Improving Value Through Heart Failure Care Coordination: The Allina Health Experience

Article Citation:

Steven M. Bradley, Pam Rush, Kim Wolf, Amin Rahmatullah, Robin Braun, Michael Samara, Alan J. Bank, Steven Bergeson, Wendy Gunderson, Craig E. Strauss, Shaina Witt, Melissa Hutchinson, TC Tong, Denise Mueller, Peter Eckman, and Spencer H. Kubo (2018) Improving Value Through Heart Failure Care Coordination: The Allina Health Experience. Journal of the Minneapolis Heart Institute Foundation: Fall/Winter 2018, Vol. 2, No. 2, pp. 9-13.

Original Article

Steven M. Bradley, MD, MPH

Pam Rush, RN, MS,

Kim Wolf, MBA, BSN

Amin Rahmatullah, MD

Robin Braun, RN

Michael Samara, MD

Alan J. Bank, MD

Steven Bergeson, MD

Wendy Gunderson, APRN, CNS

Craig E. Strauss, MD, MPH

Shaina Witt, MA

Melissa Hutchinson, MPH

TC Tong, BA, BS

Denise Mueller, MA

Peter Eckman, MD

Spencer H. Kubo, MD

Minneapolis Heart Institute, Minneapolis, MN; Minneapolis Heart Institute, Center for Healthcare Delivery Innovation, Minneapolis, MN; Allina Health, Minneapolis, MN; Essentia Health West Region, Fargo, ND (work completed while at Allina Health, Minneapolis, MN); Metropolitan Heart and Vascular Institute, Coon Rapids, MN; United Heart and Vascular Clinic, Saint Paul, MN; Cope Health Solutions, New York, NY (work completed while at Allina Health, Minneapolis, MN).

Disclosures: None.

Address for Correspondence: Steven M. Bradley, MD, MPH, Minneapolis Heart Institute, 920 East 28th Street, Suite 300, Minneapolis, MN 55047, E-mail:


Reducing readmission following heart failure is emerging as a target of quality initiatives in the Centers for Medicare & Medicaid Services Hospital Readmission Reduction Program. In response to higher readmission rates than desired following heart failure (HF) hospitalization, Allina Health hospital’s cardiovascular subspecialty care developed an HF nurse care coordinator program to reduce readmissions. The nurse HF care coordinator serves as the bridge to help manage care following hospital discharge and ensure adherence to protocols developed by the HF management program. This effort was initially developed and implemented at Mercy and Unity Hospital and was associated with a 4.3% reduction in HF readmissions. Subsequent expansion of the HF nurse care coordinator program to United and Abbott Northwestern Hospitals was associated similar reductions in HF readmissions. Concurrently, all-cause mortality at 6 months post-discharge was also significantly lower following implementation of the program (mortality pre-HF care coordinator program 12.6% vs. post-HF care coordinator program 18.8%, P = .047) in propensity matched analysis. The findings of this effort suggest the potential for care coordination programs to improve the care and outcomes of patients with HF.

Keywords: heart failure, readmissions, quality of care, outcomes


Goals and Vision of the Program

Readmission following hospitalization for heart failure (HF) is a common, high-cost, and potentially avoidable adverse event. Accordingly, the Centers for Medicare & Medicaid Services (CMS) Hospital Readmission Reduction Program has sought to incentivize efforts to minimize readmissions through reduced payments to hospitals that have excess readmissions. Specific strategies to improve readmission, including early follow-up after discharge and increased patient self-efficacy, have shown promise within programs such as the American College of Cardiology Hospital to Home (H2H) initiative.1 However, these strategies have not fully addressed persisting challenges to post-hospital HF care.

Allina Health is a not-for-profit health care system in Minnesota and western Wisconsin with over 5,000 affiliated and employed physicians, 13 hospitals, over 90 clinics, and numerous specialty medical services. Although Allina has a history of providing high quality episodic care, readmission rates following HF hospitalization were higher than desired. This brief report summarizes the development, implementation, evaluation, and dissemination of a HF nurse care coordinator program to reduce readmission rates following HF hospitalization toward achieving better patient health and satisfaction. Concurrently, we assessed for potential negative impacts of the program by tracking patient mortality.Design of the Initiative and Challenges in Implementation

Soon after CMS announced a focus on HF readmissions, cardiologists at Allina’s Mercy and Unity Hospitals (Coon Rapids and Fridley, MN, respectively) launched an effort to reduce 30-day HF readmissions. These providers envisioned a HF management program to overcome persistent challenges with care coordination. The resultant program was modeled after a prior successful randomized controlled trial of transitional care coordination.2

Central to this effort was the creation of a nurse HF care coordinator role. The nurse HF care coordinator was tasked with seeing patients with HF during their hospitalization, understanding their care plans, and ensuring that the care plan is followed after discharge. In this way, the care coordinator served as the bridge to help manage care following hospital discharge and ensure adherence to protocols developed by the HF management program. Furthermore, the HF nurse care coordinator helped facilitate earlier outpatient follow-up among patients with evidence of recurrent signs and symptoms of HF. This approach was based on the premise that the typical HF patient is complex, with multiple comorbid conditions and multiple providers, such that standard instructions to outpatient providers could not capture important nuances relevant for a successful transition from hospital to home and facilitate communication across multiple providers. For example, many HF patients require continued diuresis at higher than maintenance doses upon discharge from the hospital. The transition back to maintenance diuretic dosing requires close clinical surveillance and judgment that is challenging to navigate and communicate across providers with written discharge instructions alone.

Although coordinated care is conceptually appealing, the creation of a nurse HF coordinator role represented a significant investment of resources without clear evidence that the approach would be successful. In fact, randomized studies of care coordination for patients with multimorbid disease have largely failed in the absence of substantial in-person contact.3 As such, no other Allina hospitals joined Mercy and Unity in the initial implementation of the nurse HF care coordinator program. Subsequently, an evaluation of the HF care coordinator model was undertaken to both evaluate the effectiveness of the program and, if deemed successful, to inform how the intervention could be successfully implemented at other Allina hospitals.Implementation and Success of the Initiative

The HF care coordinator model was implemented at Mercy and Unity Hospital in January of 2010. The HF care coordinators were responsible for screening admission diagnosis codes within the electronic medical record to identify patients admitted with a primary diagnosis of congestive HF. From this list of patients, HF care coordinators then completed brief chart reviews to ensure accuracy of the diagnostic code before engaging patients directly.

Patients with a confirmed diagnosis of HF were then seen and evaluated by the HF care coordinators in the hospital. In these meetings, the HF care coordinator provided HF education, assessed self-management skills, developed patient-centered goals of care, coordinated optimal medication therapy, scheduled follow-up appointments, coordinated visits across specialties, and identified and began to address discharge concerns. Following discharge, the HF care coordinators called patients within 24 to 48 hours of discharge and at least weekly in the month post-discharge. During these follow-up calls, the HF care coordinator reviewed medication and diet adherence, scheduled appointments, addressed patient concerns and goals, and coordinated communication between patients and providers. Whenever possible, care coordinators would also attend follow-up cardiology visits with patients. HF care coordinators also sought to identify progressive or urgent symptoms and coordinate appropriate clinical action. For patients who remained clinically stable, the HF care coordinators worked toward a transition to self-management.

Following implementation of the program at Mercy and Unity hospitals, readmission rates were observed to decrease by 4.3% (22.9% in the 2 years prior implementation compared with 18.6% in the 2 years after implementation). Based on this success, cardiologists at United Hospital sought to replicate the program. The United program was launched in June 2012 with an associated 5.1% reduction in readmission rates (19.8% in the 18 months prior to implementation down to 14.7% in the 18 months after implementation). This suggested that the structure and methods of the original Mercy and Unity program were scalable and transferrable.

The next step was the creation of an Allina wide team to coordinate and oversee a broader implementation of the HF readmission program, led by a program director. The program focused on five major functional areas: nursing, care coordination, protocols and guidelines, measurement and reporting, and education. Each functional area was led by a cardiologist, with care coordination co-led by a cardiologist and a primary care physician. The overriding goals of the program were to develop processes that supported a consistent approach in the care and discharge of HF patients, integrate appropriate services such as palliative and rehab care with the care process, and provide coordinated care across the continuum.

Additionally, resources were provided to hire HF nurse care coordinators who provided this type of specialty care at specific hospitals. In January 2013, these nurse coordinators implemented the program at Abbott-Northwestern Hospital, the largest cardiovascular hospital in the Allina system. Following implementation at Abbott Northwestern, readmission was observed to decrease by 3.9% (19.3% in the 21 months prior to implementation down to 15.4% in the 15 months after implementation).

To further evaluate the effectiveness of the HF care coordination program, Allina Health’s Economics & Program Evaluation Team undertook a formal assessment of the impact of the HF care coordination program on healthcare value following implementation at these sites. A logic model was created to represent the initiative’s inputs/resources, activities, outputs, and expected outcomes (Figure 1). Informed by this logic model and data from the Allina Enterprise Data Warehouse, the outcomes and patient experience of HF patients following implementation of the program were compared to those of HF patients hospitalized prior to implementation of the HF care coordination program. To facilitate a more meaningful comparison, patients were matched on the presence or absence of common clinical conditions, distance from home to the hospital, and insurance coverage (Table 1). Aspects of temporal trends in care delivery that could contribute to differences in HF patient experience and outcomes over time were acknowledged as limitations, but not formally addressed in the evaluation. The results of this evaluation suggested a significant reduction in rates of 30-day all-cause readmission after HF discharge (12.3% vs 21.7%, P = .003) (Table 2). In this matched analysis, evidence of the impact of care coordination was also apparent in the higher rate of follow-up within 5 days of discharge (50.5% vs. 20.2%, P< .001) and cardiology follow-up within 30 days of discharge (44.8% vs. 15.9%, P < .001). At six months after discharge, readmission rates remained lower (36.5% vs. 45.5%, P = .031) and all-cause mortality was also significantly lower (12.6% vs. 18.8%, P = .047) following implementation of HF care coordination. Patient experience scores were not significantly different in sampled comparisons. Personnel costs of the heart failure care coordinator program in the post-implementation period of evaluation was estimated at $200,300. During this period, HF care coordinators were associated with an estimated 34 prevented readmissions. Furthermore, from an Allina Health perspective, the HF care coordination program was associated with an estimated savings of more than $365,000 annually due to greater care coordination and efficiency (eg, reduction in missed appointments).

Heart failure care coordination logic model.

Creation of comparison cohort for evaluation of heart failure care coordination program.

Outcomes following implementation of heart failure care coordination.

Despite the success of the HF care coordination program at cardiovascular hubs of care in the Allina system, we observed several barriers to extending the HF care coordinator approach to other Allina hospitals. One barrier related to inadequate HF hospitalization volume to support on-site nurse care coordinators. To address this barrier, strategies were developed to expand the reach of HF nurse care coordinators to other hospitals. For example, patients admitted for HF at St. Francis Hospital (Shakopee, MN) were referred to Abbott Northwestern HF nurse care coordinators for telephonic follow-up. A care coordinator at St. Francis introduced the HF care coordinator program to the patient and gave a warm hand-off to the Abbott Northwestern team. The Abbott Northwestern HF care coordinator who provided this telephonic coverage would occasionally visit St. Francis to establish relationships with providers and care coordinators and to encourage use of the program. Furthermore, the cardiology group that provides services at Abbott Northwestern Hospital also provides inpatient cardiology services to St. Francis, thus increasing visibility of the program.

An additional challenge in the development and expansion of the HF nurse care coordinator program related to difficulties in developing clinical expertise. Due to close connections with the cardiology groups at Mercy and Unity, United, and Abbott Northwestern Hospitals, the HF nurse care coordinators at these sites were able to rapidly develop expertise necessary for independent work that was additive to the patient’s care. Training HF nurse care coordinators at other locations was challenged by the lack of on-site clinical support and guidance from cardiology providers.Summary of the Experience, Future Directions and Challenges

The involvement of care coordinators to assist HF patients in the hospital to home transition has been successful in addressing HF readmissions in Allina Health. Furthermore, coordinated care was associated with cost savings for the health system and, as such, was a financially viable solution to improve patient outcomes.

Despite the success of the program, we experienced challenges that limited dissemination beyond our hubs of cardiovascular care. This including the need for adequate heart failure admission volume to support on-site investment in the program and development of clinical expertise. The use of services to improve specialty care access in remote access—such as video teleconferencing during hospitalization for initial introduction, education, and care planning with telephone follow-up—may serve as a future modality to spread the impact of the HF care coordinator program. It is also important to acknowledge that the impact of the HF care coordinator program cannot be separated from the importance of early follow-up. Our anecdotal experience suggests HF care coordinators are best as a supplement, rather than a surrogate, to close in-person clinical follow-up. Accordingly, a concurrent effort has focused on ensuring early follow-up (within 5 days of discharge) with the patient’s primary care provider or cardiology provider to assess volume status, symptoms, medications, and dietary adherence. A future challenge to the HF care coordinator program relates to Allina Health’s increasing investment in telemonitoring (ie, remote scales for weight/volume monitoring) under the direction of Advanced Care Teams in primary care. As these efforts move forward, it will be important to coordinate efforts across teams to optimize patient care management and patient outcomes. Finally, concerns have been raised about a potential inverse relationship between readmission rates and patient mortality.4,5 Mortality following HF hospitalization in Allina Health has improved significantly as efforts at care coordination have progressed, highlighting the potential benefit of this low-tech but human resource-intensive effort. Moving forward, it will remain important to monitor for unintended adverse consequences of efforts to promote outpatient care whenever possible.


1.American College of Cardiology. Hospital to home. Available from: Accessed October 3, 2017. [Google Scholar]
2.Naylor MD, Brooten DA, Campbell RL, Maislin G, McCauley KM, Schwartz JS. Transitional care of older adults hospitalized with heart failure: a randomized, controlled trial. J Am Geriatr Soc. 2004;52:675–684. [Crossref] [Google Scholar]
3.Peikes D, Chen A, Schore J, Brown R. Effects of care coordination on hospitalization, quality of care, and health care expenditures among Medicare beneficiaries: 15 randomized trials. JAMA. 2009;301:603–618. [Crossref] [Google Scholar]
4.Dharmarajan K, Wang Y, Lin Z, et al. Association of changing hospital readmission rates with mortality rates after hospital discharge. JAMA. 2017;318:270–278. [Crossref] [Google Scholar]
5.Gupta A, Allen LA, Bhatt DL, et al. Association of the hospital readmissions reduction program implementation with readmission and mortality outcomes in heart failure. JAMA Cardiol. 2018Fall/Winter 2018;3:44–53. [Crossref] [Google Scholar]
Print ISSN: