Overview of Accountable Care Organizations for Oncology Specialists

The authors review the economic factors leading to the growth of accountable care organizations and describe how oncology specialists may participate in the model to ensure success for physicians and patients.

Abstract

Accountable care organizations (ACOs) are poised to become major components of health care delivery in the United States. The practice of oncology, often laden with high charges, is likely to undergo major shifts as ACOs become widespread. In this article, we review the economic factors leading to the growth of ACOs and discuss some elements of the current ACO model proposed in the Affordable Care Act. Oncology specialists—in medicine, surgery, and radiation oncology—will have important roles in determining the place of specialty care in an ACO framework and will have to take the lead in educating patients, primary care physicians, and administrators on the value propositions related to their activities. We also describe how oncology specialists may participate in the model to ensure success for physicians and patients.

Introduction

The Patient Protection and Affordable Care Act of 2010 is scheduled to become fully effective by 2014. By that time, a large proportion of Medicare payments will be delivered within the rubric of accountable care organizations (ACOs). Private insurance companies, like Blue Cross/Blue Shield, are also forming ACO-like agreements with health care providers, betting that the future of health care will rely on this model.¹ As of July 2012, 153 organizations in 45 states across the United States had joined the program with Medicare.² Another 162 health care organizations had identified themselves as ACOs or begun the process of transitioning to ACOs with private insurers less than 1 year after the announcement of the finalized ACO regulations.³ Although the model is clearly poised to become a major force in health care delivery, much uncertainty remains as to its implications for health care providers.

Under the ACO model, panels of primary care practitioners (PCPs) may vet the necessity, quality, value, and accountable delivery of specialty diagnostic and therapeutic procedures, including cancer care.⁴ The estimate for the direct cost of cancer care in 2010 was more than $124 billion, making cancer one of the costliest medical conditions in the United States.⁵ In light of pressure to reduce spending, there is a clear risk that ACOs may disrupt the traditional practice of oncology specialists. By and large, the oncology community has remained rather skeptical that ACOs will be beneficial for physicians and patients.⁶ How will the implementation of the ACO model affect the roles of cancer care specialists (medical, surgical, and radiation oncologists) involved in care delivery? This article will provide a historical context for ACOs, give an overview for the model, and consider the expected or optimal roles of cancer care specialists within an ACO.

Scope of Health Care Costs

The United States is currently in the midst of a “massive, multifront war on federal government deficits,”^7(p1655) and health care expenditures are a critical battlefield.⁷ Despite attempts to reign in spending, health care expenditures by public and private entities remain staggeringly high. The total national health expenditures for 2012 neared $2.8 trillion, which is 17.9% of the gross domestic product (GDP) and almost $9,000 per capita.⁸ These figures are the continuation of a trend going back decades (Fig 1).⁹

Figure 1.

National health expenditures per capita (adjusted for inflation). Data adapted.⁹

The unrelenting increase in national health care spending has been attributed to many culprits (Table 1). However, a number of studies have suggested that medical technology remains the greatest driver of health care spending over time. This includes the wealth of new medications, devices, diagnostic tools, procedures, and specialists available for patient care.¹⁰ On average, total national health care spending has risen 2% faster than the GDP over the past 30 years. The period from 1993 to 2000, however, managed to buck the trend, with the increase in health care spending lower than the growth of the GDP.¹³ This period is marked by the growth, and subsequent decline, of health maintenance organizations (HMOs) and has led many lawmakers to conclude that managed care is a successful method of controlling health care costs.

Table 1.

Estimated Contributions of Selected Factors to Growth in Real per Capita Spending on Health Care, 1940 to 1990

Factor	Studies Estimating Contributions of Selected Factors (%)
	Smith et al (2000)11	Cutler (1995)*	Newhouse et al (1992)12
Aging of population	2	2	2
Changes in third-party payment	10	13	10
Growth of personal income	11-18	5	< 23
Prices in health care sector	11-22	19	†
Administrative costs	3-10	13	†
Defensive medicine and supplier-induced demand	0	†	0
Technology-related changes in medical practice	38-62	49	> 65

NOTE. Data in table represent the estimated percentage share of long-term growth for which each factor accounts. Data adapted.¹⁰

^*Cuter, unpublished report for National Institutes of Health Economic Roundtable, 1995.

^†Not estimated.

Expansion and Contraction of HMOs

ACOs are a response to the positive and negative components of the managed care movement of the last century. The roots of managed care date back to 1929, when farmers in a rural cooperative in Oklahoma paid a fixed fee for direct access to medical care.¹⁴ However, managed care did not begin to take off until the 1970s, when the Nixon administration promoted planning grants and loan guarantees for newly developed HMOs. In contrast to the traditional fee-for-service infrastructure, an infrastructure that seemed to provide perverse financial incentives for continuing disease management fees, HMOs were meant to establish financial incentives for continued health rather than sickness. HMOs would ostensibly shift the curve of rapidly escalating health care costs.¹⁵

When enrolled in an HMO plan, patients had access to a restricted network of physicians and services, as determined by the HMO. Access to specialists typically came through referrals by the PCP, who served as the gatekeeper to additional services.¹⁶ With these organizations sprouting across the country during the 1990s, costs did initially seem to come under more reasonable control. Access to specialists, including oncology services, was often contingent on burdensome preapproval for treatment and diagnostic procedures. The process created a perception that the HMOs were restricting specialists' independent ability to provide adequate care. Subsequent studies have shown that diagnosis, treatment, and outcome in cancer care did not differ in HMO plans versus traditional fee-for-service plans.^17,18 Despite this, the attempt by HMOs to hold down costs was viewed with suspicion by the public.¹⁹ A full-scale political backlash ensued, featuring television advertisements showing seemingly scared senior citizens protesting against this corporate intrusion on the physician-patient relationship.²⁰ Many providers moved away from the HMO model and relaxed the controversial restrictive elements of managed care of the 1990s. As the cost curves began to accelerate once again, lawmakers searched to devise methods of holding down health care costs.

Formation of ACOs

The current ACO movement can be considered Managed Care 2.0. Although private insurance companies are also forming ACO-like agreements with providers, the scheme put forward by Medicare is the national prototype. Enrolling in the Medicare program is optional for patients and providers, but both will have significant built-in incentives and disincentives to pursue this alternative to routine fee-for-service offerings.²¹

As currently envisioned, the foundation of any new ACO will be a network of providers in the organization. The providers will be the PCPs in charge of patient care and will subsequently be held accountable for patient outcomes. Unlike previous iterations of managed care, 75% of an ACO must be owned by member health care providers and governed by a board elected from these providers. Although a physician may join multiple ACOs, he or she must have a separate contract and tax identification number with each.²¹ ACOs are encouraged to make meaningful use of electronic medical records and collect extensive patient data. The reliance on patient data to measure outcomes is the most important distinction between ACOs today and HMOs in the past.²²

A Medicare patient will not be required to join an ACO but will be prospectively placed in one if his or her physician joins. The total cost of caring for the Medicare patient is assigned to the PCP who provides the plurality of his or her care. It is therefore the role of that PCP to hold down this cost. The patient's data are shared within the ACO and with the Centers for Medicare and Medicaid Services (CMS) to monitor health outcomes. Unlike an HMO, an ACO is not a closed network. Patients are free to see additional providers outside of the ACO, to opt out of data sharing, and to see non-ACO providers as their individual needs demand.²¹

Reimbursement in an ACO

Under the current ACO reimbursement models, each ACO agrees to a 3-year contract with CMS, with the option to renew the contract after 3 years. The ACO may enter into a one-sided or two-sided shared savings contract. Under the one-sided model, CMS will project future spending for all patients in the ACO based on spending from previous years. If the actual costs of the ACO are less than the projection, then the ACO will retain 50% of savings (with the remainder 50% going to CMS). With the two-sided model, the ACO will retain 60% of the savings. But the ACO must also pay penalties if actual costs exceed projected costs (Appendix Fig A1, online only). After an initial 3-year contract period, all one-sided ACOs must transition to the two-sided model.²¹

For now, physicians in an ACO are still reimbursed under the fee-for-service model by CMS. The improved coordination of care management and health information technology is expected to reduce the providers' cost of delivering health care. Additionally, improved patient health should decrease use of health care services and therefore reduce CMS spending. ACOs are expected to transition away from fee-for-service reimbursement to episode-based or bundled payments after the initial 3-year contract. Although the details of bundled payment schemes have not yet been defined, this is also expected to lower spending.

Value and Accountability in an ACO

To ensure that quality of care is maintained, the ACO must report 33 quality measurements to CMS each year (Appendix Table A1, online only). However, cancer-specific guidelines are not included in the quality measurements. An ACO is eligible for shared savings only if it meets a quality threshold on at least 70% of measures by the third year.^21,22 The ACO structure also promotes shared accountability, because the spending projections are calculated as an aggregate of patient costs. Providers are encouraged to work together to coordinate care and reduce costs to maximize savings for the ACO as a whole.²³

Limitations and Risks in Forming an ACO

One of the major questions surrounding the rise of ACOs is also the most obvious: Will it save money? The Physician Group Practice Demonstration was a trial run by CMS to determine the specifics of cost savings and quality reporting and was the predecessor to the current ACO design. The Physician Group Practice Demonstration included 10 multispecialty groups in a simplified shared savings program. If physician groups in the demonstration met certain quality standards and reduced their costs, they could share in a portion of the savings. However, the savings over a 5-year period from 2005 to 2010 were generally small, averaging $114 per beneficiary per year on average in the program. The savings were also inconsistent among the multiple practices, with half generating substantial savings and half generating none (Table 2). The rather modest decline in spending has called into question whether transitioning Medicare beneficiaries to ACOs on a national scale will truly reduce government health care expenditures.^24–26

Table 2.

Physicians Group Practice Demonstration, 2005 to 2010: Total Shared Savings

Physicians Group Practice	Savings ($)
	Year 1	Year 2	Year 3	Year 4
Billings Clinic, Billings, MT	0	0	0	0
Dartmouth-Hitchcock Clinic, Lebanon, NH	0	6,689,879	3,570,173	328,798
Everett Clinic, Everett, WA	0	129,268	0	0
Forsyth Medical Group, Winston-Salem, NC	0	0	0	0
Geisinger Clinic, Danville, PA	0	0	1,950,649	1,788,196
Marshfield Clinic, Marshfield, WI	4,565,327	5,781,573	13,816,922	16,154,242
Middlesex Health System, Middletown, CT	0	0	0	0
Park Nicollet Clinic, St Louis Park, MN	0	0	0	0
St John's Clinic, Springfield, MO	0	0	3,143,044	8,185,757
University of Michigan Faculty Group Practice, Ann Arbor, MI	2,758,370	1,239,294	2,789,006	5,222,852

Data adapted.²²

Formation of an ACO also brings considerable risk to providers beyond financial penalties. The ACO guidelines were released with relaxed antitrust concessions, because many ACOs will form from the consolidation of smaller provider groups. These large organizations may wield increased market power and demand higher reimbursement rates from insurers, thus inadvertently increasing health care spending.²⁷ The formation of an ACO will also require immense upfront costs. This includes startup costs for establishing methods of collecting and reporting data, establishing bodies of governance, and developing internal legal agreements on reimbursements and quality metrics. Health information technology costs alone have ranged from $1 to $4 million for recently formed ACOs.²⁸ If an ACO does not meet quality standards, it may be removed from the shared savings program, further exposing providers to risk both in terms of lost capital and lost patient trust.²⁹ Lastly, ACOs may relegate specialists to a role to which they are unaccustomed and that they are untrained to fill. Although much of the talk around ACOs has been centered on the impact on primary care, the role of the specialist can also be expected to shift dramatically.³⁰

Role of Oncology Specialists

The optimal role of the oncology specialist will vary based on the type of ACO he or she joins. ACOs can take different shapes based on the needs of the community. Four general types of ACOs have been described^31–33: one, fully integrated health delivery systems similar to Kaiser Permanente in California or Norton Healthcare in Kentucky; two, hospital-led multispecialty group practices, like the Cleveland Clinic Foundation in Ohio or Mayo Clinic in Minnesota; three, joint physician-hospital partnerships, like Tucson Medical Center in Arizona or Advocate Healthcare in Illinois; and four, independent physician associations (IPAs), such as Monarch Healthcare in California.

An oncologist choosing to remain unaffiliated with an ACO in his or her community may still rely on referrals from ACOs. The referrals will come from IPAs or primary physician–led ACOs and will still require oncologists to make dramatic changes in how their practices are run. Oncology specialists must remain aware that the costs of treating patients with cancer will often reflect on those patients' PCPs. Oncology specialists will need to educate PCPs on the value of costly drug procedures or imaging or treatment techniques. Unlike chronic conditions like heart disease or diabetes, cancer in any form is outside of the purview of PCPs, and there is little reluctance to consult a specialist for cancer care. However, the referral patterns will change toward oncology specialists who share a common vision for low-cost, high-quality patient care.

The Alternative Quality Contract is a global payment system similar to ACOs implemented by Blue Cross/Blue Shield of Massachusetts. It is composed mostly of independent primary practices connected through the ACO, with outside specialists receiving referrals as needed. PCPs in the Alternative Quality Contract demonstrated a shift in their referral patterns toward specialists who maintained data on patient outcomes and effectively communicated these outcomes to the PCPs.³⁴ With a monetary incentive, PCPs will refer to outside specialists who can provide the best value (ie, quality patient care at the lowest cost).

Alternatively, an oncology specialist may join a hospital-led or fully integrated ACO as an employee with a stake in the organization. Current managed-care groups, like Intermountain Healthcare in Utah, provide a model for how specialists in fully integrated ACOs can operate. Intermountain Healthcare has developed protocols that provide a framework for making clinical decisions. The clinical protocols highlight specific best practices for high-intensity clinical conditions (ie, conditions with high costs per individual patient case). Protocols are based on regional specialist opinion and subject to regular review by specialty panels.³⁵ Oncology specialists have to be prepared to develop and follow protocols tailored to their individual communities.

Ultimately, the ACO model is tilted toward reliance on PCP decision making, and many specialists will be wary of losing independence in caring for their patients. Whether as a consultant to a PCP-led IPA or as an employee of an integrated group, the oncology specialist will rely on a PCP gatekeeper in most ACOs. Oncologists may feel pressured to curb the use of costly drugs and expensive procedures. New treatments from across all branches of oncology—from proton therapy to hyperthermic intraperitoneal chemotherapy to sipuleucel-T—will now reflect directly on the PCP.⁶ With an eye toward helping hold down costs, oncology specialists will need to be more proactive in demonstrating the value of specialized care to PCPs.

Specialty ACOs and Medical Homes

ACOs have recently been touted as the end of traditional insurance companies in featured editorials and print media across the country.³⁶ Perhaps in an effort to get ahead of the curve, private insurance companies are forming ACOs in partnerships with physicians' groups and hospitals more rapidly than the Medicare Shared Savings Program and are open to more experimentation with the model. One notable development of this is the formation of an oncology-specific ACO in southern Florida. The organization is a partnership between Florida Blue, a Blue Cross/Blue Shield insurance company, Baptist Health South Florida hospitals, and Advanced Medical Specialties, a Miami-based oncology practice.³⁷ Although Medicare does not allow specialty-led ACOs in its program, this private ACO may provide a glimpse into the future of oncology practices.

Oncology is less likely to be an organizational priority for cost savings in a PCP-driven ACO. In a recent survey of ACO chief executive officers, oncology care was not mentioned as a major target to reduce costs.³⁰ However, Advanced Medical Specialties and Baptist Health plan to reduce their costs by focusing on controllable costs of hospital care among oncology patients and share the savings with Florida Blue. This includes using evidence-based chemotherapy pathways, reducing emergency room use, and improving coordination of care among specialists.³⁷

Notably, the approach taken by the oncology ACO is similar to the approach of the oncology medical home. Oncology medical homes provide care to patients with the goals of reducing emergency room visits, reducing inpatient admissions, and using hospice care.³⁸ The oncology medical home provides a promising option for oncologists to dip their feet into the ACO movement. It can serve as a blueprint for starting an oncology-specific ACO with private insurers, similar to the South Florida oncology ACO. The oncology medical home is also more likely to be a specialist collaborator with a PCP-driven ACO because of demonstrated shared values. A recent study of an oncology medical home in eastern Pennsylvania showed almost $1 million in savings each year per physician in the group.^39,40 This can translate to significant savings for an affiliated ACO. The movement toward the oncology medical home dovetails the ACO movement and provides a foundation for future collaborations.

Future Trends in 2013

Last year, Barkley³⁰ published “Trends to Watch for in 2012” regarding the future of oncologists and ACOs. With the fate of the Patient Protection and Affordable Care Act no longer in question and the ACO movement gaining steam, we suggest similar future trends to watch for in 2013:

1. Watch for greater experimentation with reform in payment to oncology specialists. In addition to adherence to treatment pathways, factors like decreased hospitalizations, improved patient satisfaction, and end-of-life counseling may become tied to oncology specialist payments. For example, insurers are already experimenting with tiered pathways, in which the level of patient benefits increases with higher copay add-ons.

2. Watch for more large-scale oncology provider groups forming with the purpose of working as specialist collaborators with PCP-driven ACOs. This includes oncology-specific ACOs and medical homes, which can attest to sharing many of the same values as ACOs. Oncologists may have the onus to point out the value of some activities to patients and even induce patients to demand from their PCPs the inclusion of some elements of high-cost care.

3. Watch for a change in referral patterns toward oncology specialists who use health information technology to collect and analyze patient data. Newly formed ACOs are still determining how to meet CMS requirements on reporting outcomes and will need detailed patient data to meet quality measurements.

4. Watch for numerous ethical and legal arguments about quality of care to be voiced by different stakeholders. Pathway-driven medicine may lead to bare-bones cancer care, significantly reducing the universe of treatment approaches used by specialists. This homogenization of care guidelines with emphasis on proven value and parsimony is likely to make oncology a dramatically different field over the next several decades.

Appendix

Table A1.

ACO Required Measures of Quality

Quality Domain	No. of Measures
Patient/caregiver experience	7
Care coordination/patient safety	6
Preventative health	8
At-risk populations—diabetes	6
Hypertension	1
Ischemic vascular disease	2
Heart failure	1
Coronary artery disease	2
Total measures	33

Abbreviation: ACO, accountable care organization.

Figure A1.

In one-sided model, 50% of savings go to accountable care organization (ACO), with no additional risk of penalty. In two-sided model, 60% of savings go to ACO, with penalties for spending over projection. All ACOs must transition to two-sided risk model after the initial 3-year contract. Data adapted (Dorn SD: Clin Gastroenterol Hepatol 9:750-753, 2011).

Authors' Disclosures of Potential Conflicts of Interest

The author(s) indicated no potential conflicts of interest.

Author Contributions

Conception and design: All authors

Administrative support: Roger M. Macklis

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors