The readmission risk flag: Using the electronic health record to automatically identify patients at risk for 30‐day readmission

Abstract

Background Identification of patients at high risk for readmission is a crucial step toward improving care and reducing readmissions. The adoption of electronic health records (EHR) may prove important to strategies designed to risk stratify patients and introduce targeted interventions. Objective To develop and implement an automated prediction model integrated into our health system’s EHR that identifies on admission patients at high risk for readmission within 30 days of discharge. Design Retrospective and prospective cohort. Setting Healthcare system consisting of three hospitals. Patients All adult patients admitted from August 2009 to September 2012. Interventions An automated readmission risk flag integrated into the EHR. Measures Thirty-day all-cause and 7-day unplanned healthcare system readmissions. Results Using retrospective data, a single risk factor, ≥2 inpatient admissions in the past 12 months, was found to have the best balance of sensitivity (40%), positive predictive value (31%), and proportion of patients flagged (18%), with a c-statistic of 0.62. Sensitivity (39%), positive predictive value (30%), proportion of patients flagged (18%) and c-statistic (0.61) during the 12-month period after implementation of the risk flag were similar. There was no evidence for an effect of the intervention on 30-day all-cause and 7-day unplanned readmission rates in the 12-month period after implementation. Conclusions An automated prediction model was effectively integrated into an existing EHR and identified patients on admission who were at risk for readmission within 30 days of discharge.

Post–Acute Care Use and Hospital Readmission after Sepsis

“…Readmissions were considered outcomes and also served as index hospitalizations (13,21,22). Secondary outcomes included post-acute care use at discharge, 7-and 90-day readmissions, emergency department treat-and-release visits within 30 days, hospital-based acute care use within 30 days (23), and outcomes after readmission.…”

“…We hypothesized that variables captured in recently validated readmission prediction scores (21,24) would be concentrated in sepsis survivors. We also contrasted readmission rates after sepsis hospitalizations with hospitalizations associated with the high-risk conditions of acute myocardial infarction, congestive heart failure, and pneumonia, all as identified by the Centers for Medicare & Medicaid Services (CMS)-recommended ICD-9-CM codes (22).…”

“…In the final models, we additionally controlled for center and presepsis characteristics, factors present prior to the index hospitalization that are known to be, or potentially to be, associated with readmission risk (13,17,(21)(22)(23)(24)(25). We examined the following presepsis characteristics: age, sex, race, marital status, insurance status, comorbid conditions (using the Charlson Comorbidity Index score (25) and a history of malignancy specifically), and number of admissions over the prior year.…”

“…Last, we used mixedeffects multivariable logistic regression models (with random intercepts for patients) to examine risk factors associated with 30-day readmission after sepsis hospitalizations. In addition to variables included in readmission prediction scores (13,17,(21)(22)(23)(24), we examined four hospitalization factors: acute dialysis, acute neurologic dysfunction, the use of mechanical ventilation, and presence of shock. We identified these conditions using validated claims-based approaches and included them in models based on the hypothesis that these factors, which reflect illness severity and have been related to long-term cognitive and/or physical impairment after critical illness (10,(27)(28)(29), would be associated with readmission.…”

Real‐time automated clinical deterioration alerts predict thirty‐day hospital readmission

“…Their electronic model demonstrated good discrimination for 30 day mortality and readmission and performed as well, or better than, models developed by the Center for Medicaid and Medicare Services and the Acute Decompensated Heart Failure Registry. Similarly, Baillie et al developed an automated prediction model that was effectively integrated into an existing EMR and identified patients on admission who were at risk for readmission within 30 days of discharge (14). Our automated CDA differs from these previous risk predictors by surveying patients throughout their hospital stay as opposed to identifying risk for readmission at a single time point.…”