Memory (LSTM) neural network using expert features and contextual embedding of clinical concepts. This study targets key elements of an Electronic Health Record (EHR) driven prediction model in a single framework: using both expert and machine derived features, incorporating sequential patterns and addressing the class imbalance problem. We evaluate the contribution of each element towards prediction performance (ROC-AUC, F1-measure) and cost-savings. We show that the model with all key elements achieves higher discrimination ability (AUC: 0.77; F1: 0.51; Cost: 22% of maximum possible savings) outperforming the reduced models in at least two evaluation metrics. Additionally, we present a simple financial analysis to estimate annual savings if targeted interventions are offered to high risk patients.
Access to routine care and risks for 30-day readmission in patients with cardiovascular disease
2018, American Heart Journal
Studies have shown that access to routine medical care is associated with the prevention, diagnosis, and treatment of chronic diseases. However, studies have not examined whether patient-reported difficulties in access to care are associated with rehospitalization in patients with cardiovascular disease.
Electronic medical records and a standardized survey were used to examine cardiovascular patients admitted to a large medical center from January 1, 2015 through January 10, 2017 (n = 520). All-cause readmission within 30 days of discharge was the primary outcome for analysis. Logistic regression models were used to examine the association between access to care and 30-day readmission while adjusting for patient demographics, socioeconomic status, healthcare utilization, and health status.
Nearly 1-in-6 patients (15.7%) reported difficulty in accessing routine medical care; and those who were younger, male, non-white, uninsured, with heart failure, and had low social support were significantly more likely to report difficulty. Patients who reported difficulty in accessing care had significantly higher rates of 30-day readmission than patients who did not report difficulty (33.3% vs. 17.9%; P = .001); and the risks remained largely unchanged after accounting for nearly two dozen covariates (unadjusted odds ratio [OR] = 2.29; 95% CI, 1.46-3.60 vs. adjusted OR = 2.17; 95% CI, 1.29-3.66). Risks for readmission were especially high for patients who reported issues with transportation (OR = 3.24; 95% CI, 1.28-8.16) and scheduling appointments (OR = 3.56; 95% CI, 1.43-8.84), but not for other reasons (OR = 1.47; 95% CI, 0.61-3.54).
Cardiovascular patients who reported difficulty in accessing routine care had substantial risks of readmission within 30 days after discharge. These findings have important implications for identifying high-risk patients and developing interventions to improve access to routine medical care.
Insights from a machine learning model for predicting the hospital Length of Stay (LOS) at the time of admission
2017, Expert Systems with Applications
Citation Excerpt :
Others have pointed out the possibility of a strong association between hypertension and the 30-day readmission rate for CHF patients. Our results indicate that the latter association depends not only on the specific combination of present comorbidities, but also on factors such as the total length of previous hospital stays, and age, as already discussed in the literature (Hamner & Ellison, 2005; Vinson et al., 1990). We found that Ischemic Heart Disease (IHD) is a dominant consequent of many CARMA rules, within both groups (A and B).
A model that accurately predicts, at the time of admission, the Length of Stay (LOS) for hospitalized patients could be an effective tool for healthcare providers. It could enable early interventions to prevent complications, enabling more efficient utilization of manpower and facilities in hospitals. In this study, we apply a regression tree (Cubist) model for predicting the LOS, based on static inputs, that is, values that are known at the time of admission and that do not change during patient's hospital stay. The model was trained and validated on de-identified administrative data from the Veterans Health Administration (VHA) hospitals in Pittsburgh, PA. We chose to use a Cubist model because it produced more accurate predictions than did alternative techniques. In addition, tree models enable us to examine the classification rules learned from the data, in order to better understand the factors that are most correlated with hospital LOS. Cubist recursively partitions the data set as it estimates linear regressions for each partition, and the error level differs for different partitions, so that it is possible to deduce what are the characteristics of patients whose LOS can be accurately predicted at admission, and what are the characteristics of patients for whom the LOS estimate at that point in time is more highly uncertain. For example, our model indicates that the prediction error is greater for patients who had more admissions in the recent past, and for those who had longer previous hospital stays. Our approach suggests that mapping the cases into a higher dimensional space, using a Radial Basis Function (RBF) kernel, helps to separate them by their level of Cubist error, using a Support Vector Machine (SVM).
Risk factors associated with 30-day readmission and length of stay in patients with type 2 diabetes
2017, Journal of Diabetes and its Complications
Patients with type 2 diabetes mellitus (type 2 DM) are at greater risk of poor hospital outcomes. The purpose of this study was to determine the impact of type 2 DM on 30-day hospital readmission and length of stay (LOS).
We studied all inpatient admissions in Pennsylvania during 2011 using data from the Pennsylvania Health Care Cost Containment Council. Outcomes included 30-day readmission and inpatient LOS. We estimated the impact of type 2 DM on readmission and LOS, and identified risk factors for readmission and prolonged LOS.
Among inpatient admissions, patients with diabetes were more likely to be readmitted (AOR = 1.17, P < 0.001) and have longer LOS (0.19 days, P < 0.001) compared to patients without diabetes. Among those with diabetes, several factors were associated with readmission, including demographics, source of admission, and comorbidities. Patients with diabetes were more likely to be readmitted for infectious complications (9.4% vs. 7.7%), heart failure (6.0% vs. 3.1%), and chest pain/MI (5.5% vs. 3.3%) than patients without diabetes.
Diabetes is associated with risk of 30-day readmission and LOS, and several patient-specific factors are associated with outcomes for patients with diabetes. Future studies may target risk factors to develop strategies to reduce readmissions and LOS.
Prediction modeling and pattern recognition for patient readmission
2016, International Journal of Production Economics
Citation Excerpt :
Some policy makers are considering either rewarding hospitals with low readmission rates or penalizing those with high rates, and studies have been done to introduce a consensus method of calculating the readmission rate in order to make it a more reliable quality-of-care indicator for comparison and ranking of hospitals (van Walraven et al., 2012). Research has been done to determine which factors contribute to early re-hospitalization, mostly limiting the population under study to the elderly, patients with a specific disease or a certain ethnicity (Marcantonio et al., 1999; Philbin and DiSalvo, 1999; Ottenbacher et al., 2000; Shyu et al., 2002; Reuben et al., 2002; Hamner and Ellison, 2005; Wong et al., 2010; Allaudeen et al. 2011; Dunlay and Gersh, 2013; Engoren et al., 2013; Njagi et al., 2013). Much attention has been given to predicting hospital readmission within a thirty-day time frame after discharge as a binary value (Marcantonio et al., 1999; Lagoe et al., 2001; Shyu et al., 2002; Wong et al., 2010; van Walraven et al., 2012).
Readmission is a major source of cost for healthcare systems. Hospital-specific readmission rates are considered an indicator of hospital performance and generate public interest regarding the health care quality. We aimed to identify those patients who are likely to be readmitted to the hospital. The identified patients can then be considered by health care personnel for application of preventive alternative measures such as: providing intensive post-discharge care, managing the conditions of the most vulnerable in their home, supporting self-care, and integrating health services and information technology systems to avoid unnecessary readmissions. Neural Network, Classification and Regression model and Chi-squared Automatic Interaction Detection models were used for the readmission prediction. All models were able to perform with an overall accuracy above 80%, with the latter two models having the advantage of providing the user with the opportunity of selecting different misclassification costs. We employed C5.0 algorithm to search for recurring pattern in the history or demographics of patients who have been readmitted and explored if a rule of thumb can be derived to predict those at risk of future readmissions. Moreover, the key variables influencing readmission were studied based on a large data set. The most important factors contributing to readmission were determined such as age, sex, number of previous prescriptions and length of previous stays, place of service, and number of previous claims.

View all citing articles on Scopus

View full text

Issues in cardiovascular nursingPredictors of hospital readmission after discharge in patients with congestive heart failure

Purpose

Methods

Results

Conclusions

Section snippets

Review of research literature

Design

Sample and setting

Discussion

J Card Fail

Am J Cardiol

Am J Med

Am J Cardiol

J Card Fail

Lancet

Am J Cardiol

J Am Coll Cardiol

Heart Disease and Stroke Statistics-2004 Update

Race, quality of care, and outcomes of elderly patients hospitalized with heart failure

JAMA

[Hospital readmission in internal medicine]

An Med Interna

Predictors of readmission among elderly survivors of admission with heart failure

Am Heart J

Medical and socioenvironmental predictors of hospital readmission in patients with CHF

Am Heart J

Longer hospital length of stay is not related to better clinical outcomes in congestive heart failure

Am J Manag Care

Improved outcomes from a comprehensive management system for heart failure

Eur J Heart Fail

Issues in cardiovascular nursing
Predictors of hospital readmission after discharge in patients with congestive heart failure