Acute Myocardial Infarction Readmission Risk Prediction Models: A Systematic Review of Model Performance

An overview of reviews and synthesis across 440 studies examines the importance of hospital readmission predictors across various patient populations

OBJECTIVES

The scientific literature contains an abundance of prediction models for hospital readmissions. However, no review has yet synthesized their predictors across various patient populations. Therefore, our aim was to examine predictors of hospital readmissions across 13 patient populations.

STUDY DESIGN AND SETTING

An overview of systematic reviews was combined with a meta-analytical approach. Two thousand five hundred four different predictors were categorized using common ontologies to pool and examine their odds ratios and frequencies of use in prediction models across and within different patient populations.

RESULTS

Twenty-eight systematic reviews with 440 primary studies were included. Numerous predictors related to prior use of healthcare services (odds ratio; 95% confidence interval: 1.64; 1.42-1.89), diagnoses (1.41; 1.31-1.51), health status (1.35; 1.20-1.52), medications (1.28; 1.13-1.44), administrative information about the index hospitalization (1.23; 1.14-1.33), clinical procedures (1.20; 1.07-1.35), laboratory results (1.18; 1.11-1.25), demographic information (1.10; 1.06-1.14), and socioeconomic status (1.07; 1.02-1.11) were analyzed. Diagnoses were frequently used (in 37.38%) and displayed large effect sizes across all populations. Prior use of healthcare services showed the largest effect sizes but were seldomly used (in 2.57%), whereas demographic information (in 13.18%) was frequently used but displayed small effect sizes.

CONCLUSION

Diagnoses and patients' prior use of healthcare services showed large effects both across and within different populations. These results can serve as a foundation for future prediction modeling.

Prognostic impact of incident left ventricular systolic dysfunction after myocardial infarction

Introduction

We sought to investigate the prognostic impact of incident left ventricular (LV) systolic dysfunction at the chronic phase of acute myocardial infarction (AMI).

Materials and methods

Among 2,266 consecutive patients admitted for AMI, 1,330 patients with LV ejection fraction (LVEF) ≥ 40% during hospitalization who had LVEF data at 6 months after AMI were analyzed. Patients were divided into three subgroups based on LVEF at 6 months: reduced-LVEF (<40%), mid-range-LVEF (≥ 40% and < 50%) and preserved-LVEF (≥ 50%). Occurrence of a composite of hospitalization for heart failure or cardiovascular death after 6 months of AMI was the primary endpoint. The prognostic impact of LVEF at 6 months was assessed with a multivariate-adjusted Cox model.

Results

Overall, the mean patient age was 67.5 ± 11.9 years, and LVEF during initial hospitalization was 59.4 ± 9.1%. The median (interquartile range) duration of follow-up was 3.0 (1.5–4.8) years, and the primary endpoint occurred in 35/1330 (2.6%) patients (13/69 [18.8%] in the reduced-LVEF, 9/265 [3.4%] in the mid-range-LVEF, and 13/996 [1.3%] in the preserved-LVEF category). The adjusted hazard ratio for the primary endpoint in the reduced-LVEF vs. mid-range-LVEF category and in the reduced-LVEF vs. preserved-LVEF category was 4.71 (95% confidence interval [CI], 1.83 to 12.13; p < 0.001) and 14.37 (95% CI, 5.38 to 38.36; p < 0.001), respectively.

Conclusion

Incident LV systolic dysfunction at the chronic phase after AMI was significantly associated with long-term adverse outcomes. Even in AMI survivors without LV systolic dysfunction at the time of AMI, post-AMI reassessment and careful monitoring of LVEF are required to identify patients at risk.

Leveraging Advanced Data Analytics to Predict the Risk of All-Cause Seven-Day Emergency Readmissions

Introduction

Emergency readmissions have been a long-time, multifaceted, unsolved problem. Developing a predictive model calibrated with hospital-specific Electronic Health Record (EHR) data could give higher prediction accuracy and insights into high-risk patients for readmission. Thus, we need to proactively introduce the necessary interventions. This study aims to investigate the relationship between features that consider significant predictors of at-risk patients for seven-day readmission through logistic regression in addition to developing several machine learning models to test the predictability of those attributes using EHR data in a Saudi Arabia-specific ED context.

Methods

Univariate and multivariate logistic regression has been used to identify the most statistically significant features that contributed to classifying readmitted and not readmitted patients. Seven different machine learning models were trained and tested, and a comparison between the best-performing model was conducted in terms of five performance metrics. To construct the prediction model and internally validate it, the processed dataset was split into two sets: 70% for the training set and 30% for the test set or validation set.

Results

XGBoost achieved the highest accuracy (64%) in predicting early seven-day readmissions. Catboost was the second-best predictive model at 61%. XGBoost achieved the highest specificity at 70%, and all the models had a sensitivity of 57% except for XGBoost and Catboost at 32% and 38%, respectively. All predictive attributes, patient age, length of stay (LOS) in minutes, visit time (AM), marital status (married), number of medications, and number of abnormal lab results were significant predictors of early seven-day readmissions while marital status and number of vital-sign instabilities at discharge were not statistically significant predictors of seven-day readmission.

Conclusion

Although XGBoost and Catboost showed good accuracy, none of the models achieved good discriminative ability in terms of sensitivity and specificity. Thus, none can be clinically used for predicting early seven-day readmission. More predictive variables need to be fed into the model, specifically predictors approximate to the day of discharge, in order to optimize the model’s performance.

LACE Index to Predict the High Risk of 30-Day Readmission in Patients With Acute Myocardial Infarction at a University Affiliated Hospital

Background

The LACE index (length of stay, acuity of admission, comorbidity index, and emergency room visit in the past 6 months) has been used to predict the risk of 30-day readmission after hospital discharge in both medical and surgical patients. This study aimed to utilize the LACE index to predict the risk of 30-day readmission in hospitalized patients with acute myocardial infraction (AMI).

Methods

This was a retrospective study. Data were extracted from the hospital's electronic medical records of patients admitted with AMI between 2015 and 2019. LACE index was built on admission patient demographic data, and clinical and laboratory findings during the index of admission. The multivariate logistic regression was performed to determine the association and the risk prediction ability of the LACE index, and 30-day readmission were analyzed by receiver operator characteristic curves with C-statistic.

Results

Of the 3,607 patients included in the study, 5.7% (205) were readmitted within 30 days of discharge from the hospital. The adjusted odds ratio based on logistic regression of all baseline variables showed a statistically significant association with the LACE score and revealed an increased risk of readmission within 30 days of hospital discharge. However, patients with high LACE scores (≥10) had a significantly higher rate of emergency revisits within 30 days from the index discharge than those with low LACE scores. Despite this, analysis of the receiver operating characteristic curve indicated that the LACE index had favorable discrimination ability C-statistic 0.78 (95%CI; 0.75–0.81). The Hosmer–Lemeshow goodness- of-fit test P value was p = 0.920, indicating that the model was well-calibrated to predict risk of the 30-day readmission.

Conclusion

The LACE index demonstrated the good discrimination power to predict the risk of 30-day readmissions for hospitalized patients with AMI. These results can help clinicians to predict the risk of 30-day readmission at the early stage of hospitalization and pay attention during the care of high-risk patients. Future work is to be focused on additional factors to predict the risk of 30-day readmissions; they should be considered to improve the model performance of the LACE index with other acute conditions by using administrative data.

Ability of the LACE Index to Predict 30-Day Readmissions in Patients with Acute Myocardial Infarction

Aims: This study aimed to utilize the existing LACE index (length of stay, acuity of admission, comorbidity index and emergency room visit in the past six months) to predict the risk of 30-day readmission and to find the associated factors in patients with AMI. Methods: This was a retrospective study and LACE index scores were calculated for patients admitted with AMI between 2015 and 2019. Data were utilized from the hospital’s electronic medical record. Multivariate logistic regression was performed to find the association between covariates and 30-day readmission. The risk prediction ability of the LACE index for 30-day readmission was analyzed by receiver operating characteristic curves with the C statistic. Results: A total of 205 (5.7%) patients were readmitted within 30 days. The odds ratio of older age group (OR = 1.78, 95% CI: 1.54–2.05), admission via emergency ward (OR = 1.45; 95% CI: 1.42–1.54) and LACE score ≥10 (OR = 2.71; 95% CI: 1.03–4.37) were highly associated with 30-day readmissions and statistically significant. The receiver operating characteristic curve C statistic of the LACE index for AMI patients was 0.78 (95% CI: 0.75–0.80) and showed favorable discrimination in the prediction of 30-day readmission. Conclusion: The LACE index showed a good discrimination to predict the risk of 30-day readmission for hospitalized patients with AMI. Further study would be recommended to focus on additional factors that can be used to predict the risk of 30-day readmission; this should be considered to improve the model performance of the LACE index for other acute conditions by using the national-based administrative data.

Readmission After ACS: Burden, Epidemiology, and Mitigation

PURPOSE OF REVIEW

The past decade has brought increased efforts to better understand causes for ACS readmissions and strategies to minimize them. This review seeks to provide a critical appraisal of this rapidly growing body of literature.

RECENT FINDINGS

Prior to 2010, readmission rates for patients suffering from ACS remained relatively constant. More recently, several strategies have been implemented to mitigate this including improved risk assessment models, transition care bundles, and development of targeted programs by federal organizations and professional societies. These strategies have been associated with a significant reduction in ACS readmission rates in more recent years. With this, improvements in 30-day post-discharge mortality rates are also being appreciated. As we continue to expand our knowledge on independent risk factors for ACS readmissions, further strategies targeting at-risk populations may further decrease the rate of readmissions. Efforts to understand and reduce 30-day ACS readmission rates have resulted in overall improved quality of care for patients.

Prediction of Cardiovascular and All-Cause Mortality After Myocardial Infarction in US Veterans

Risk prediction models for cardiovascular disease (CVD) death developed from patients without vascular disease may not be suitable for myocardial infarction (MI) survivors. Prediction of mortality risk after MI may help to guide secondary prevention. Using national electronic record data from the Veterans Health Administration 2002 to 2012, we developed risk prediction models for CVD death and all-cause death based on 5-year follow-up data of 100,601 survivors of MI using Cox proportional hazards models. Model performance was evaluated using a cross-validation approach. During follow-up, there were 31,622 deaths and 12,901 CVD deaths. In men, older age, current smoking, atrial fibrillation, heart failure, peripheral artery disease, and lower body mass index were associated with greater risk of death from CVD or all-causes, and statin treatment, hypertension medication, estimated glomerular filtration rate level, and high body mass index were significantly associated with reduced risk of fatal outcomes. Similar associations and slightly different predictors were observed in women. The estimated Harrell's C-statistics of the final model versus the cross-validation estimates were 0.77 versus 0.77 in men and 0.81 versus 0.77 in women for CVD death. Similarly, the C-statistics were 0.75 versus 0.75 in men, 0.78 versus 0.75 in women for all-cause mortality. The predicted risk of death was well calibrated compared with the observed risk. In conclusion, we developed and internally validated risk prediction models of 5-year risk for CVD and all-cause death for outpatient survivors of MI. Traditional risk factors, co-morbidities, and lack of blood pressure or lipid treatment were all associated with greater risk of CVD and all-cause mortality.

Weather Impact on Acute Myocardial Infarction Hospital Admissions With a New Model for Prediction: A Nationwide Study

Introduction: Cardiovascular disease is one of the leading causes of mortality worldwide. Acute myocardial infarction (AMI) is associated with weather change. The study aimed to investigate if weather change was among the risk factors of coronary artery disease to influence AMI occurrence in Taiwan and to generate a model to predict the probabilities of AMI in specific weather and clinical conditions.

Method: This observational study utilized the National Health Insurance Research Database and daily weather reports from Taiwan Central Weather Bureau to evaluate the discharge records of patients diagnosed with AMI from various hospitals in Taiwan between January 1, 2008 and December 31, 2011. Generalized additive models (GAMs) were used to estimate the effective parameters on the trend of the AMI incidence rate with respect to the weather and health factors in the time-series data and to build a model for predicting AMI probabilities.

Results: A total of 40,328 discharges were listed. The minimum temperature, maximum wind speed, and antiplatelet therapy were negatively related to the daily AMI incidence; however, a drop of 1° when the air temperature was below 15°C was associated with an increase of 1.6% of AMI incidence. By using the meaningful parameters including medical and weather factors, an estimated GAM was built. The model showed an adequate correlation in both internal and external validation.

Conclusion: An increase in AMI occurrence in colder weather has been evidenced in the study, but the influence of wind speed remains uncertain. Our analysis demonstrated that the novel GAM model can predict daily onset rates of AMI in specific weather conditions.

Did the Hospital Readmissions Reduction Program Reduce Readmissions? An Assessment of Prior Evidence and New Estimates

In this article, we provide a comprehensive, empirical assessment of the hypothesis that the Hospital Readmissions Reduction Program (HRRP) affected hospital readmissions. In doing so, we provide evidence as to the validity of prior empirical approaches used to evaluate the HRRP and we present results from a previously unused approach to study this research question-a regression-kink design. Results of our analysis document that the empirical approaches used in most prior research assessing the efficacy of the HRRP often lack internal validity. Therefore, results from these studies may not be informative about the causal consequences of the HRRP. Results from our regression-kink analysis, which we validate, suggest that the HRRP had little effect on hospital readmissions. This finding contrasts with the results of most prior studies, which report that the HRRP significantly reduced readmissions. Our finding is consistent with conceptual considerations related to the assumptions underlying HRRP penalty: in particular, the difficulty of identifying preventable readmissions, the highly imperfect risk adjustment that affects the penalty determination, and the absence of proven tools to reduce readmissions.

Development and Validation of a Risk Prediction Model for 1-Year Readmission Among Young Adults Hospitalized for Acute Myocardial Infarction

Background

Readmission over the first year following hospitalization for acute myocardial infarction (AMI) is common among younger adults (≤55 years). Our aim was to develop/validate a risk prediction model that considered a broad range of factors for readmission within 1 year.

Methods and Results

We used data from the VIRGO (Variation in Recovery: Role of Gender on Outcomes of Young AMI Patients) study, which enrolled young adults aged 18 to 55 years hospitalized with AMI across 103 US hospitals (N=2979). The primary outcome was ≥1 all‐cause readmissions within 1 year of hospital discharge. Bayesian model averaging was used to select the risk model. The mean age of participants was 47.1 years, 67.4% were women, and 23.2% were Black. Within 1 year of discharge for AMI, 905 (30.4%) of participants were readmitted and were more likely to be female, Black, and nonmarried. The final risk model consisted of 10 predictors: depressive symptoms (odds ratio [OR], 1.03; 95% CI, 1.01–1.05), better physical health (OR, 0.98; 95% CI, 0.97–0.99), in‐hospital complication of heart failure (OR, 1.44; 95% CI, 0.99–2.08), chronic obstructive pulmomary disease (OR, 1.29; 95% CI, 0.96–1.74), diabetes mellitus (OR, 1.23; 95% CI, 1.00–1.52), female sex (OR, 1.31; 95% CI, 1.05–1.65), low income (OR, 1.13; 95% CI, 0.89–1.42), prior AMI (OR, 1.47; 95% CI, 1.15–1.87), in‐hospital length of stay (OR, 1.13; 95% CI, 1.04–1.23), and being employed (OR, 0.88; 95% CI, 0.69–1.12). The model had excellent calibration and modest discrimination (C statistic=0.67 in development/validation cohorts).

Conclusions

Women and those with a prior AMI, increased depressive symptoms, longer inpatient length of stay and diabetes may be more likely to be readmitted. Notably, several predictors of readmission were psychosocial characteristics rather than markers of AMI severity. This finding may inform the development of interventions to reduce readmissions in young patients with AMI.

Predictive modelling of hospital readmission: Evaluation of different preprocessing techniques on machine learning classifiers

Hospital readmission is a major cost for healthcare systems worldwide. If patients with a higher potential of readmission could be identified at the start, existing resources could be used more efficiently, and appropriate plans could be implemented to reduce the risk of readmission. Therefore, it is important to predict the right target patients. Medical data is usually noisy, incomplete, and inconsistent. Hence, before developing a prediction model, it is crucial to efficiently set up the predictive model so that improved predictive performance is achieved. The current study aims to analyse the impact of different preprocessing methods on the performance of different machine learning classifiers. The preprocessing applied by previous hospital readmission studies were compared, and the most common approaches highlighted such as missing value imputation, feature selection, data balancing, and feature scaling. The hyperparameters were selected using Bayesian optimisation. The different preprocessing pipelines were assessed using various performance metrics and computational costs. The results indicated that the preprocessing approaches helped improve the model’s prediction of hospital readmission.

Development of Acute Myocardial Infarction Mortality and Readmission Models for Public Reporting on Hospital Performance in Canada

Background

Given changes in the care and outcomes of acute myocardial infarction (AMI) patients over the past several decades, we sought to develop prediction models that could be used to generate accurate risk-adjusted mortality and readmission outcomes for hospitals in current practice across Canada.

Methods

A Canadian national expert panel was convened to define appropriate AMI patients for reporting and develop prediction models. Preliminary candidate variable evaluation was conducted using Ontario patients hospitalized with a most responsible diagnosis of AMI from April 1, 2015 to March 31, 2018. National data from the Canadian Institute for Health Information was used to develop AMI prediction models. The main outcomes were 30-day all-cause in-hospital mortality and 30-day urgent all-cause readmission. Discrimination of these models (measured by c-statistics) was compared with that of existing Canadian Institute for Health Information models in the same study cohort.

Results

The AMI mortality model was assessed in 54,240 Ontario AMI patients and 153,523 AMI patients across Canada. We observed a 30-day in-hospital mortality rate of 6.3%, and a 30-day all-cause urgent readmission rate of 10.7% in Canada. The final Canadian AMI mortality model included 12 variables and had a c-statistic of 0.834. For readmission, the model had 13 variables and a c-statistic of 0.679. Discrimination of the new AMI models had higher c-statistics compared with existing models (c-statistic 0.814 for mortality; 0.673 for readmission).

Conclusions

In this national collaboration, we developed mortality and readmission models that are suitable for profiling performance of hospitals treating AMI patients in Canada.

Man vs. Machine: Comparing Physician vs. Electronic Health Record-Based Model Predictions for 30-Day Hospital Readmissions

BACKGROUND

Electronic health record (EHR)-based readmission risk prediction models can be automated in real-time but have modest discrimination and may be missing important readmission risk factors. Clinician predictions of readmissions may incorporate information unavailable in the EHR, but the comparative usefulness is unknown. We sought to compare clinicians versus a validated EHR-based prediction model in predicting 30-day hospital readmissions.

METHODS

We conducted a prospective survey of internal medicine clinicians in an urban safety-net hospital. Clinicians prospectively predicted patients' 30-day readmission risk on 5-point Likert scales, subsequently dichotomized into low- vs. high-risk. We compared human with machine predictions using discrimination, net reclassification, and diagnostic test characteristics. Observed readmissions were ascertained from a regional hospitalization database. We also developed and assessed a "human-plus-machine" logistic regression model incorporating both human and machine predictions.

RESULTS

We included 1183 hospitalizations from 106 clinicians, with a readmission rate of 20.8%. Both clinicians and the EHR model had similar discrimination (C-statistic 0.66 vs. 0.66, p = 0.91). Clinicians had higher specificity (79.0% vs. 48.9%, p < 0.001) but lower sensitivity (43.9 vs. 75.2%, p < 0.001) than EHR model predictions. Compared with machine, human was better at reclassifying non-readmissions (non-event NRI + 30.1%) but worse at reclassifying readmissions (event NRI - 31.3%). A human-plus-machine approach best optimized discrimination (C-statistic 0.70, 95% CI 0.67-0.74), sensitivity (65.5%), and specificity (66.7%).

CONCLUSION

Clinicians had similar discrimination but higher specificity and lower sensitivity than EHR model predictions. Human-plus-machine was better than either alone. Readmission risk prediction strategies should incorporate clinician assessments to optimize the accuracy of readmission predictions.

Home health care after discharge is associated with lower readmission rates for patients with acute myocardial infarction

OBJECTIVE

We studied the utilization of home health care (HHC) among acute myocardial infarction (AMI) patients, impact of HHC on and predictors of 30-day readmission.

METHODS

We queried the National Readmission Database (NRD) from 2012 to 2014identify patients with AMI discharged home with (HHC+) and without HHC (HHC-). Linkage provided in the data identified patients who had 30-day readmission, our primary end-point. The probability for each patient to receive HHC was calculated by a multivariable logistic regression. Average treatment of treated weights were derived from propensity scores. Weight-adjusted logistic regression was used to determine impact of HHC on readmission.

RESULTS

A total of 406 237 patients with AMI were discharged home. Patients in the HHC+ cohort (38 215 patients, 9.4%) were older (mean age 77 vs. 60 years P < 0.001), more likely women (53 vs. 26%, P < 0.001), have heart failure (5 vs. 0.5%, P < 0.001), chronic kidney disease (26 vs. 6%, P < 0.001) and diabetes (35 vs. 26%, P < 0.001). Patients readmitted within 30-days were older with higher rates of diabetes (RR = 1.4, 95% CI: 1.37-1.48) and heart failure (RR = 5.8, 95% CI: 5.5-6.2). Unadjusted 30-day readmission rates were 21 and 8% for HHC+ and HHC- patients, respectively. After adjustment, readmission was lower with HHC (21 vs. 24%, RR = 0.89, 95% CI: 0.82-0.96; P < 0.001).

CONCLUSION

In the United States, AMI patients receiving HHC are older and have more comorbidities; however, HHC was associated with a lower 30-day readmission rate.

Prediction models for hospital readmissions in patients with heart disease: a systematic review and meta-analysis

OBJECTIVE

To describe the discrimination and calibration of clinical prediction models, identify characteristics that contribute to better predictions and investigate predictors that are associated with unplanned hospital readmissions.

DESIGN

Systematic review and meta-analysis.

DATA SOURCE

Medline, EMBASE, ICTPR (for study protocols) and Web of Science (for conference proceedings) were searched up to 25 August 2020.

ELIGIBILITY CRITERIA FOR SELECTING STUDIES

Studies were eligible if they reported on (1) hospitalised adult patients with acute heart disease; (2) a clinical presentation of prediction models with c-statistic; (3) unplanned hospital readmission within 6 months.

PRIMARY AND SECONDARY OUTCOME MEASURES

Model discrimination for unplanned hospital readmission within 6 months measured using concordance (c) statistics and model calibration. Meta-regression and subgroup analyses were performed to investigate predefined sources of heterogeneity. Outcome measures from models reported in multiple independent cohorts and similarly defined risk predictors were pooled.

RESULTS

Sixty studies describing 81 models were included: 43 models were newly developed, and 38 were externally validated. Included populations were mainly patients with heart failure (HF) (n=29). The average age ranged between 56.5 and 84 years. The incidence of readmission ranged from 3% to 43%. Risk of bias (RoB) was high in almost all studies. The c-statistic was <0.7 in 72 models, between 0.7 and 0.8 in 16 models and >0.8 in 5 models. The study population, data source and number of predictors were significant moderators for the discrimination. Calibration was reported for 27 models. Only the GRACE (Global Registration of Acute Coronary Events) score had adequate discrimination in independent cohorts (0.78, 95% CI 0.63 to 0.86). Eighteen predictors were pooled.

CONCLUSION

Some promising models require updating and validation before use in clinical practice. The lack of independent validation studies, high RoB and low consistency in measured predictors limit their applicability.

PROSPERO REGISTRATION NUMBER

CRD42020159839.

Current Trends in Readmission Prediction: An Overview of Approaches

Hospital readmission shortly after discharge threatens the quality of patient care and leads to increased medical care costs. In the United States, hospitals with high readmission rates are subject to federal financial penalties. This concern calls for incentives for healthcare facilities to reduce their readmission rates by predicting patients who are at high risk of readmission. Conventional practices involve the use of rule-based assessment scores and traditional statistical methods, such as logistic regression, in developing risk prediction models. The recent advancements in machine learning driven by improved computing power and sophisticated algorithms have the potential to produce highly accurate predictions. However, the value of such models could be overrated. Meanwhile, the use of other flexible models that leverage simple algorithms offer great transparency in terms of feature interpretation, which is beneficial in clinical settings. This work presents an overview of the current trends in risk prediction models developed in the field of readmission. The various techniques adopted by researchers in recent years are described, and the topic of whether complex models outperform simple ones in readmission risk stratification is investigated.

Patient-Reported Outcomes Predict Future Emergency Department Visits and Hospital Admissions in Patients With Stroke

Background

Identification of stroke patients at increased risk of emergency department (ED) visits or hospital admissions allows implementation of mitigation strategies. We evaluated the ability of the Patient‐Reported Outcomes Information Measurement System (PROMIS) patient‐reported outcomes (PROs) collected as part of routine care to predict 1‐year emergency department (ED) visits and admissions when added to other readily available clinical variables.

Methods and Results

This was a cohort study of 1696 patients with ischemic stroke, intracerebral hemorrhage, subarachnoid hemorrhage, or transient ischemic attack seen in a cerebrovascular clinic from February 17, 2015, to June 11, 2018, who completed the following PROs at the visit: Patient Health Questionnaire‐9, Quality of Life in Neurological Disorders cognitive function, PROMIS Global Health, sleep disturbance, fatigue, anxiety, social role satisfaction, physical function, and pain interference. A series of logistic regression models was constructed to determine the ability of models that include PRO scores to predict 1‐year ED visits and all‐cause and unplanned admissions. In the 1 year following the PRO encounter date, 1046 ED visits occurred in 548 patients; 751 admissions occurred in 453 patients. All PROs were significantly associated with future ED visits and admissions except PROMIS sleep. Models predicting unplanned admissions had highest optimism‐corrected area under the curve (range, 0.684–0.724), followed by ED visits (range, 0.674–0.691) and then all‐cause admissions (range, 0.628–0.671). PROs measuring domains of mental health had stronger associations with ED visits; PROs measuring domains of physical health had stronger associations with admissions.

Conclusions

PROMIS scales improve the ability to predict ED visits and admissions in patients with stroke. The differences in model performance and the most influential PROs in the prediction models suggest differences in factors influencing future hospital admissions and ED visits.

Development of Electronic Health Record-Based Prediction Models for 30-Day Readmission Risk Among Patients Hospitalized for Acute Myocardial Infarction

IMPORTANCE

In the US, more than 600 000 adults will experience an acute myocardial infarction (AMI) each year, and up to 20% of the patients will be rehospitalized within 30 days. This study highlights the need for consideration of calibration in these risk models.

OBJECTIVE

To compare multiple machine learning risk prediction models using an electronic health record (EHR)-derived data set standardized to a common data model.

DESIGN, SETTING, AND PARTICIPANTS

This was a retrospective cohort study that developed risk prediction models for 30-day readmission among all inpatients discharged from Vanderbilt University Medical Center between January 1, 2007, and December 31, 2016, with a primary diagnosis of AMI who were not transferred from another facility. The model was externally validated at Dartmouth-Hitchcock Medical Center from April 2, 2011, to December 31, 2016. Data analysis occurred between January 4, 2019, and November 15, 2020.

EXPOSURES

Acute myocardial infarction that required hospital admission.

MAIN OUTCOMES AND MEASURES

The main outcome was thirty-day hospital readmission. A total of 141 candidate variables were considered from administrative codes, medication orders, and laboratory tests. Multiple risk prediction models were developed using parametric models (elastic net, least absolute shrinkage and selection operator, and ridge regression) and nonparametric models (random forest and gradient boosting). The models were assessed using holdout data with area under the receiver operating characteristic curve (AUROC), percentage of calibration, and calibration curve belts.

RESULTS

The final Vanderbilt University Medical Center cohort included 6163 unique patients, among whom the mean (SD) age was 67 (13) years, 4137 were male (67.1%), 1019 (16.5%) were Black or other race, and 933 (15.1%) were rehospitalized within 30 days. The final Dartmouth-Hitchcock Medical Center cohort included 4024 unique patients, with mean (SD) age of 68 (12) years; 2584 (64.2%) were male, 412 (10.2%) were rehospitalized within 30 days, and most of the cohort were non-Hispanic and White. The final test set AUROC performance was between 0.686 to 0.695 for the parametric models and 0.686 to 0.704 for the nonparametric models. In the validation cohort, AUROC performance was between 0.558 to 0.655 for parametric models and 0.606 to 0.608 for nonparametric models.

CONCLUSIONS AND RELEVANCE

In this study, 5 machine learning models were developed and externally validated to predict 30-day readmission AMI hospitalization. These models can be deployed within an EHR using routinely collected data.

A stacking-based model for predicting 30-day all-cause hospital readmissions of patients with acute myocardial infarction

BACKGROUND

Acute myocardial infarction (AMI) is a serious cardiovascular disease, followed by a high readmission rate within 30-days of discharge. Accurate prediction of AMI readmission is a crucial way to identify the high-risk group and optimize the distribution of medical resources.

METHODS

In this study, we propose a stacking-based model to predict the risk of 30-day unplanned all-cause hospital readmissions for AMI patients based on clinical data. Firstly, we conducted an under-sampling method of neighborhood cleaning rule (NCR) to alleviate the class imbalance and then utilized a feature selection method of SelectFromModel (SFM) to select effective features. Secondly, we adopted a self-adaptive approach to select base classifiers from eight candidate models according to their performances in datasets. Finally, we constructed a three-layer stacking model in which layer 1 and layer 2 were base-layer and level 3 was meta-layer. The predictions of the base-layer were used to train the meta-layer in order to make the final forecast.

RESULTS

The results show that the proposed model exhibits the highest AUC (0.720), which is higher than that of decision tree (0.681), support vector machine (0.707), random forest (0.701), extra trees (0.709), adaBoost (0.702), bootstrap aggregating (0.704), gradient boosting decision tree (0.710) and extreme gradient enhancement (0.713).

CONCLUSION

It is evident that our model could effectively predict the risk of 30-day all cause hospital readmissions for AMI patients and provide decision support for the administration.

Clinical Predictive Models of Sudden Cardiac Arrest: A Survey of the Current Science and Analysis of Model Performances

Background

More than 500 000 sudden cardiac arrests (SCAs) occur annually in the United States. Clinical predictive models (CPMs) may be helpful tools to differentiate between patients who are likely to survive or have good neurologic recovery and those who are not. However, which CPMs are most reliable for discriminating between outcomes in SCA is not known.

Methods and Results

We performed a systematic review of the literature using the Tufts PACE (Predictive Analytics and Comparative Effectiveness) CPM Registry through February 1, 2020, and identified 81 unique CPMs of SCA and 62 subsequent external validation studies. Initial cardiac rhythm, age, and duration of cardiopulmonary resuscitation were the 3 most commonly used predictive variables. Only 33 of the 81 novel SCA CPMs (41%) were validated at least once. Of 81 novel SCA CPMs, 56 (69%) and 61 of 62 validation studies (98%) reported discrimination, with median c‐statistics of 0.84 and 0.81, respectively. Calibration was reported in only 29 of 62 validation studies (41.9%). For those novel models that both reported discrimination and were validated (26 models), the median percentage change in discrimination was −1.6%. We identified 3 CPMs that had undergone at least 3 external validation studies: the out‐of‐hospital cardiac arrest score (9 validations; median c‐statistic, 0.79), the cardiac arrest hospital prognosis score (6 validations; median c‐statistic, 0.83), and the good outcome following attempted resuscitation score (6 validations; median c‐statistic, 0.76).

Conclusions

Although only a small number of SCA CPMs have been rigorously validated, the ones that have been demonstrate good discrimination.

Comorbid vision and cognitive impairments in older adults hospitalized for acute myocardial infarction

Older patients presenting with acute myocardial infarction (AMI) often have comorbidities. Our objective was to examine how outcomes differ by cognitive and vision status in older AMI patients. We use data from a prospective cohort study conducted at 94 hospitals in the United States between January 2013 and October 2016 that enrolled men and women aged ≥75 years with AMI. Cognitive impairment (CI) was defined as telephone interview for cognitive status (TICS) score <27; vision impairment (VI) and activities of daily living (ADLs) were assessed by questionnaire. Of 2988 senior AMI patients, 260 (8.7%) had CI but no VI, 858 (28.7%) had VI but no CI, and 251 (8.4%) had both CI/VI. Patients in the VI/CI group were most likely to exhibit geriatric syndromes. More severe VI was associated with lower (worse) scores on the TICS (β −1.53, 95% confidence interval (CI) −1.87 to −1.18). In adjusted models, compared to participants with neither impairment, participants with VI/CI were more likely to die (hazard ratio 1.61, 95% CI 1.10–2.37) and experience ADL decline (odds ratio 2.11, 95% CI 1.39–3.21) at 180 days. Comorbid CIs and VIs were associated with high rates of death and worsening disability after discharge among seniors hospitalized for AMI. Future research should evaluate protocols to accommodate these impairments during AMI presentations and optimize decision-making and outcomes.

Thirty-Day Hospital Readmission After Acute Myocardial Infarction in China

Background Readmission after acute myocardial infarction in low- and middle-income countries like China is not well characterized. Methods and Results We approached consecutive patients with acute myocardial infarction hospitalized within 24 hours of symptom onset and discharged alive from 53 geographically diverse hospitals in China. We described rates of unplanned 30-day readmission, their timing and admitting diagnoses, and fit Cox proportional hazards models to identify factors associated with readmission. Among 3387 patients, median (interquartile range) age was 61 (52-69) years, and 76.9% were men. The index median length of stay was 11 (8-14) days. Unplanned 30-day readmission occurred in 6.3% of the cohort; most readmissions (77.7%) were for cardiovascular diagnoses. Nearly half (41.9% of all-cause readmissions; 44.3% of cardiovascular readmissions) occurred within 5 days of discharge. Mini-Global Registry of Acute Coronary Events scores at admission (hazard ratio [HR], 1.15 for every 10-point increase; 95% CI, 1.05-1.25), longer length of stay (HR, 1.03; 95% CI, 1.00-1.06 for each extra day), and in-hospital recurrent angina (HR, 1.40; 95% CI, 1.04-1.89) were associated with higher unplanned all-cause readmission. Revascularization during the index hospitalization (70.2% of the cohort) was associated with lower risks of all-cause readmission (HR, 0.27; 95% CI, 0.18-0.42). In addition, left ventricular ejection fraction <0.4 (HR, 1.79; 95% CI, 1.05-3.07) and in-hospital complication (HR, 1.20; 95% CI, 1.03-1.39) were associated with higher risk of unplanned cardiovascular readmission, and ST-segment-elevation myocardial infarction (HR, 0.60; 95% CI, 0.36-0.98) was associated with lower risk of unplanned cardiovascular readmission. Sex, family income, depression, stress level, lower social support, disease-specific health status, and medications were not associated with readmission. Conclusions In China, most readmissions are for cardiovascular events, and almost half occur within 5 days of discharge. Clinical factors identify patients at higher and lower unplanned readmissions. Clinical Trial Registration URL: https://www.clinicaltrials.gov . Unique identifier: NCT01624909.

Thirty-Day Readmission Risk Model for Older Adults Hospitalized With Acute Myocardial Infarction

BACKGROUND

Early readmissions among older adults hospitalized for acute myocardial infarction (AMI) are costly and difficult to predict. Aging-related functional impairments may inform risk prediction but are unavailable in most studies. Our objective was to, therefore, develop and validate an AMI readmission risk model for older patients who considered functional impairments and was suitable for use before hospital discharge.

METHODS AND RESULTS

SILVER-AMI (Comprehensive Evaluation of Risk in Older Adults with AMI) is a prospective cohort study of 3006 patients of age ≥75 years hospitalized with AMI at 94 US hospitals. Participants underwent in-hospital assessment of functional impairments including cognition, vision, hearing, and mobility. Other variables plausibly associated with readmissions were also collected. The outcome was all-cause readmission at 30 days. We used backward selection and Bayesian model averaging to derive (N=2004) a risk model that was subsequently validated (N=1002). Mean age was 81.5 years, 44.4% were women, and 10.5% were nonwhite. Within 30 days, 547 participants (18.2%) were readmitted. Readmitted participants were older, had more comorbidities, and had a higher prevalence of functional impairments, including activities of daily living disability (17.0% versus 13.0%; P=0.013) and impaired functional mobility (72.5% versus 53.6%; P<0.001). The final risk model included 8 variables: functional mobility, ejection fraction, chronic obstructive pulmonary disease, arrhythmia, acute kidney injury, first diastolic blood pressure, P2Y12 inhibitor use, and general health status. Functional mobility was the only functional impairment variable retained but was the strongest predictor. The model was well calibrated (Hosmer-Lemeshow P value >0.05) with moderate discrimination (C statistics: 0.65 derivation cohort and 0.63 validation cohort). Functional mobility significantly improved performance of the risk model (net reclassification improvement index =20%; P<0.001).

CONCLUSIONS

In our final risk model, functional mobility, previously not included in readmission risk models, was the strongest predictor of 30-day readmission among older adults after AMI. The modest discrimination indicates that much of the variability in readmission risk among this population remains unexplained by patient-level factors.

CLINICAL TRIAL REGISTRATION

URL: https://www.clinicaltrials.gov. Unique identifier: NCT01755052.

Clinical and procedural implications of congenital vena cava anomalies in adults: A systematic review

BACKGROUND

Although congenital vena cava (CVC) anomalies in adults have implications for surgical and radiological interventions, the literature is scare and disparate. The aim of this systematic review was to assess cardiovascular clinical and procedural implications of CVC anomalies in adults without congenital heart disease.

METHODS AND RESULTS

We searched PubMed and EMBASE from database conception through October 2018 for English-language studies describing the epidemiology of CVC anomalies or their clinical or procedural implications in humans. Two independent reviewers screened 7093 records and identified 16 relevant studies. We found two major implications of CVC anomalies: 1) congenital inferior vena cava (CIVC) anomalies are associated with a 50-100-fold higher risk of deep venous thrombosis, particularly among younger patients, and 2) persistent left superior vena cava (PLSVC) is associated with a 2-3-fold higher risk of supraventricular arrhythmias. PLSVC also poses technical challenges to cardiovascular electronic device implantation, requiring alterations in surgical approach and lengthening procedure and X-ray exposure times. Due to the large disparity in reported prevalence rates of CIVC anomalies, we performed a meta-analysis of CIVC anomaly prevalence including 8 studies, which showed a weighted prevalence of 6.8% (95% CI, 4.5-9.2%).

CONCLUSION

These findings challenge the notion that CVC anomalies are rare and asymptomatic in adults. Rather, the literature indicates that CVC anomalies are not uncommon and have important clinical and procedural implications. To further understand the prevalence and implications of CVC anomalies, a robust US population-based study and nationwide registry is warranted in the current era of venous interventions.

Thirty-Day Readmission After Medical Versus Endovascular Therapy for Atherosclerotic Renal Artery Stenosis

Whether renal artery stenting (RAS) confers benefit over medical therapy (MT) alone in patients with atherosclerotic renal artery stenosis admitted with acute coronary syndromes (ACS), congestive heart failure (CHF), or hypertensive crisis remains unknown. We identified a nationally-weighted cohort of 116,056 patients from the Nationwide Readmissions Database with a preexisting diagnosis of atherosclerotic renal artery stenosis and an index hospitalization diagnosis of ACS, CHF, or hypertensive crisis, and propensity-matched on the likelihood of undergoing inpatient RAS. Thirty-day readmission rates, index hospitalization complications, hospital lengths-of-stay, and cost were compared between treatment groups. Overall, all-cause, nonelective 30-day readmission rates did not differ between RAS and MT alone (18.2% vs 18.7%, respectively, p = 0.49). RAS was associated with higher index rates of acute kidney injury, major bleeding, transfusion, and vascular complications, and were similar irrespective of index hospitalization diagnosis. Index hospitalization length of stay (6 vs 4 days; p <0.001) and cost ($23,020 vs. $11,459; p <0.001) were higher with RAS. In conclusion, nearly 1-in-5 patients hospitalized with atherosclerotic renal artery stenosis and ACS, CHF, or hypertensive crisis were readmitted within 30-days. Index hospitalization complications occurred more frequently among those treated with RAS than MT alone, but the likelihood of readmission did not differ by treatment strategy.

Usefulness of Social Support in Older Adults After Hospitalization for Acute Myocardial Infarction (from the SILVER-AMI Study)

The availability of social support is associated with health outcomes after acute myocardial infarction (AMI), yet previous studies have largely considered social support as a single entity, rather than examining its discrete domains. Furthermore, few studies have investigated the impact of social support in older AMI patients, in whom it may be especially important. We aimed to determine the associations between 5 discrete domains of social support - emotional support, informational support, tangible support, positive social interaction, and affectionate support - with 6-month readmission and mortality in older patients hospitalized for AMI, adjusting for known predictors of post-AMI outcomes. Three thousand six participants 75 years and older were recruited from a network of 94 hospitals across the United States. A 5-item version of the Medical Outcomes Study Social Support Survey was used to measure perceived social support, and readmission and mortality were ascertained 6 months after initial hospitalization. Independent associations were determined using multivariable regression. Among 3,006 participants, mean age was 82 years, 44% were female, and 11% non-white. Participants who were female, non-white, less educated, and lived alone tended to report lower social support. In multivariable analyses, low informational support was associated with readmission (odds ratio 1.22; 95% confidence interval 1.01 to 1.47), and low emotional support with mortality (odds ratio 1.43; 95% confidence interval 1.04 to 1.97). In conclusion, individual domains of social support had distinct, independent associations with post-AMI outcomes, lending a more nuanced and precise understanding of this important social determinant of health. Understanding these distinct associations can inform the development of interventions and policies to improve post-AMI outcomes.

Acute Kidney Injury Among Older Patients Undergoing Coronary Angiography for Acute Myocardial Infarction: The SILVER-AMI Study

BACKGROUND

Among older adults (age ≥75 years) hospitalized for acute myocardial infarction, acute kidney injury after coronary angiography is common. Aging-related conditions may independently predict acute kidney injury, but have not yet been analyzed in large acute myocardial infarction cohorts.

METHODS

We analyzed data from 2212 participants age ≥75 years in the Comprehensive Evaluation of Risk Factors in Older Patients with Acute Myocardial Infarction (SILVER-AMI) study who underwent coronary angiography. Acute kidney injury was defined using Kidney Disease Improving Global Outcomes (KDIGO) criteria (serum Cr increase ≥0.3 mg/dL from baseline or ≥1.5 times baseline). We analyzed the associations of traditional acute kidney injury risk factors and aging-related conditions (activities of daily living impairment, prior falls, cachexia, low physical activity) with acute kidney injury, and then performed logistic regression to identify independent predictors.

RESULTS

Participants' mean age was 81.3 years, 45.2% were female, and 9.5% were nonwhite; 421 (19.0%) experienced acute kidney injury. Comorbid diseases and aging-related conditions were both more common among individuals experiencing acute kidney injury. However, after multivariable adjustment, no aging-related conditions were retained. There were 11 risk factors in the final model; the strongest were heart failure on presentation (odds ratio [OR] 1.91; 95% confidence interval [CI], 1.41-2.59), body mass index [BMI] >30 (vs BMI 18-25: OR 1.75; 95% CI, 1.27-2.42), and nonwhite race (OR 1.65; 95% CI, 1.16-2.33). The final model achieved an area under the receiver operating characteristic curve of 0.72 and was well calibrated (Hosmer-Lemeshow P = .50). Acute kidney injury was independently associated with 6-month mortality (OR 1.98; 95% CI, 1.36-2.88) but not readmission (OR 1.26; 95% CI, 0.98-1.61).

CONCLUSIONS

Acute kidney injury is common among older adults with acute myocardial infarction undergoing coronary angiography. Predictors largely mirrored those in previous studies of younger individuals, which suggests that geriatric conditions mediate their influence through other risk factors.

A Framework for Integrating Domain Knowledge in Logistic Regression with Application to Hospital Readmission Prediction

It is commonly understood that machine learning algorithms discover and extract knowledge based on data at hand. However, a huge amount of knowledge is available which is in machine-readable format and ready for inclusion in machine learning algorithms and models. In this paper, we propose a framework that integrates domain knowledge in form of ontologies/hierarchies into logistic regression using stacked generalization. Namely, relations from ontology/hierarchy are used in stacking manner in order to obtain higher, more abstract concepts. Obtained concepts are further used for prediction. The problem we solved is unplanned 30-days hospital readmission, which is considered as one of the major problems in healthcare. Proposed framework yields better results compared to Ridge, Lasso, and Tree Lasso Logistic Regression. Results suggest that the proposed framework improves AUC by up to 9.5% on pediatric datasets and up to 4% on morbidly obese patients’ datasets and also improves AUPRC by up to 5.7% on pediatric datasets and up to 2.6% on morbidly obese patients’ datasets on average. This indicates that the inclusion of domain knowledge improves the predictive performance of Logistic Regression.

When a Short-Term Outlook Is the Best Long-Term Strategy: Time-Varying Risk of Readmission After Acute Myocardial Infarction

See Article by Khot et al

The prevalence of 30-day readmission after acute myocardial infarction: A systematic review and meta-analysis

Objective

The 30‐day readmission is associated with increased medical costs, which has become an important quality metric in several medical institutions. This current study is aimed at clarifying the prevalence, the underlying risk factors, and reasons of the 30‐day readmission after acute myocardial infarction (AMI).

Methods

PubMed, Cochrane Library, and EMBASE were systematically searched to identify eligible studies. Random‐effect models were employed to perform pooled analyses. Means and 95% confidence intervals (CIs) were used to estimate prevalence and reasons for 30‐day readmission. We also used Odds ratios (ORs) to explore the potential significant predictors of risk factors of 30‐day readmission after AMI. Potential publication bias was assessed using funnel plot and Begg'test.

Results

A total of 14 relevant studies were included in this systematic review and meta‐analysis. The pooled 30‐day readmission rate of AMI was 12% (95% CI 0.11‐0.14). Acute coronary syndrome (ACS), angina and acute ischemic heart disease, and heart failure (HF) were the principal cardiovascular reasons of 30‐day readmission. Meanwhile, non‐specific chest pain was regarded as the significant cause among non‐cardiovascular reasons. The common co‐morbidities kidney disease, HF and diabetes mellitus were significant risk factors for 30‐day readmission. No significant publication bias was found by funnel plot and statistical tests.

Conclusions

The 30‐day readmission rate of post‐AMI ranged from 11% to 14% and can be mainly attributed to cardiovascular and non‐cardiovascular events. The common co‐morbidities, such as kidney disease, HF, and diabetes mellitus were significant risk factors for 30‐day readmission.

Predictors and Risk Calculator of Early Unplanned Hospital Readmission Following Contemporary Self-Expanding Transcatheter Aortic Valve Replacement from the STS/ACC TVT Registry

BACKGROUND

Predictors of hospital readmissions and tools to predict readmissions after TAVR are scarce. Our objective was to identify predictors of early hospital readmission following TAVR in contemporary clinical practice and develop a risk calculator.

METHODS

Patients with a contemporary self-expanding TAVR between 2015 and 2017 in the STS/ACC/TVT Registry™ database were included. Patients were divided into a derivation and validation cohort (2:1). A risk score was calculated using the derivation cohort based on multivariable predictors of 30-day unplanned readmissions and applied to the validation cohort.

RESULTS

A total of 10,345 TAVR patients at 350 centers were included. Unplanned 30-day hospital readmission was 9.2%. Patients with an early readmission had higher 30-day rates for mortality (2.3% vs. 0.8%, p ≪ 0.001), stroke (4.1% vs. 2.7% p = 0.009), major vascular complications (2.0% vs. 1.0%, p = 0.003) and new pacemaker implantation (25.7% vs. 18.6%, p ≪ 0.001). Multivariable predictors of 30-day readmission included diabetes, atrial fibrillation, advanced heart failure symptoms, home oxygen, decreased 5-m gait speed or the inability to walk, serum creatinine ≫1.6 mg/dL, index hospitalization length of stay ≫5 days, major vascular complication and ≥ moderate post-procedure aortic or mitral valve regurgitation. Based on these predictors, we stratified 30-day readmission risk into low-, moderate- and high-risk subsets. There was a 2.5× difference in readmission rates between the low- (5.8%) and high-risk subsets (14.6%).

CONCLUSION

We stratified the risk of early hospital readmission after TAVR based on a simple scoring system. This score may improve discharge planning centered on the individual's readmission risk.

SUMMARY

Unplanned readmissions in the United States are prevalent and costly accounting for $41.3 billion in annual hospital payments and are associated with adverse clinical outcomes. We found that diabetes, atrial fibrillation, advanced heart failure symptoms, home oxygen, frailty, acute kidney injury, prolonged hospitalization, major vascular complications, and moderate or worse post-procedure aortic or mitral valve regurgitation predicted of 30-day readmission following self-expanding TAVR. This information may improve discharge planning centered on each patient's readmission risk.

Unplanned Readmissions After Acute Myocardial Infarction: 1-Year Trajectory Following Discharge From a Safety Net Hospital

BACKGROUND

Financial penalties rendered by the Centers for Medicare and Medicaid Services have brought about new challenges for safety net hospitals that serve a vulnerable patient population with risk factors associated with high readmission rates. Our goal was to determine the 1-year trajectory of unplanned readmissions in post-myocardial infarction (MI) patients, and to identify factors associated with readmission.

METHODS

A total of 261 acute MI patients admitted from April 2015 to April 2016 were evaluated in a multidisciplinary cardiology clinic within 10 days of hospital discharge and baseline characteristics and medical comorbidities were collected. Readmission and mortality data were obtained at 1 year through chart review and telephone follow-up.

RESULTS

At 1 year, there were 90 (34%) unplanned readmissions of which half were for noncardiac diagnoses. Of these, 69 patients (77%) were readmitted once, 16 (18%) were readmitted twice, 2 (2%) were readmitted 3 times, and 3 (3%) were readmitted 4 times over the subsequent year. Cardiac causes of 1-year readmission included recurrent MI in 23 (9%) and decompensated heart failure in 18 (7%) patients. Depressed left ventricular systolic function (hazard ratio, 2.23; 95% confidence interval, 2.00-2.44; P = 0.0003) and diabetes mellitus (hazard ratio, 1.60; 95% confidence interval, 1.38-1.82; P = 0.029) were associated with a significantly higher risk of readmission at 1 year.

CONCLUSION

Following acute MI, patients are readmitted for cardiac and noncardiac diagnoses well beyond the 30-day mark. This is likely a function of the vulnerability of the patient population rather than a reflection of the medical care provided. More frequent surveillance may attenuate this problem.

Predicting 30-Day Hospital Readmissions in Acute Myocardial Infarction: The AMI "READMITS" (Renal Function, Elevated Brain Natriuretic Peptide, Age, Diabetes Mellitus, Nonmale Sex, Intervention with Timely Percutaneous Coronary Intervention, and Low Systolic Blood Pressure) Score

BACKGROUND

Readmissions after hospitalization for acute myocardial infarction (AMI) are common. However, the few currently available AMI readmission risk prediction models have poor-to-modest predictive ability and are not readily actionable in real time. We sought to develop an actionable and accurate AMI readmission risk prediction model to identify high-risk patients as early as possible during hospitalization.

METHODS AND RESULTS

We used electronic health record data from consecutive AMI hospitalizations from 6 hospitals in north Texas from 2009 to 2010 to derive and validate models predicting all-cause nonelective 30-day readmissions, using stepwise backward selection and 5-fold cross-validation. Of 826 patients hospitalized with AMI, 13% had a 30-day readmission. The first-day AMI model (the AMI "READMITS" score) included 7 predictors: renal function, elevated brain natriuretic peptide, age, diabetes mellitus, nonmale sex, intervention with timely percutaneous coronary intervention, and low systolic blood pressure, had an optimism-corrected C-statistic of 0.73 (95% confidence interval, 0.71-0.74) and was well calibrated. The full-stay AMI model, which included 3 additional predictors (use of intravenous diuretics, anemia on discharge, and discharge to postacute care), had an optimism-corrected C-statistic of 0.75 (95% confidence interval, 0.74-0.76) with minimally improved net reclassification and calibration. Both AMI models outperformed corresponding multicondition readmission models.

CONCLUSIONS

The parsimonious AMI READMITS score enables early prospective identification of high-risk AMI patients for targeted readmissions reduction interventions within the first 24 hours of hospitalization. A full-stay AMI readmission model only modestly outperformed the AMI READMITS score in terms of discrimination, but surprisingly did not meaningfully improve reclassification.