• Many organizations in healthcare are moving rapidly to adopt value-based reimbursement (VBR) models that reward value instead of vol­ ume. By shifting a significant percentage of clinical and financial risk from payers to providers, VBR programs can help reduce costs significantly, improve the quality of care, and increase efficiency. The experience of Banner Health Network (BHN) with VBR in commercial and government plans indicates that a large health system can transition to VBR with renewed focus on quality, cost efficiencies, population health management tactics, and member engagement.

In December 2011, BHN was selected in a competitive bid process as one of 32 organizations to participate in a Centers for Medicare & Medicaid Ser­ vices (CMS) value-based demonstration initiative called the Pioneer account­ able care organization (ACO). As a Pioneer ACO with consistent and positive results, BHN, based in Phoenix, Arizona, has demonstrated that VBR can lead to results that are beneficial to the member, the healthcare organization, and the community. BHN has been a top performer in returning shared savings to Medicare while improving appropriate service utilization and performance on quality metrics.

The development of simultaneous commercial ACO products was a re­ quirement of the Pioneer agreement with CMS, as well as a clear goal for BHN. The initial period of ACO partnership formation revealed an uncomfortable re­ ality: Payers and providers would need to collaborate and share information as never before. Further, many payers were uncertain about working closely with providers who were working simultaneously with other payers on similar part­ nerships. Before long, however, there were enough successful VBR arrange­ ments to allay these initial payer anxieties. Today, BHN has high-value network arrangements with nearly every major payer in Arizona.

Becky Kuhn, FACHE, is executive vice president for com m un ity delivery at Banner Health
in Phoenix, Arizona. Chuck Lehn is executive vice president o f strategic growth at Banner Health.
m

After moving steadily toward value- based reimbursement (VBR) for more than a decade, BHN has amassed a mem­ bership of about 350,000 covered lives. Moreover, 92 percent of BHN’s member­ ship is insured through a risk-bearing relationship.

BHN’s strategies and tools for VBR range from clinical to technical. Engaged provider partners have been an essential component without whom success would not be possible. Banner Health care man­ agement teams working with our provid­
ers are focused on the development and imple­ mentation of evidence- based clinical practices across the enterprise to make sure that members receive the most expedi­ ent and efficient care with the fewest complications.

Facility-based and community-based nurse case management allows members to re­ ceive the guidance, education, and clinical support they need to achieve the highest possible level of personal wellness and avoid unnecessary hospital admissions. Enabling technologies, such as telehealth, are showing great potential in creating virtual connections between patients and their caregivers. Claims data and an inter­ nal enterprise data warehouse drive popu­ lation health tactics and feed electronic health records (EHRs) to enable provi­ sion of the best possible care. Finally, a new area of focus for BHN is a post-acute strategy to ensure that the continuum of care does not end abruptly when members transition to post-acute care providers in the community.

Although the move to VBR requires a commitment to meeting cost and qual­ ity benchmarks, integration of care at
Enabling technologies,
such as telehealth, are
showing great p o ten tia l in
creating v irtua l connections
between patients and the ir
caregivers.

all levels, and a significant information technology (IT) infrastructure, BHN’s ex­ perience demonstrates a clear potential for organizations that successfully navigate the gap between a fee-for-service (FFS) world and VBR.
I n t r o d u c t i o n After several years of slow but steady adop­ tion, VBR models appear to be on their way to replacing FFS as the predominant reimbursement system in US healthcare. McKesson (2014) estimates that VBR will overtake FFS by 2020, with 90 percent of payers and 81 percent of providers already using some mix of both. In the first score- card on payment reform released in March 2013 by the not-for-profit Catalyst for Payment Reform (2013), only 11 percent of payments to providers were not under an FFS model. One year later, the second report card (Catalyst for Payment Reform and Health Care Incentives Improvement Institute 2014) reported that 40 percent of commercial health plan payments were made through methods designed to improve quality and reduce waste (see Exhibit 1 for details about waste in the U S healthcare system).

The U S Department of Health & Human Services (HHS20i5)is encourag­ ing the adoption of VBR by mandating that 30 percent of Medicare outlays by 2016 and half by 2018 be routed through alternative payment models such as ACOs and bundled payments. HHS (2015) also set a goal of tying 85 percent of all tradi­ tional Medicare payments to quality or value by 2016 and 90 percent by 2018 via programs such as the Hospital Value-Based Purchasing and the Hospital Readmis­ sions Reduction Programs. This is the first time in the history of the Medicare program that HHS has set explicit goals

18 • F R O N T I E R S O F H E A L T H S E R V I C E S M A N A G E M E N T 3 2 1 2

Source: Health Affairs (2012). “ Health Policy Briefs: Reducing Waste in Health Care.’’
for alternative payment models and value- based payments.
The Affordable Care Act created a num­ ber of new quality-based payment models, including ACOs, patient-centered medical homes, and new methods of bundling pay­ ments for episodes of care. In each pay­ ment model, healthcare providers are held accountable for the quality and cost of the care they deliver to patients, giving them a financial incentive to better coordinate care for their patients and avoid unneces­ sary expenses. Pay for performance (P4P) is emerging as the most commonly used model. Also known as value-based pur­ chasing, P4P measures quality by moni­ toring process outcomes, such as lowering blood pressure and counseling patients to stop smoking. In a 2014 McKesson report, payers using a mix of reimbursement
models reported that the proportion of their business aligned with P4P would increase from 10 percent to 18 percent in five years. Providers using a mix of reim­ bursement models projected that their alignment with P4P would increase at an even greater rate from 9 percent to 21 percent in five years.
Adopting a VBR model requires or­ ganizations to take into account clinical markers and best practices to ultimately produce better population-based out­ comes. Shifting focus to these long-range goals can be challenging and will require investment in contemporary healthcare IT and system-wide improvements in integra­ tion, such as a commitment to a common EHR platform that allows the dynamic exchange of patient data. BHN has moved toward VBR using several approaches.
T l
m
> H
73
m
B e c k y K u h n , FACHE, a n d C h u c k L e h n • 19

Ba n n e r ’s Jo u r n e y to VBR Parent company Banner Health operates 29 acute, specialty, and critical-access hospitals in Alaska, Arizona, Califor­ nia, Colorado, Nebraska, Nevada, and Wyoming. In 2014, the system had $5.4 billion in revenue. Banner Health’s senior leadership strongly believes that VBR, supported by population health manage­ ment, has the potential to create greater sustainability in our nation’s healthcare system. To that end, we have developed a strategic plan and supporting infrastruc­

ture in accordance with that belief. Banner Health has the infrastructure for and 20 years of experience managing its own self- funded health plan in all the states in which we have operations. Banner Health
was recently named the largest private employer in Arizona (Wiles 2015), so the opportunity to learn from our employee health plan members and their depen­ dents has been a significant benefit. We can apply new learning and capabilities to other types of arrangements and demon­ strate how these elements have played out in our own plans.

In addition, the health system histori­ cally used a model that included taking on risk from several Medicare Advantage plans, but these plans and the self-funded health plan combined made up less than 5 percent of Banner’s business. To further prepare for VBR, Banner Health moved to­ ward a case-rate model of reimbursement (i.e., fixed payment for a defined episode of care), similar to diagnosis-related group reimbursement, by taking on more of the risk, especially with inpatient care. This model has provided incentives for deliv­ ering the right care in the most efficient way, preparing the patient for discharge,
In 2012, B H N began
entering in to contracts th a t
carried significantly more
fin a n c ia l risk than those it
had entered into previously.
managing resources and length of stay, and supporting care that would prevent avoidable readmission.
BHN was incorporated in July 2011, designed specifically to be the vehicle for a collaborative model between Ban­ ner Health, independent physicians in the community, and physicians in the employed model used at Banner Medical Group. The network supports the assump­ tion of much more risk, making possible Banner’s participation as one of the origi­ nal 32 Pioneer ACOs in a demonstration program announced by the Centers for Medicare & Medicaid Services (CMS) in December 2011. Our integrated delivery network is composed of the Banner Health hospitals and ambulatory facilities, Banner Medical Group (employed providers), and two affiliated partners—Banner Physician Hospital Organization (BPHO) and Ari­ zona Integrated Physicians (AIP). These entities joined together with the goal of improving quality and the care experience while reducing costs for employers and the members we serve.
In 2012, BHN began entering into contracts that carried significantly more financial risk than those it had entered into previously, especially relative to expecta­ tions around advancing quality measures, such as reduced avoidable admissions and readmissions and improved diabetic indicators. Today, nearly all of BHN’s book of business involves some degree of risk. To date, performance on various contracts has been uneven. Attributed populations in which member engagement and network utilization do not get traction can be chal­ lenging. Medicare Advantage and Market­ place plans are highly price sensitive. Yet, together, they have yielded a neutral effect financially, even after we consider the initial investments associated with the launch of a new division of the organization—BHN.
20 • F R O N T I E R S O F H E A L T H S E R V I C E S M A N A G E M E N T 3 2 : 2

P i o n e e r ACO a n d C o m m e r c i a l ACO Pr o d u c t s In performance year i, more than 50,000 traditional Medicare beneficiaries were attributed to BHN. We became responsible for managing the health of these members on January 1, 2012. The Medicare Pioneer ACO is a value-based model designed to provide members with an improved patient care experience (including qual­ ity and satisfaction), improved health of populations, and a reduction in the cost of healthcare.
In the first year, BHN demonstrated an ability to reduce hospital admissions, hospital length of stay, and the need for hospital readmissions by supporting beneficiaries in transitions when they were most at risk and in need of care and advocacy (see Exhibit 2 for more detail on performance year 1 results). Deploying nurse case managers and social workers was an important first step in developing a structure to better support the medical and social needs of the chronically ill benefi­ ciaries with complex medical needs who represented the top 5 percent of the overall spend. BHN proved to be a top performer
EXHIBIT 2 Pioneer ACO First-Year Results
in terms of shared savings compared with the results for other Pioneer organizations that year, returning $19 million in savings shared with CMS.
L&M Policy Research (2013), an inde­ pendent research firm in Washington, DC, compiled a report for CMS titled Effect of Pioneer ACOs on Medicare Spending in the First Year. The report credited BHN’s focus on providing highly collaborative care— especially for chronically ill patients with complex medical needs—as a key to early success. Surrounding at-risk patients with supportive care, particularly during care transitions, was an important strategy that helped reduce hospital admissions and readmissions. Advanced IT was another important component of the ACO infra­ structure, giving providers more informa­ tion about the beneficiaries in their care.
In performance year 2 (2013), BHN’s quality score reflected a gain of nearly 19 percent from the previous year, the result of an educational focus, monitoring of metrics, and provider engagement. As a top performer in the Medicare Pioneer ACO program for the second consecutive year, BHN shared savings of more than
Reduction in overall hospital admissions
Reduction in average length o f stay
Reduction in avoidable hospital readmissions
Reduction in high-tech imaging services
1 • Reduction in CMS paid amount per beneficiary
^ Banner Health Network
m
> H
7> m
B e c k y K u h n , FACHE, a n d C h u c k L e h n • 21

B H N has been able
to develop strong and
productive relationships
w ith our payer partners.
$15 million. This was possible as a result of delivering the right care, at the right time, in the right setting through more coordinated care, as well as having IT tools and a large enterprise that was able to set a new direction in support of value-based care.
BHN is among a handful of organiza­ tions in the country that have established through early success that ACOs can improve the quality of care and the experi­ ence of care, as well as reduce Medicare costs. In the report released by CMS in late January 2014, BHN was one of two Pio­
neer ACOs that together achieved 41 percent of the total overall reduction in 2012 spending growth in this nationwide demon­ stration project.
BHN is one of the remaining 19 Pio­ neer ACOs. In performance year 3, BHN returned more than $29 million in savings shared with CMS, and its quality score in 2014 improved by almost 10 percent from the preceding year. To date, results from performance year 4 have not been released by the Center for Medicare & Medicaid Innovation, but BHN anticipates similar results.
Starting in 2012, BHN and Aetna, a diversified healthcare benefits company serving 36.4 million people, jointly of­ fered a shared-risk product, Aetna Whole Health, in Arizona. Lower prices incen- tivized members to stay in network, and BHN was able to generate savings through reduced hospital readmissions, expanded access to primary care physicians, and increased use of preventive screenings. In this way, both BHN and Aetna have a stake in the quality of the care, the cost, and the outcomes. After only the first full year of this relationship, BHN and Aetna
announced that their accountable care col­ laboration resulted in a shared savings of approximately $5 million on Aetna Whole Health fully insured commercial member­ ship in 2013 and a 5 percent decline in average medical cost for members.
In 2013, Aetna Whole Health plan members served by BHN attained the following:
• Improvements in cancer screening rates, including cervical and colorectal cancer screening
• Better blood sugar management in more members with diabetes
• Reduction in radiology services of approximately 9 percent
• Increase of almost 4 percent in the generic drug prescribing rate
• Reduction of about 9 percent in avoidable admissions
Aetna and BHN also saw savings and an improved medical cost trend among Aetna members in the ACO outside the Aetna Whole Health product. Nearly two-thirds of Aetna’s 47 commercial ACO agreements currently feature or expect to feature risk-sharing arrangements by 2015.
In September 2013, Banner Health finalized a joint venture with Blue Cross Blue Shield of Arizona (BCBSAZ) that included the joint ownership of a Medi­ care Advantage plan formerly managed by a Banner Health legacy organization. As a result of this arrangement, which combines the capabilities of a payer and a provider, the membership in BCBSAZ Advantage nearly doubled from 23,000 to 45,000 in a two-year period. The success of this relationship has led to two other plans in the commercial marketplace: BCBSAZ Alliance and BCBSAZ Acclaim.
2 2 • F R O N T I E R S O F H E A L T H S E R V I C E S M A N A G E M E N T 3 2 : 2

O v e r a l l Pa y e r Re l a t i o n s h i p s BHN has announced a variety of value- based relationships with payers, including UnitedHealthcare, Cigna, Humana, and Health Net. An important first step in these relationships was building trust and a true spirit of collaboration. Sharing data and working hand in hand from the sale of the product through the delivery of care was new territory for all parties.
Ultimately, BHN has been able to de­ velop strong and productive relationships with our payer partners. The organization now has high-value network or attributed network contracts with nearly every major payer in our market, as well as an agree­ ment with a Medicaid plan. BHN also is the provider network for a variety of prod­ ucts on the Health Insurance Marketplace in coordination with Aetna, BCBSAZ, Cigna, Health Net, and Meritus. We work with each of our valued partners to de­ velop the tools and services that will bring the highest-quality care, service, and value to members.
T o o l s t o S u p p o r t VBR BHN has used a wide range of VBR- related strategies and tools to improve quality, reduce costs, and achieve better population health management. Some of the most important factors in our progress are described here.
Care Management Continuous clinical improvement has become an integral part of the care deliv­ ered throughout Banner Health as a result of its care management function. In 2001, Banner created a dozen functional teams that focused on patient safety, utilization management, and clinical data analysis, as well as a few clinical improvement efforts aimed at CMS core measures.
The improvement process has evolved to include clinical consensus groups (CCGs) and Teams for Rapid Innovation and De­ sign (TRIaDs).
To date, more than 50 teams, including 22 CCGs and six TRIaDs, are dedicated to improving clinical quality. The CCGs are multidisciplinary groups of physicians and other clinicians who examine emerg­ ing issues, research, and improvement opportunities in one specific clinical area. Many of these members are community providers who participate on a voluntary basis. TRIaDs focus on improving patient outcomes across the organization, espe­ cially within BHN’s membership. Each of these teams has a dedicated medical direc­ tor, clinical director, and process engineer.
TRIaDs have focused accountable leadership for improving patient outcomes across the organization. The process engineer specializes in the development of metrics and design processes. These teams are now organized around health management for members and patients. The six TRIaDs focus on the following clinical areas:
• Member needs assessment • Prevention/minor episodic care • Chronic care • Complex episodic care • Acute care • Complex chronic care
The 22 CCGs serve the following clini­ cal areas:
• Anesthesia • Behavioral health • Cardiology • Cardiovascular surgery • Critical care • Emergency medicine
T 1
m
> H
73
m
B e c k y K u h n , FACHE, a n d C h u c k L e h n • 23

Hospital medicine Infectious disease Medical imaging Neonatal intensive care unit (ICU) and newborn care Nephrology Neuroscience Oncology Orthopedics Palliative care Pediatrics Pharmacy and therapeutics Post-acute care Primary care Pulmonary care Surgery Women’s health
involves a strategic approach, execution is key. Execution, has, in fact, become a core competence of our organization. We accomplish it by means of a three-step process:
1. Define the clinical practice (using clinical experts to study and determine practices that will significantly enhance care throughout Banner Health)
2. Design how it will be performed (including education and communication pieces)
3. Implement the clinical practice using project management tools and techniques
O n e o f B a n n e r H e a lth ‘s core
s tre n g th s is its in n o v a tiv e
a p p ro a c h to re d u c in g
th e k n o w n la g t im e
b e tw e e n th e id e n t if ic a t io n
o f a n ev id en ce -b ase d
c lin ic a l p ra c tic e a n d its
w id e a c c e p ta n c e a n d
im p le m e n ta t io n .
Banner Health also has 27 discipline teams ranging from noninvasive cardiology to medical/ surgical care and environ­ mental services to spiri­ tual care. These teams bring together leaders in a single discipline or service delivery area to standardize and improve clinical and operational performance
system-wide by addressing care and service delivery; standards of professional practice; standards of patient care; and in­ formation, technology, and workflow. Over time, some of the earlier discipline teams have disbanded, making room for addi­ tional CCGs and TRIaDs to delve deeper into process improvement.
One of Banner Health’s core strengths is its innovative approach to reducing the known lag time between the identification of an evidence-based clinical practice and its wide acceptance and implementation. Although clinical process improvement
Clinical practices describe a spectrum of care that can be delivered to a group of patients who have a specified medical condition. On the basis of the strength of the evidence, these practices can be further categorized as expected (to be done every time) or recommended (based on the clinician’s judgment). Expected practices are developed from the highest level of evidence-based research or strong consen­ sus among practitioners across Banner.
Telehealth and Enabling Technologies In 2006, Banner Health launched a tele­ health program based primarily around the Philips elCU. Two-way audio/video capabilities are now available in a large number of our ICU rooms across the orga­ nization. EHR information and physiologi­ cal data from the bedside monitors are streamed in a real-time feed to a remote monitoring center where Philips software is used to analyze the information and pro­ vide early signs of deterioration, allowing intensivists and critical care nurses to ob­ serve patients and notify on-site clinicians
24 • F R O N T I E R S O F H E A L T H S E R V I C E S M A N A G E M E N T 32U

of potentially consequential changes in a patient’s condition.
The telehealth system, called TelelCU, covers 650 acute care beds in five states in which Banner has operations. When the system is activated, patients can see and hear the remote clinician via a monitor lo­ cated across from their beds. The effect on quality of care and mortality rates has been stunning. Banner has augmented a strong bedside team with elCU technology and a 24/7 TelelCU team. This combination has saved Banner 46,435 ICU days and 1,890 lives in 2014, according to the APACHE IV actual-to-prediction metrics (Cerner 2013). Exhibits 3 and 4 depict the reduced length of stay and mortality, respectively.
Banner’s relationship with Philips began in the ICU setting but has been extended to
the ambulatory setting for in-home moni­ toring of the complex, acute population who may have difficulty traveling to medi­ cal appointments or receiving the care they need. More than 500 patients, primarily Medicare beneficiaries, are now accessing care through specially programmed tablets. These tablets, which we provide to program participants, allow them to initiate video consults with caregivers. This virtual care is coupled with live visits from health coaches and an ongoing primary care relationship. The program, which began in 2013, has demonstrated a 45 percent reduction in hospitalizations and a 27 percent reduction in the cost of care.
Other areas of telehealth that are show­ ing great promise include telestroke care to connect stroke specialists with patients
E X H I B I T 3 Longitudinal ICU and Hospital Length of Stay
ICU Length o f Stay ——- APACHE Predicted
“o ngn
0.60 •
< 0.00
2009 2010 2011 2012 2013
Hospital Length o f Stay ——- APACHE Predicted
s 120 .y 1.00 “U S u 0.60
4-* <
0.00
2009 2010 2011 2012 2013
^ Banner Health Network
Source: Banner H ea lth and Philips.
Note: APACHE is a reg istered tradem ark o f Cerner Corp.
L igh t-b lue lines = Banner’s ICU and hosp ita l length o f stay.
“Tl
m
> H
C
7° m
B e c k y K u h n , FACHE, a n d C h u c k L e h n • 25