Currently Viewing:
The American Journal of Accountable Care June 2014
Patient-Centered Physician Selection: A Necessary First Step for Accountable Care
Brian Powers, BA and Sachin H. Jain MD, MBA
Patient-Centered Medical Home Recognized Practices Provide a Strong Front-Line for Accountable Care Organizations
Paige Cooke, assistant director, customer engagement, National Commitee for Quality Assurance
Patient-Centered Outcome Assessment May Lead to Different Conclusions and Different Treatment Decisions
Robert M. Kaplan, PhD, chief science officer, Agency for Healthcare Research and Quality
Connected Care Is Key to Accountable Care: The Case for Supporting Telehealth in ACOs
Krista Drobac, executive director, Alliance for Connected Care, and Clif Gaus, president and CEO, National Association of ACOs
How We Did It: How One Physician-Owned ACO Earned Shared Savings
Kelly Conroy, executive director, Palm Beach Accountable Care Organization, CEO Triple Aim Advisory Group
Study: Health Recovery Solutions Uses Tablets to Reduce Cardiac Readmissions
Judith Kutzleb, DNP; and Joan Shea, MBA, JD
Currently Reading
A Transitional Care Model for Patients With Acute Coronary Syndrome
Sherry Bumpus*, PhD, FNP-BC; Barbara L. Brush*, PhD, ANP- BC, FAAN; Susan J. Pressler*, PhD, RN, FAAN; Jack Wheeler, PhD; Kim A. Eagle*, MD; and Melvyn Rubenfire*, MD *These authors contributed equall
AJMC's ACO and Emerging Healthcare Delivery Coalition: First Live Meeting Coverage
Katie Sullivan

A Transitional Care Model for Patients With Acute Coronary Syndrome

Sherry Bumpus*, PhD, FNP-BC; Barbara L. Brush*, PhD, ANP- BC, FAAN; Susan J. Pressler*, PhD, RN, FAAN; Jack Wheeler, PhD; Kim A. Eagle*, MD; and Melvyn Rubenfire*, MD *These authors contributed equall
This study suggests that the Bridging the Discharge Gap Effectively (BRIDGE) program can help decrease the number of hospital readmissions in patients with acute coronary syndrome that cause unnecessary and substantial healthcare systems costs.

IMPORTANCE: Hospital readmissions place a substantial burden on the US healthcare system, especially those related to cardiovascular disease. Patients with acute coronary syndrome (ACS) constitute many of these readmissions, yet little is known about reducing such events.

OBJECTIVES: This study aimed to measure the effectiveness of the Bridging the Discharge Gap Effectively (BRIDGE) program in reducing readmissions and improving medication persistence in patients with ACS.

DESIGN: This was a retrospective outcomes study that used consecutive data for all patients referred to the BRIDGE program between March 2008 and March 2009.

SETTING: Ambulatory visit post-discharge from an academic hospital.

PARTICIPANTS: A total of 424 patients were referred to the BRIDGE program. ACS comprised 25.2% (n = 107) of the diagnoses. Patients were excluded if they died or were rehospitalized prior to their scheduled BRIDGE appointment (n = 9, 8.4%).The final study sample included 80 patients after excluding patients with missing data (n = 18, 18.4%).

INTERVENTION: BRIDGE is a nurse practitioner–led model for providing transitional care to cardiac patients.The intervention consists of a 1-time visit within 14 days of discharge. Patients received AHA/ACCF guideline-based care, examinations, education, follow-up, and referrals.

MAIN OUTCOME MEASURES: Prior to data collection, it was hypothesized that patients who attended BRIDGE would have lower 30-day readmission rates and superior 6-month medication persistence rates over nonparticipants (ie, follow-up only with primary care providers or cardiologists).

RESULTS: Of 80 patients, 77.5% (n = 62) attended. Their mean age was 62.5 years, 58.7% were female, and 86.3% were white. Patients who attended BRIDGE had lower rates of readmission at 30 (9.7% vs 27.8%, P = .112), 60 (11.3% vs 38.9%, P = .012), 90 (16.1% vs 38.9%, P = .052), and 180 (27.4% vs 50.0%, P = .072) days postdischarge compared with nonattenders.There were no significant differences between groups in medication persistence 6 months after discharge.

CONCLUSIONS: Patients who attended their BRIDGE appointment had fewer readmissions than patients who received usual care. Further, these reductions were not explained by better medication persistence. Models such as this should be developed and analyzed across institutions and patient types to ensure patient safety at home after hospital discharge and reduce excessive and unnecessary health-system costs.
It was estimated that 15.4 million Americans suffered from coronary heart disease between 2007 and 2010, and nearly 1 635,000 more would have a new coronary event. Of the latter, 419,100 will likely survive and approximately 67% of those (280,000) will suffer a recurrent event warranting rehospitalization (not including those who will have a silent event).1 Despite these projections, little is known about the circumstances culminating in a hospital readmission for patients with acute coronary syndrome (ACS). Possible contributors, however, have been speculated to include premature discharge, lack of prompt access to cardiology follow-up after hospital discharge,2 insufficient discharge education, lack of patient understanding of education provided, and poor patient and provider adherence to American Heart Association/American College of Cardiology Foundation (AHA/ACCF) guidelines.3 Whatever the cause of these hospital readmissions, especially those deemed avoidable,4 they place a substantial burden on an already stressed US healthcare system.5 To address these high rates of readmission, many clinical scientists and care providers are looking to new models of care that focus specifically on the transition period between hospital discharge and home.

Transitional care begins prior to hospital discharge and terminates once an outpatient care team has seen the patient and assumed care responsibility. Most transitional care models are designed around direct patient assessment, diagnosis, treatment, and education during this interim phase.6-9 This is achieved by facilitating communication among providers, improving discharge education and medication management, resolving outstanding diagnostics, and instructing patients when to seek care.10

Healthcare providers have long recognized the consequences of poor patient follow-up, but lack of provider accountability between discharge and ambulatory care follow-up has allowed a gap in care transitions to go largely unchecked.11 As an example, Jencks and colleagues5 reported that 50.2% of Medicare patients rehospitalized within 30 days of discharge had no record of being seen by a healthcare provider postdischarge. One consequence of such oversight is the rising number of potentially preventable hospital readmissions.12 Initially, the aim of reporting hospital readmission rates was to draw attention to the dilemma and allow patients to make informed choices about where to seek care.13 Later, readmission rates became part of pay-for-performance programs.14 Now, as a result of the Patient Protection and Affordable Care Act (ACA), hospitals with above-average readmission rates will be penalized with a reduction in the percentage of their Medicare reimbursement. Initial fines were imposed in 2013 and levied against 2012 readmission rates.

We developed the Bridging the Discharge Gap Effectively (BRIDGE) program to facilitate timely postdischarge care for patients discharged with a cardiac diagnosis. Operating since 2008, BRIDGE provides a 1-time ambulatory transitional care visit within 14 days of discharge, and stresses the importance of developing trusting relationships between patients and providers.15 The BRIDGE clinic is staffed by 5 specialty-certified cardiovascular nurse practitioners (NPs) who function in colaboration with the discharging cardiologist. The goal of the NPs is to eliminate many of the aforementioned contributors to hospital readmissions by conducting thorough examinations, reviewing diagnostic tests postdischarge, evaluating response to treatment, performing medication reconciliation, making therapeutic adjustments when necessary, and ensuring that appropriate follow-up and referrals are scheduled. The NPs tailor education about the individual’s event, condition, disease process, and signs and symptoms that should trigger a call to a physician or an emergency department visit. The BRIDGE visit functions as an extension of the hospital discharge team, differing from usual care in 3 distinct ways: (1) the BRIDGE clinic ensures that the time between a patient’s hospital discharge to their first outpatient follow-up is no longer than 14 days; (2) the visit with a NP, while providing traditional evaluation and treatment, emphasizes education and support; and (3) patients are seen by NPs who are integrated within the health system and better able to facilitate and coordinate care within that system compared with those external to the system.

This study aimed to measure the effectiveness of the BRIDGE program by comparing the 30-day to 180-day readmission rates for ACS patients who attended BRIDGE with those who chose not to attend in lieu of usual care. Additionally, it aimed to determine whether medication persistence contributed to the readmission rates. It was hypothesized that patients who attended BRIDGE would have lower 30-day readmission rates and superior 6-month medication persistence rates over nonparticipants.


This was a retrospective study using consecutive data (extracted from an electronic medical record) for all patients referred to the BRIDGE program in a deidentified clinical database. The Human Subjects Internal Review Board of the University of Michigan Medical Center approved this study (HUM00035421).

All patients discharged with a diagnosis of ACS (acute myocardial infarction or unstable angina) from the inpatient adult cardiology service between March 30, 2008, and March 30, 2009, were eligible for this study. Referrals were made to the BRIDGE program based on the lack of availability of a cardiology or primary care follow-up appointment within 14 days of discharge. Patients included in the analysis were divided into 2 cohorts: those who were referred and attended (“attenders”) and those who were referred and did not attend (“nonattenders”). Patients were excluded from the study if they became pregnant, sought follow-up outside the institution, or died within 30 days of discharge. The Social Security Death Index was queried for patients lost to follow-up; only the result of this query was recorded.

The study cohort (Figure 1) included 424 patients referred to the BRIDGE program. ACS comprised 25.2% (n = 107) of the diagnoses referred. Patients were further excluded if they died or were rehospitalized prior to their scheduled BRIDGE appointment (n = 9; 8.4%). The final study sample included 80 patients after excluding patients with missing variables (n = 18; 18.4%).


Two principal instruments were used to describe patients’ overall state of health. The first measure, the Charlson Comorbidity Index (CCI), quantified the combined effects of 19 comorbid conditions on mortality.16 Tested across a variety of populations, the CCI has consistently proved valid for predicting mortality.16-18 The second measure, the GRACE Risk Model for Discharge, predicted the risk of death in patients with ACS from hospital discharge to 6 months out19 with excellent discriminatory and predictive validity (C statistic, 0.70-0.80).20,21

Data Analysis

Data were analyzed using SPSS 18.0 (IBM, Armonk, New York). All variables were assessed for compliance with statistical assumptions. Missing data were excluded from the sample.

Baseline equivalencies were assessed for all demographic variables, comorbidities, CCI scores, and GRACE risk scores. Associations between variables were assessed with Pearson’s correlation coefficient. Independent student t tests and χ2 were used to compare differences between the BRIDGE intervention group and the usual care group. Pearson’s χ2 test for significance was reported except in cases where the expected count would violate an underlying assumption; in those cases, Fisher’s exact test was reported. Hierarchical logistic regression models were used to evaluate whether BRIDGE attendance influenced readmissions and/or medication persistence, after adjusting for comorbidities (CCI) and severity of events (GRACE). ORs and 95% CIs were reported for all independent variables. The significance level was set at .05 for all analyses.

Readmission rate. The hospital readmission rate was calculated as the number of patients discharged from the hospital with a diagnosis of an ACS event and readmitted within 30, 60, 90, or 180 days, divided by the total number of people who were discharged alive with the same diagnosis.5 The BRIDGE and non-BRIDGE– specific rates of readmission were calculated as the total number of readmissions for the BRIDGE and non-BRIDGE groups divided by the total number of subjects in each group. Only the first readmission following discharge for an ACS event was counted.

Medication persistence. Medication persistence was defined as self-reported continued use of prescribed AHA/ACCF-recommended pharmacotherapy regimens of β-blockers, angiotensin-converting enzyme (ACE) inhibitors (or angiotensin II receptor blockers [ARBs]), aspirin, statins, or clopidogrel at 6 months, postdischarge. Six-month persistence rates were determined by comparing self-reported medication regimens from the most complete cardiology or primary care follow-up notes at 6 months, postdischarge (no more than 2 weeks before and no more than 1 month beyond the 6-month discharge date) to the medication regimen prescribed at discharge. Six-month medication persistence rates for each drug and for the 4 (β-blockers, ACE inhibitors [or ARBs], aspirin, statins) and 5 (β-blockers, ACE inhibitors [or ARBs], aspirin, statins, plus clopidogrel) drug combination regimens were compared between BRIDGE attendees and nonattendees. This methodology is consistent with the methodology employed by the GRACE Registry.22 Patients with known contraindications or hypersensitivity reactions to certain medications, such that they could not be prescribed or continued, were excluded.


Of 107 ACS patients discharged from the study hospital and referred to the BRIDGE program, 80 were included in the final readmission analysis. To isolate the BRIDGE effect, patients who died or were readmitted prior to their scheduled appointment were excluded, as were patients with missing data (n = 27; Figure 1). Excluded patients, compared with patients remaining in the study, were noted to have higher CCI and GRACE risk scores, suggesting that these patients either had more comorbid conditions, worse events, or both.

The mean age of the final sample was 62.4 years. The majority (58.8%) were female, and white (86.3%), with a median length of initial hospital stay ranging from 2 to 4 days. Table 1 shows a comparison of patients who attended BRIDGE versus non-attenders. With the exception of a higher percentage of dyslipidemia among BRIDGE users, there were no significant differences between the 2 groups.


Copyright AJMC 2006-2017 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
Welcome the the new and improved, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up

Sign In

Not a member? Sign up now!