Searching for the Value of Accountable Care Organizations in Cancer Care

Precis

The impact of value-based healthcare payment programs on cancer care is not well understood. As Accountable Care Organizations and other alternative payment models continue to grow, understanding their intended and unintended effects will be essential to refining these policies.

Accountable care organizations (ACOs) represent the major policy effort of the Affordable Care Act to transition medical care towards a value-based payment system. In these alternative payment models, groups of health care providers agree to undertake responsibility for the health care quality and spending for a particular population. If ACOs achieve preset quality and spending reduction benchmarks, they receive bonus payments. The largest Medicare ACO program is the Shared Savings Program, which serves 9.7 million beneficiaries through one of 428 participating ACOs.¹ Early evidence suggests that these ACOs deliver quality care and have achieved modest savings, particularly through reductions in spending for inpatient and post-acute care (e.g., rehabilitation and skilled nursing facility care).^2,3 Thus far, only sparse evidence is available that assesses the impact of ACOs on oncologic care.

In part, the modest results of ACO policy thus far may be a consequence of the primary care focus of Medicare ACOs. Patients are attributed to Medicare ACOs based on whether they receive the plurality of primary care services from ACO participating providers.⁴ Additionally, ACOs are judged on primary care focused quality measures in 4 domains: patient and caregiver experience, care coordination and patient safety, preventive health (including cancer screening), and at-risk populations (specific measures for patients with diabetes, depression, hypertension, and ischemic vascular disease).⁵ While these quality metrics are applied to all ACO patients, it is not clear if primary care physicians and specialists are equally engaged in achieving these benchmarks. Indeed, an early analysis of Medicare ACOs found that ACOs with more primary care physicians and fewer specialists performed better on composite measures of disease prevention and wellness screening.⁶ As a result, it is not clear to what extent ACOs are motivated to influence the spending and quality of cancer care, which is driven primarily by specialists.

In this issue of Cancer, Resnick and colleagues explore this important paradigm by evaluating the impact of ACO participation on one of the ways primary care physicians directly impact cancer care: screening for breast, colorectal, and prostate cancers.⁷ They found small, but statistically significant, differences in screening practices among ACO and non-ACO patients. In the year after ACO initiation, patients attributed to an ACO had a 1.8% reduction in the use of breast cancer screening and a 3.4% reduction in prostate cancer screening, relative to non-ACO patients. Additionally, these patients had a 2.4% relative increase in colorectal cancer screening as compared to those not in an ACO. The authors confirmed the robust nature of these findings by employing both more sensitive and more specific algorithms to define screening.

While these differential effects of ACO implementation are small, they account for changing background trends in screening and may, as the authors note, result in significant changes to the population of patients with screen-detected cancers. Furthermore, these findings may underestimate the full impact ACOs will have on cancer screening. The ACO post-contract period evaluated in this study included the first year after ACO initiation. Existing evidence suggests that ACOs progress in their ability to reduce spending and improve quality with more years of experience in the program.¹ Additionally, the first year after the start of an ACO contract required reporting of breast and colorectal cancer screening, but ACOs were not evaluated on this metric until the third year of ACO participation, which is not captured in the current analysis.⁵ Therefore, the significant findings noted in the first year of ACO participation in this study are intriguing and may increase in magnitude in future years of the Shared Savings Program. Longer term follow up will also be needed to assess the impact of these differences in screening on cancer diagnosis and treatment as well as overall spending.

Despite its inclusion as a quality measure, ACO participation was associated with a significant decrease in the use of breast cancer screening. As the authors posit: this may represent a reduction in breast cancer overscreening. More work is still needed to understand how changes in volume of screening interact with appropriateness of screening. While colorectal and breast cancer screening are included within the performance measures in the Medicare Shared Savings Program, prostate cancer screening is not.⁵ Notwithstanding the U.S. Preventive Services Task Force’s recent draft recommendation for shared decision-making, the debate regarding the value of prostate specific antigen (PSA) screening for prostate cancer continues. As is evident in the overall trend found in this study, PSA screening in the Medicare population has decreased considerably from 2007 to 2014. It is not surprising that ACO implementation resulted in an even greater reduction in prostate cancer screening, given the ongoing debate, unclear clinical benefit, and considerable spending that follows a prostate cancer diagnosis.

As the authors astutely point out, however, it is essential that future work consider the impact of these screening trends in diverse populations. There is considerable evidence that African-American men are at a higher risk of prostate cancer mortality than the general population and may be better served by more screening.⁸ Similarly, there is evidence for biological differences, limitations to access, and poorer outcomes for black men and women with colorectal and breast cancers.^9,10 As we better understand the impacts of ACO participation on cancer screening trends, we must ensure that the one-size-fits-all quality measures and focus on constraining costs do not worsen existing disparities in the delivery of high quality care.

The current study and others have demonstrated the potential for ACOs to impact cancer care.^11–13 Thus far, however, this impact seems confined to areas in which primary care providers can effect change. This is particularly noteworthy as specialists are often the gatekeepers of expensive diagnostic tests and treatments in oncology. While ACOs may be able to promote changes in the appropriateness of cancer screening by primary care providers, their ability to improve the quality and spending associated with specialty oncologic care may be limited. Early analysis of Medicare Shared Savings Programs ACOs found that participation in an ACO did not lead to significant improvements in mortality, complications, readmissions, or length-of-stay after major cancer surgery.¹⁴ Similarly, ACOs did not impact the large scale treatment or spending for men with prostate cancer.¹¹ In part, these findings may reflect the lack of surgeon and specialist engagement in ACO programs. Resnick and colleagues have previously demonstrated that fewer than a quarter of surgeons in the US participated in an ACO in 2015.¹⁵ Despite the lack of formal specialist participation, ACOs may still be able to promote value in oncologic care. Specialists who do not participate in an ACO may, nonetheless, practice in a network with ACO-participating clinicians. Thus, primary care providers in an ACO could attempt to influence specialty care by directing referrals to specific providers thought to provide high quality care, with less spending and improved care coordination.¹⁶

The development and dissemination of new technology is a major driver of health care spending growth in the United States. In no area is this more evident than oncology, where surgical devices, genomic testing, and novel immunotherapy agents are being aggressively pursued. Accountable care organizations and other value based payment models have the potential to change the paradigm of technology adoption in oncology by altering the incentives of hospitals and health care providers. Under a fee-for-service system, health care providers are incentivized to increase the number of patients treated and the volume of treatments delivered. In this setting, new devices, advanced imaging modalities, and expensive therapeutics are adopted and marketed to provide cutting edge options for patients and also to help hospitals and physician groups generate a competitive advantage. As payment models shift towards capitation (i.e., holding health care providers responsible for the costs of care), hospitals and provider groups are increasingly incentivized to constrain costs and limit access to treatments that do not offer sufficient value.

It is evident that the transition to a value-based model for health care reimbursement will be gradual. Despite the considerable potential of Medicare’s ACO programs, early spending reductions and quality improvements have been modest.^2,17 Nevertheless, oncology is an important area for the development of value-based initiatives for the Centers for Medicare & Medicaid Services, as evinced by recent payment programs specifically targeting the delivery of cancer care (e.g., Oncology Care Model and Shared Savings Program). Additionally, the Medicare Access and CHIP Reauthorization Act (MACRA) mandates participation in either an advanced alternative payment model or the Merit-based Incentive Payment System (MIPS) for all providers treating more than a nominal number of Medicare patients. This policy will further promote the transition to ACOs and other value-based payment models. In light of this ongoing expansion of the ACO programs, it is increasingly essential that we better understand the intended and unintended effects of this push towards financial stewardship in the care of patients with cancer.