Strategic Review Process for an Accountable Care Organization and Emerging Accountable Care Best Practices

Abstract

Accountable Care Organizations (ACOs), like other care entities, must be strategic about which initiatives they support in the quest for higher value. This article reviews the current strategic planning process for the Johns Hopkins Medicine Alliance for Patients (JMAP), a Medicare Shared Savings Program Track 1 ACO. It reviews the 3 focus areas for the 2017 strategic review process – (1) optimizing care coordination for complex, at-risk patients, (2) post-acute care, and (3) specialty care integration – reviewing cost savings and quality improvement opportunities, associated best practices from the literature, and opportunities to leverage and advance existing ACO and health system efforts in each area. It then reviews the ultimate selection of priorities for the coming year and early thoughts on implementation. After the robust review process, key stakeholders voted to select interventions targeted at care coordination, post-acute care, and specialty integration including Part B drug and imaging costs. The interventions selected incorporate a mixture of enhancing current ACO initiatives, working collaboratively and synergistically on other health system initiatives, and taking on new projects deemed targeted, cost-effective, and manageable in scope. The annual strategic review has been an essential and iterative process based on performance data and informed by the collective experience of other organizations. The process allows for an evidence-based strategic plan for the ACO in pursuit of the best care for patients.

The Medicare Shared Savings Program (MSSP) was established by the Affordable Care Act to facilitate efforts to improve the quality of care for Medicare beneficiaries while simultaneously reducing unnecessary costs. In 2015, MSSP accountable care organizations (ACOs) generated total program savings of $429 million with 31% receiving shared savings bonuses.¹ Among those that reported quality outcomes in 2014 and 2015, there was improvement in 84% of quality measures reported in both years.¹ Program results show substantial variation in financial performance and quality results among the ACOs in the program. As an example, the Johns Hopkins Medicine Alliance for Patients (JMAP) achieved a high quality score of 96.2² – placing JMAP among the top quarter nationally and among the top 3 in Maryland for ACOs that had a quality score in 2015 – with declining utilization of inpatient hospitalizations and emergency department (ED) visits (Table 1), but did not achieve shared savings. High variation and lack of any strong relationship between net spending and quality has been a consistent feature of the MSSP ACO program, suggesting a continued need for better evidence of which interventions are most impactful.³

Table 1.

Johns Hopkins Medicine Alliance for Patients Performance vs Medicare Shared Savings Program Accountable Care Organizations in Performance Year 2015 ^a

Measure	JMAP	MSSP ACO Avg^h
Size
Total assigned beneficiaries^b	39,140	18,547
Total participating providers^c	2700	539
Number of participating PCPs	554	155
Number of participating specialists	1888	308
Summary of Quality, Utilization, and Expenditures
Quality Score	96%	91%
Hospital discharges per 1,000 person-years	317	331
ED visits per 1000 person-years	754	732
Per capita total expenditures	$ 12,806	$ 10,625
Disabled
Per capita disabled expenditures	$ 15,956	$ 10,371
Average disabled HCC^d risk score	1.30	1.12
Disabled person-years^e	6915	2473
Acute Expenditures
Total inpatient expenditures^f	$ 5028	$ 3361
Post-Acute Care Expenditures
Hospice expenditures	$ 117	$ 262
Skilled nursing facility or unit expenditures	$ 296	$ 904
Home health expenditures	$ 337	$ 621
Imaging
CT events^g	946	745
MRI events^g	364	287

Data from CMS 2015 Shared Savings Program Public Use File.² All expenditures are annualized, truncated, weighted mean expenditures per assigned beneficiary person-years in the performance year.

Number of assigned beneficiaries, performance year.

Includes participating PCPs, specialists, nurse practitioners, physician assistants, and clinical nurse specialists.

Hierarchical Condition Category (HCC) Coding.

Number of assigned beneficiaries with disabled enrollment type in the performance year adjusted for the total number of months that each beneficiary was classified as disabled. Number of disabled person-months divided by 12.

Includes all hospital provider types including but not limited to short-term acute care hospital, long-term care hospital, rehabilitation hospital or unit, and psychiatric hospital or unit.

Total number of events per 1000 person-years in the performance year.

Non-weighted average of MSSP ACOs.

ACO, Accountable Care Organization; CMS, Centers for Medicare & Medicaid Services; CT, computed tomography; ED, emergency department; JMAP, Johns Hopkins Medicine Alliance for Patients; MRI, magnetic resonance imaging; MSSP, Medicare Shared Savings Program; PCP, primary care physician.

Without a clear pathway to shared savings and with budgetary constraints, ACOs, like other care delivery entities, must be strategic about which initiatives they support in the quest for higher value. Although reviews of general ACO best practices are available, less has been published on continual strategy refinement for existing entities. This article will overview the current JMAP strategic planning process, which has evolved over the 3 years of operation. It will then review the 3 general areas focused on for the 2017 strategic review and the associated best practices identified for these care delivery opportunities (Table 2). Finally, it will review the ultimate selection of priorities for the coming year and offer early thoughts on implementation. JMAP believes these priorities likely will align with those of many ACOs as they strive to deliver higher value care.

Table 2.

Summary of Select Accountable Care Organization Best Practices

Topic	Targeted population	Key best practices	Best practice outcomes	JMAP proposed initiatives
Complex case coordination for complex, at-risk patients	Highest cost, highest risk patients	• Frequent in-person meetings • Direct interaction with providers • Role as communication hub • Transitional care support • Evidence-based education • Strong medication management⁷	• Most Medicare Demonstration programs led to unchanged or increased Medicare spending⁶ • MGH demonstrated savings ($2.65-$3.35 per dollar spent), decreased mortality and acute care utilization, but not readmissions. High participation levels of 84–88%⁸	• Define new protocols and targets for high-risk CCM to align with best practices
Disabled Populations	Disabled and functionally limited population	• CAPABLE program to provide home-based assessments and repairs¹³	• CAPABLE improved activities of daily living by 75% and depressive symptoms by 53%. Saved an average of $867 per member per month (∼$10,000 per year) compared to matched controls¹³	• Explore opportunities for partnership and expansion of CAPABLE program
Behavioral Health Integration	Patients with mental illness or active substance abuse	• Implementation of IMPACT Model¹⁸	• UCLA Behavioral Health Associates increased number of patients receiving behavioral health services from 4% to 13% and reduced ED use 13% (260 visits per 1000 patient-years before the intervention to 226 visits after treatment, P < 0.01)¹⁸	• Expand HBS referrals with risk-based models • Explore reimbursement opportunities
Skilled Nursing Facilities (SNFs)	Hospitals and high-volume regional SNFs	• Identify and partner with high-quality SNFs • Supplement publically available data with self-reported measures for partners²²	• Publically available metrics are not consistently associated with adjusted risk of readmission or death²³ • Partners HealthCare SNF Collaborative used 2-step screening, inclusive of self-reported measures. SNF network had higher CMS Five-Star Scores (4.6 vs 3.2, P < 0.001) and more likely to have 5 days of clinical coverage (17.0 vs 2.9%, P = 0.02) and to have physician see admits within 24 (38.3 vs 17.1%, P = 0.02) to 48 hours (93.6 vs 80.0%, P = 0.03)²²	• Align with health system efforts to improve SNF transitions and apply best practice care pathways to create a broader SNF collaborative and preferred network
Home-based primary care (HBPC)	Chronically ill, frail, and physically or cognitively disabled patients	• Interdisciplinary care team • Target at-risk patients for program²⁵ • Integrate palliative care strategy⁴⁰	• HBPC reduces hospitalizations and hospital days (moderate strength of evidence) and can decrease ED and specialty visits (low strength of evidence)²⁴ • Medstar showed 17% lower costs, saving $8477 per beneficiary (P = 0.003) over 2 years²⁵ • ProHealth home palliative care saved $12,000 per beneficiary in the final 3 months of life ($20,420 vs $32,420; P = 0.0002) with a 35% reduction in Part A and 37% in Part B Medicare expenditures⁴⁰	• Align with Maryland state-based initiative for HBPC via Community Health Partnership of Baltimore
Imaging appropriate-ness	All ACO affiliated ordering providers	• “Hard stop” interventions are more effective²⁹ • Supplement guidelines with local expert opinion • Provide physician feedback and educational sessions^31,32 • Select narrow list of targeted studies³³	• Decision Support Systems (DSS) improved imaging appropriateness (effect size, −0.49; CI −0.71 to −0.26) with lesser effect on total usage (effect size, −0.13; CI −0.23 to −0.04)²⁹ • MGH decreased low utility exams from 6% to 2%^31,32 • Virginia Mason significantly reduced utilization for lumbar MRIs for low back pain (RR 0.77; CI 0.87–0.67; P = 0.0001), head MRIs for headache (RR 0.76; CI 0.91–0.64; P = 0.001) and sinus CTs for sinusitis (RR 0.73; CI 0.82–0.65; P < 0.0001)³³	• Champion Choosing Wisely efforts and DSS implementation • Review specific opportunities for targeted educational interventions
Part B Drug Expenditure	Medicare Part B medications	• High-cost drug substitutions with less expensive alternatives • Transfer venue of care to lower cost settings³⁹	• Single drug substitution of ranibizumab ($2023 per dose) with bevacizumab represents potential for $18 billion in Part B savings over 10 years³⁷	• Explore opportunities for high-cost drug substitutions in appropriate circumstances

ACO, Accountable Care Organization; CCM, complex care management; CI, confidence interval; CMS, Centers for Medicare & Medicaid Services; CT, computed tomography; ED; emergency department; HBS, Health Behavior Specialist; JMAP, Johns Hopkins Medicine Alliance for Patients; MGH, Massachusetts General Hospital; MRI, magnetic resonance imaging; RR, risk ratio.

Overview of the JMAP Strategic Planning Process

JMAP includes primary and specialty care providers from the Johns Hopkins University (JHU) faculty, Johns Hopkins Community Physicians, and 3 community-based practices – Potomac Physician Associates, Columbia Medical Practice, and Cardiovascular Specialists of Central Maryland – as well as the 5 Johns Hopkins Medicine (JHM) hospitals in Maryland and Washington, DC. In total, there are approximately 2700 providers responsible for managing the quality and care of nearly 40,000 Medicare fee-for-service beneficiaries. JMAP launched in 2014 and renewed for an additional 3-year term starting January 1, 2017.

The JMAP strategic review occurs annually, via a retreat-like format. It is a collaborative process that seeks to build on the current infrastructure and determine where new efforts are needed to drive toward value in the coming year. The 2017 process began with a critical review of prior year performance in relation to quality and cost to generate a broad list of potential target areas. A core group of JMAP leadership reviewed available Centers for Medicare & Medicaid Services (CMS) reports and benchmarks, internal monitoring dashboards, and ACO best practices (Table 2). In addition, a JMAP provider survey completed in the first year of operations provides ongoing guidance in identifying provider preferences for promoting patient-centered care.⁴ For each potential target area, the core group summarized best practice information and quantified cost savings and quality improvement opportunities for review by strategic retreat attendees.

The opportunities identified for the retreat were grouped into 3 broad categories: (1) optimizing care coordination for at-risk patients, (2) post-acute care (PAC), and (3) specialty care integration. In a collaborative meeting, approximately 40 clinical and administrative representatives were gathered representing key partners including Johns Hopkins HealthCare, the hospitals, the ambulatory practices, care coordination, quality, and analytics as well as beneficiaries. JMAP 2016 performance data were reviewed along with current ACO best practices in each area (Tables 1 & 2). The group was then subdivided and reviewed the data in interdisciplinary discussion groups from which evidence-driven proposals were generated. The proposals were subsequently voted on by all attendees and other key stakeholders and ranked. Each proposal was reviewed based on opportunities to leverage and advance current efforts as well as incremental budget and staff requirements. The priorities and resource demands identified were used to drive the 2017 budget process. Priorities for 2017 were then reviewed and approved by the JMAP Board of Directors.

A charter and project plan are developed for each initiative in order to outline project scope and objectives, name accountable leaders, identify key milestones, determine timelines, and define process and outcome metrics. Project teams meet with established frequency to move initiatives forward. The teams present on their progress and challenges through appropriate committees and with periodic board presentations to review progress and accomplishments to date.

Optimizing Care Coordination for Complex, At-Risk Patients

In a risk-based environment, controlling cost for the highest risk, highest cost patients has been a key strategy for controlling overall expenditures for nearly all ACOs.⁵ For JMAP, the annual planning process was an opportunity to review the cornerstone complex care coordination program, opportunities for the disabled population, and behavioral health integration efforts.

Complex care coordination

Most ACOs have focused on managing the costs of their most expensive, highest-need patients through complex care coordination programs aimed at improving beneficiary health with the hope of offsetting program costs through reductions in acute care utilization.⁶ Various demonstrations looking at these programs more broadly have shown limited effect on hospital admissions and Medicare expenditures. After factoring in the cost of administering the program, Medicare spending was either unchanged or increased in nearly all programs.⁶

A review of the Medicare Coordinated Care Demonstration found 6 care coordinator practices associated with program success: (1) supplementing patient telephone calls with frequent in-person meetings; (2) meeting in person with providers; (3) acting as a communications hub for providers; (4) delivering evidence-based education to patients; (5) providing comprehensive medication management; and (6) providing timely and comprehensive transitional care after hospitalizations.⁷ The Massachusetts General Hospital (MGH) Care Management Program incorporated many of these features and reported patient enrollment rates as high as 84%–88%.⁸

For JMAP, the complex care coordination program was built on preexisting programs utilizing a hybrid of site-specific and regional care coordinators to address the needs of complex, at-risk patients. Patients are identified by a combination of provider referrals and the JHU Adjusted Clinical Group risk-prediction model reflecting likelihood for hospitalization within the next year.^9,10 Patients with the highest predicted risk were reviewed and opportunities were found to integrate the aforementioned best practices into the program through redesigned protocols and new benchmarks.

The intervention implemented included in-depth chart reviews of the highest risk patients by both a physician and program leadership to identify areas of continuing opportunity in education, documentation, and further refinement of care management delivery strategies for all enrolled patients. Although implementation costs (beyond those of the existing program) were minimal, there was a significant time investment by current leadership for chart reviews, strategy sessions, update of care management training materials, and ongoing education of care managers on best practices. Notable improvements in meeting these best practices have already been observed and are currently being assessed more formally. In addition, given questions of financial impact of care coordination programs, a larger analysis of the impact of care coordination efforts on cost and utilization within the ACO, through a difference-in-difference analysis, is currently under way.

Disabled populations

Individuals with functional limitations and chronic conditions are more than 4 times as likely to be among the top 5% of users of all health services compared to the general population.¹¹ Despite this, functional difficulties with activities of daily living (ADLs) remain a cost driver that typically is not addressed or compensated for in traditional medical care.¹²

The Community Aging in Place, Advancing Better Living for Elders (CAPABLE) program was a demonstration product funded by the Center for Medicare and Medicaid Innovation (CMMI).¹³ CAPABLE addressed the home environment and modifiable individual limitations for dually eligible older adults with physical disabilities with an interprofessional team that performed home assessments and repairs.¹³ During the 5-month intervention the program was able to improve performance of ADLs by 75% and depressive symptoms by 53%.¹³ Compared to matched controls, the program reduced total Medicare expenditures by more than $10,000 per patient per year driven by reductions in inpatient and outpatient expenditures, reduced readmissions, and fewer observation stays.¹⁴ At an intervention cost of approximately $2825 per beneficiary, this implies approximately $7000 in Medicare savings per beneficiary enrolled in the pilot studied.¹³

The JMAP population was noted to have a higher proportion of disabled patients and these patients were, on average, higher risk and higher cost than other MSSP ACOs (Table 1).² JMAP is partnering with the CAPABLE program to implement a pilot program to specifically address the needs of the disabled population and demonstrate proof of concept within the Medicare ACO context. Although a proportion of JMAP beneficiaries are dually eligible, this was not a criteria for enrollment as in the original pilot. Further analysis of the JMAP pilot will be needed to determine cost savings in this population.

Behavioral health integration

Individuals with high health care utilization have disproportionately high rates of comorbid psychiatric conditions and substance use.¹⁵ Depression is more than 3 times as common among high health care utilizers in primary care.¹⁶ ACOs have a mixed degree of engagement in behavioral health, with 43% having some behavioral health initiative between 2012 and 2015.¹⁷ UCLA Health System Behavioral Health Associates implemented an adapted IMPACT (Improving Mood-Promoting Access to Collaborative Treatment) model within the ACO framework, more than tripling the number of patients receiving behavioral health services and reducing ED use by 13%.¹⁸

For JMAP, opportunities related to substance abuse and behavioral health were identified by an early provider survey demonstrating that 82% of JMAP providers felt provision of behavioral therapists would facilitate high-quality and cost-effective care within the ACO.⁴ Data from the Johns Hopkins Community Health Partnership (J-CHiP) – a prior regional transformation initiative supported by a CMMI Health Care Innovation Award – revealed 33% of Medicare high-risk patients struggle with depression, 18% with anxiety, and 8% each with bipolar and schizophrenia.¹⁰ Building on the J-CHiP model, JMAP deploys Health Behavior Specialists (HBSs) – licensed clinical social workers – to support network primary care sites through a regional allocation model. A psychiatry consultant performs telephonic case reviews and serves as a supervisory layer for the HBSs, and a dedicated Behavioral Health Medical Director oversees all efforts. To date, there have been more than a thousand HBS encounters. To further strategically target provision of these services, JMAP is exploring opportunities to supplement primary care referrals with risk-based identification. Barriers for the program include recruitment of HBSs, operationalizing an appropriate billing process, and data limitations, particularly around the sharing of claims data related to substance use disorders. These barriers are not unique to JMAP, particularly the challenge of data availability on full integration of behavioral health services.¹⁷ JMAP is currently looking at the feasibility of direct billing for aspects of HBS services, but to date, this does not occur. JMAP is conducting an in-depth analysis of the behavioral health intervention to identify the impact on cost and utilization to facilitate ongoing strategic planning.

Post Acute Care

ACOs are responsible for both total cost of care and quality across the care continuum. Thus management of the post-discharge period is essential to improving quality and reducing costs. PAC is a broad category that can be inclusive of rehabilitation, skilled nursing facilities (SNFs), and home health care, and is relied on heavily for the care of Medicare beneficiaries. In 2011, 43% of Medicare beneficiaries were discharged from hospitals to PAC settings, accounting for 17% of 2012 Medicare spending.¹⁹

JMAP's utilization in these areas is below both its MSSP ACO peers and the national average (Table 1).² Although it spends less on PAC, JMAP's total inpatient per beneficiary spend exceeds the MSSP ACO average.² This likely reflects an opportunity to shift care to the optimal care setting as well as the impact of increased Medicare hospitalization costs associated with the Maryland all-payer rate setting.²⁰ Additionally, compared to ACOs nationally, JMAP feels there is opportunity to decrease readmissions. Investment in increasing a high-quality post-discharge network and associated resources may allow for earlier hospital discharges, decreased SNF length of stay (LOS), and fewer readmissions through improved coordination of care with the post-discharge setting.

Skilled nursing facility initiatives

SNFs are a high-cost, high-volume provider of PAC accounting for $28.7 billion in Medicare costs in 2012.²¹ There has been an increase in SNF readmission rates over time, with evidence that stronger linkages between hospitals and SNFs are associated with lower readmissions.²² To combat these cost and quality concerns, some ACOs are beginning to explore the potential benefits of SNF partnership.²² One study found that publicly available quality and performance metrics were not consistently associated with differences in the adjusted risk of readmission or death.²³

In the context of its Pioneer ACO, Partners HealthCare System launched a SNF Collaborative Network. Initial screening criteria were based on public scores with secondary criteria based on self-reported measures.²² Their network of selected SNFs had higher CMS Five-Star Scores and was more likely to have greater than 5 days of clinical coverage and a physician see admitted individuals within 24 to 48 hours.²²

Within JMAP, 83% of ACO providers felt that improving communication and coordination of care throughout the care continuum was important.⁴ To capitalize on these opportunities for improved care transition, JMAP has partnered with other Johns Hopkins Health System entities to launch a JHM Skilled Nursing Facility Collaborative extending across the full network. Initial collaborative efforts are directed toward improving and unifying transitions into and out of SNFs. JMAP has partnered with the local Quality Improvement Organization to facilitate quality improvement efforts in collaboration with its SNF partners. Anticipated cost reduction for the collaborative would be achieved through decreased SNF readmissions and ED visits as well as decreased LOS at the hospitals and SNFs. Both the interventions and measures will need ongoing refinement as data are analyzed.

Home-based services

Home-Based Primary Care (HBPC), a model that combines home-based care for medical needs with care coordination efforts, is an alternative way of organizing and delivering care that may better address the needs of chronically ill, frail, and physically or cognitively disabled patients who have difficulty accessing traditional office-based primary care. A review of the literature by the Agency for Healthcare Research and Quality found that HBPC reduced hospitalization and hospital days and can decrease ED and specialty visits.²⁴ The MedStar Health System HBPC program demonstrated 17% lower Medicare costs in comparison to a matched cohort, driven primarily by lower hospital and SNF costs.²⁵

When surveyed, 80% of JMAP providers felt provision of home-based services would facilitate ACO effectiveness.⁴ JMAP interventions to date include a small remote monitoring program related to at-risk heart failure patients. Efforts are under way to develop a HBPC program through Johns Hopkins' involvement in a parallel Maryland state-based initiative, the Community Health Partnership of Baltimore.²⁶ In light of these efforts and the significant investment to initiate such a program, it was decided to defer additional investment until the impact of the home-based initiative could be better understood.

Specialty Care Integration

Specialty care integration with primary care can have a key role in ACO success as many of the most complicated patients are comanaged by specialists who drive a large portion of their costs. For example, although 46% of office visits in 2009 were to primary care, they accounted for only 30% of visit expenditures.²⁷ Coordination of care between primary and specialty care remains essential, and extends to non–ACO specialty providers. A review of JMAP's data illustrated elevated spending in 2 areas typically associated with specialty prescribing: high-cost outpatient imaging and Part B drug expenditures.

Imaging appropriateness

Increasing use of advanced medical imaging is often cited as a key driver of medical spending growth, with an estimated 20% to 50% of imaging considered unnecessary.²⁸ Various decision support systems (DSS) have been implemented and overall have shown a moderate effect on imaging appropriateness but a small effect on total usage.²⁹ With DSS implementation, physicians had concerns about the appropriateness of guidelines, added time to work, and the frequency with which orders could not be linked to a guideline.³⁰

Two hospitals with successful DSS implementations include MGH and Virginia Mason Medical Center. MGH decreased low utility exams from 6% to 2%, incorporating feedback to physicians, educational sessions, and supplementation of American College of Radiology appropriateness criteria with local expert opinion.^31,32 Virginia Mason Medical Center reduced utilization by focusing on 3 imaging modalities that are high yield for savings: (1) lumbar magnetic resonance imaging (MRIs) for low back pain, (2) head MRIs for headache, and (3) sinus computed tomography scans (CTs) for sinusitis.³³

For JMAP, data revealed elevated imaging expenditures driven by higher CT and MRI rates compared to the MSSP ACO average (Table 1).² Multiple efforts are under way at Johns Hopkins to support decreased utilization of low-value imaging driven largely by the Johns Hopkins High Value Care Committee and the Department of Radiology. These include a series of quality improvement projects, early steps toward integration of embedded decision support software, and a diagnostic radiology telephonic consult service.³⁴ Although not driven by JMAP, JMAP has continued to champion these various high-value imaging related efforts and supported dissemination of information to its providers. As a joint effort, JMAP has targeted 3 highly utilized imaging studies in the ambulatory setting – lumbar spine MRI, brain and sinus CT, and abdominal CT with and without contrast. In addition, JMAP sought to increase engagement and education within the ED, where health system imaging utilization may be the greatest. Further analysis of the proportion of utilization amenable to intervention, including location, study, and indication, are needed to identify realistic cost targets.

Part B drug expenditures

Included in the total cost of care for which Medicare ACOs are responsible is Part B drug expenditures, which cover infusible and injectable drugs and biologics administered in hospital outpatient departments and physician offices. Overall Part B prescription drug spending has increased from $9.4 to $18.5 billion from 2005–2014.³⁵ The top 20 drugs account for 57% of the total 2014 expenditure, with 13 of the 20 having a mean annual per-patient reimbursement more than $10,000.^35,36 They include 11 drugs for cancer, 3 for macular degeneration, and 2 each for hematopoietic growth factors and rheumatoid arthritis. Opportunities to decrease costs have focused on appropriate high-value substitutions. Although oncology practice guidelines reveal limited opportunity for such substitutions, ophthalmology may represent an area of promise. Ranibizumab, for example, has similar efficacy to bevacizumab, but at a price of $2023 per dose is 40 times more expensive than the alternative (representing potential for $18 billion in Part B savings over 10 years with a single drug substitution).³⁷

Additional efforts have focused on the location of Part B medication administration, typically referred to as venue of care. Nationally, the percentage of spending occurring in the hospital outpatient center grew rapidly from 21% to 34% from 2005–2014.³⁵ However, in most instances, the hospital outpatient center is the highest cost site of drug administration, costing over 50% more than administration in physicians' offices or patients' homes.³⁸ Thus, efforts have been focused on redirecting the location of infusions from hospital outpatient centers to offices or homes.³⁹

Part B expenditures for JMAP are higher than the MSSP ACO average and trending upward. Opportunities for savings were identified in ophthalmology, particularly around possible substitution for macular degeneration treatments. The Maryland Global Budget Revenue²⁰ with capped all-payer per capita cost growth provides unique leverage as JMAP and hospital incentives are aligned in reducing Part B drug expenditures in regulated space. Led by the pharmacy team, JMAP has engaged ophthalmology leadership to review the opportunity, provided a detailed breakdown of variations in usage patterns both between sites and between providers, and identified learnings from peer institutions that have undertaken similar projects. Broader ophthalmology faculty education and engagement initiatives are under way. Additionally, ophthalmology site-of-service conversations have already resulted in care shifts of injections from facility to office-based ambulatory locations.

Discussion

In October 2016, JMAP held its third annual medical and quality strategic retreat and planning process, reviewed current interventions and prior year performance in the domains of quality and cost of care, and surveyed the evolving evidence base for best practices in the field. The program opportunities that were ultimately selected reflect an element of prioritization based on personnel and budgetary realities, as well as the goal of leveraging and connecting with other ongoing system efforts.

The annual strategic review has been an essential and iterative process that is now more informed by performance data than in previous years. As ACOs seek to simultaneously improve quality and decrease costs, they will need an evidence-based tool kit of high-value services applied to their current circumstances and environment. Although it is not always the principal goal, generating shared savings remains an elusive goal for nearly 75% of Medicare ACOs as it can be a key driver of sustainability. Although the strategic planning process is tailored for performance in a Medicare ACO, the larger opportunity is around establishing care delivery infrastructure and processes that can be applied more broadly to additional segments of the population, leveraging commercial or other payment arrangements. As more organizations transition to Advanced Alternative Payment Models with both upside and downside risk and as commercial payers seek to introduce various value-based shared savings arrangements, this will become an increasingly valuable process. The authors hope that this review of the efforts of a single ACO, informed by the experiences of many, can help other ACOs develop their own processes to evolve in each one's pursuit of the best possible care for patients.⁴⁰

Footnotes

Acknowledgments

The authors wish to acknowledge all those who participated in the 2016 JMAP Medical and Quality Retreat and planning process as well as the Board of Directors and other members of the broader JMAP team.

Author Disclosure Statement

The authors declare that there are no conflicts of interest. The authors received no financial support for the research, authorship, and/or publication of this article.

References

Centers for Medicare & Medicaid Services. Medicare Accountable Care Organizations 2015 Performance Year Quality and Financial Results. Published August 25, 2016. https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2016-Fact-sheets-items/2016-08-25.html Accessed November 7, 2016 .

Centers for Medicare & Medicaid Services. 2015. Shared Savings Program (SSP) Accountable Care Organizations (ACO) PUF. https://data.cms.gov/Special-Programs-Initiatives-Accountable-Care-Orga/2015-Shared-Savings-Program-SSP-Accountable-Care-O/7rrf-3gxr#column-menu Accessed January 11, 2017 .

Muhlestein

, Saunders

, McClellan

. Medicare Accountable Care Organization Results for 2015: the Journey to Better Quality and Lower Costs Continues. Published September 9, 2016. http://healthaffairs.org/blog/2016/09/09/medicare-accountable-care-organization-results-for-2015-the-journey-to-better-quality-and-lower-costs-continues/ Accessed February 10, 2017 .

Marsteller

, Young

, Fakeye

, et al. Early provider perspectives within an accountable care organization. Am J Accountable Care, 2016:4:27–37.

Hong

, Siegel

, Ferris

. Caring for High-Need, High-Cost Patients: What Makes for a Successful Care Management Program? Published August 2014. www.commonwealthfund.org/∼/media/files/publications/issue-brief/2014/aug/1764_hong_caring_for_high_need_high_cost_patients_ccm_ib.pdf Accessed January 11, 2017 .

Nelson

, Buntin

, Baumgardner

, et al. Working Paper Series Congressional Budget Office: Lessons from Medicare's Demonstration Projects on Disease Management and Care Coordination Lessons from Medicare's Demonstration Projects on Disease Management and Care Coordination. January 2012. https://www.cbo.gov/sites/default/files/112th-congress-2011-2012/workingpaper/WP2012-01_Nelson_Medicare_DMCC_Demonstrations_1.pdf Accessed December 26, 2016 .

Brown

, Peikes

, Peterson

, Schore

, Razafindrakoto

. Six features of Medicare coordinated care demonstration programs that cut hospital admissions of high-risk patients. Health Aff, 2012; 31:1156–1166.

Kodner

. Managing high-risk patients: the Mass General care management programme. Int J Integr Care, 2015; 15:e017.

Weiner

. The Johns Hopkins ACG ® System Technical Reference Guide. 2011. http://acg2.jhsph.org/public-docs/ACGv10.0TechRefGuide.pdf Accessed February 18, 2017 .

10.

Berkowitz

, Brown

, Brotman

, et al. Case study: Johns Hopkins community health partnership: a model for transformation. Healthc (Amst), 2016; 4:264–270.

11.

Aleckxih

, Shen

, Chan

, Taylor

, Drabek

. Individuals Living in the Community with Chronic Conditions and Functional Limitations: A Closer Look. Lewin Gr. 2010. https://aspe-hhs-gov.ezp.welch.jhmi.edu/sites/default/files/pdf/75961/closerlook.pdf Accessed January 30, 2017 .

12.

Reuben

, Tinetti

. Goal-oriented patient care—an alternative health outcomes paradigm. N Engl J Med, 2012; 366:777–779.

13.

Szanton

, Leff

, Wolff

, Roberts

, Gitlin

. Home-based care program reduces disability and promotes aging in place. Health Aff (Millwood), 2016; 35:1558–1563.

14.

Ruiz

, Snyder

, Rotondo

, Cross-Barnet

, Colligan

, Giuriceo

. Innovative home visit models associated with reductions in costs, hospitalizations, and emergency department use. Health Aff (Millwood), 2017; 36:425–432.

15.

Everett

, Reese

, Coughlin

, et al. Behavioural health interventions in the Johns Hopkins community health partnership: integrated care as a component of health systems transformation. Int Rev Psychiatry, 2014; 26:648–656.

16.

Berghofer

, Roll

, Bauer

, Willich

, Pfennig

. Screening for depression and high utilization of health care resources among patients in primary care. Community Ment Health J, 2014; 50:753–758.

17.

Fullerton

, Henke

, Crable

, Hohlbauch

, Cummings

. The impact of Medicare ACOs on improving integration and coordination of physical and behavioral health care. Health Aff (Millwood), 2016; 35:1257–1265.

18.

Clarke

, Jeffrey

, Grossman

, Strouse

, Gitlin

, Skootsky

. Delivering on accountable care: lessons from a behavioral health program to improve access and outcomes. Health Aff (Millwood), 2016; 35:1487–1493.

19.

Medicare Payment Advisory Commission (MedPAC). Medicare Post-Acute Care Reforms. June 14, 2013. www.medpac.gov/docs/default-source/congressional-testimony/20130614_wandm_testimony_pac.pdf?sfvrsn=0 Accessed February 10, 2017 .

20.

Rajkumar

, Patel

, Murphy

, et al. Maryland's all-payer approach to delivery-system reform. N Engl J Med, 2014; 370:493–495.

21.

Medicare Payment Advisory Commission (MedPAC). Report to the Congress: Medicare Payment Policy. March 2014. www.medpac.gov/docs/default-source/reports/Mar14_EntireReport.pdf Accessed February 18, 2017 .

22.

Lage

, Rusinak

, Carr

, Grabowski

, Ackerly

. Creating a network of high-quality skilled nursing facilities: preliminary data on the postacute care quality improvement experiences of an accountable care organization. J Am Geriatr Soc, 2015; 63:804–808.

23.

Neuman

, Wirtalla

, Werner

. Association between skilled nursing facility quality indicators and hospital readmissions. JAMA, 2014; 312:1542–1551.

24.

Totten

, White-Chu

, Wasson

, et al. Home-based primary care interventions. Rockville, MD: Agency for Healthcare Research and Quality, 2016.

25.

Eric De Jonge

, Jamshed

, Gilden

, Kubisiak

, Bruce

, Taler

. Effects of home-based primary care on Medicare costs in high-risk elders. J Am Geriatr Soc, 2014; 62:1825–1831.

26.

The Johns Hopkins Hospital. HSCRC Transformation Implementation Program; The Community Health Partnership of Baltimore. Published 2015. www.hscrc.state.md.us/documents/md-maphs/plan/RFP/CHP/1. Community Health Partnership Proposal.pdf Accessed February 10, 2017 .

27.

Davis

, Carper

. Statistical Brief #381: Use and Expenses for Office-Based Physician Visits by Specialty, 2009: Estimates for the U.S. Civilian Noninstitutionalized Population. Survey, Agency for Healthcare Research and Quality Medical Expenditure Panel. Published 2012. https://meps.ahrq.gov/data_files/publications/st381/stat381.pdf Accessed February 10, 2017 .

28.

Durand

, Lewin

, Berkowitz

. Medical-imaging stewardship in the accountable care era. N Engl J Med, 2015; 373:1691–1693.

29.

Goldzweig

, Orshansky

, Paige

, et al. Electronic health record-based interventions for improving appropriate diagnostic imaging: A systematic review and meta-analysis. Ann Intern Med, 2015; 162:557–565.

30.

Timbie

, Hussey

, Burgette

, et al. Medicare Imaging Demonstration Final Evaluation: Report to Congress. 2014. www.rand.org/pubs/research_reports/RR706.html Accessed November 28, 2016 .

31.

Rosenthal

, Weilburg

, Schultz

, et al. Radiology order entry with decision support: initial clinical experience. J Am Coll Radiol, 2006; 3:799–806.

32.

Sistrom

, Dang

, Weilburg

, Dreyer

, Rosenthal

, Thrall

. Effect of computerized order entry with integrated decision support on the growth of outpatient procedure volumes: seven-year time series analysis. Radiology, 2009; 251:147–155.

33.

Blackmore

, Mecklenburg

, Kaplan

. Effectiveness of clinical decision support in controlling inappropriate imaging. J Am Coll Radiol, 2011; 8:19–25.

34.

Trakhtenbroit

, Abdollahian

, Horton

, Johnson

. Choosing wisely and imaging 3.0 professional development: radiology resident consultant service. J Am Coll Radiol, 2016; 13:730–732.

35.

Sheingold

, Marchetti-Bowick

, Nguyen

, Yabroff

. Medicare Part B Drugs: Pricing and Incentives. 2016. https://aspe.hhs.gov/sites/default/files/pdf/187581/PartBDrug.pdf Accessed February 10, 2017 .

36.

Schrag

. Reimbursing wisely? CMS's trial of Medicare Part B drug payment reform. N Engl J Med, 2016; 374:2101–2105.

37.

Hutton

, Newman-Casey

, Tavag

, Zacks

, Stein

. Switching to less expensive blindness drug could save Medicare Part B $18 billion over a ten-year period. Health Aff (Millwood), 2014; 33:931–939.

38.

Lotvin

, Shrank

, Singh

, Falit

, Brennan

. Specialty medications: traditional and novel tools can address rising spending on these costly drugs. Health Aff (Millwood), 2014; 33:1736–1744.

39.

Amara

, Adamson

, Lew

, Slonim

. Accountable care organizations: impact on pharmacy. Hosp Pharm, 2014; 49:253–259.

40.

Lustbader

, Mudra

, Romano

, et al. The impact of a home-based palliative care program in an accountable care organization. J Palliat Med, 2017; 20:23–28.