Is a network of doctors and hospitals that shares responsibility for providing care to patients?

Accountable care organizations

An ACO is a group of providers (e.g., hospitals, physicians, and others involved in client care) that work together to coordinate care and manage chronic disease for Medicare beneficiaries.17 ACOs should be client-centered and involve patients in making decisions about their care. ACOs are expected to control the growth of costs and improve the quality of care.

The ACA of 2010 provided funding for ACOs to serve Medicare beneficiaries. Currently, there are over 250 Medicare ACOs and an equal number under private insurance. An ACO can operate within one health-care organization (e.g., an integrated delivery system) or across several organizations. ACOs receive higher payment if they achieve quality goals, and they share in the savings if costs are controlled. The payment received by an ACO is shared across all providers.

ACOs are expected to face several challenges.50 First, ACOs made up of separate organizations will need to negotiate responsibilities and payment and information exchange processes. ACOs within an integrated delivery network will have fewer difficulties with this aspect of implementation. Second, payments are based on outcomes, not on volume, as has been the case. Therefore, ACOs will need to change their focus from treatments to the overall health of their clients. Third, the metrics for evaluating ACO performance will likely need to be modified as we gain better understanding of the processes. Finally, ACOs are a new form of health-care delivery, and providers will need to learn how to work within the system. Because we have limited experience with this form of delivery, the learning curve is likely to be steep.50

Prospects for Accountable Care Organizations

ACOs are an organizational innovation created as part of the Medicare Shared Savings Program of the Patient Protection and Affordable Care Act that was signed into law by President Obama in 2010. Although ACOs are only a small part of a huge piece of legislation, they have attracted a great deal of attention from policy-makers, physicians, and managers.

ACOs are a network of hospitals and providers that contract with the Center for Medicare and Medicaid Services (CMS) to provide care to a large bloc of Medicare patients (5000 or more). The contracts, which last for 3 years, create a single risk-bearing entity with incentives to control costs. ACOs that come in under their specified cost benchmarks earn a fraction of the savings. To receive these payments the ACO must also meet stringent standards on 65 quality indicators that reflect patient and caregiver experience, care coordination, patient safety, preventative health, and health of at-risk frail and elderly populations.

It is interesting to consider the ACO experiment from the perspective of organizational economics. For the statistical reasons we discussed in the Section on principal-agent models ACOs must enroll large numbers of Medicare patients in order to generate reliable measures of savings. But, as emphasized, implementing pay for performance in large groups creates free-riding problems that can dramatically weaken incentives. Put differently, if the ACO is comprised of independent contractor physicians connected only by a common hospital and a common incentive plan, they are unlikely to achieve the desired changes in provider behaviors. Selection, socialization, training, and careful job design are what gives a large organization the ability to influence the behavior of physicians in large groups. If these elements are missing, it is hard to see ACOs having much effect on the way healthcare is delivered.

To achieve savings, the ACO has to manage the capabilities of hospitals and the primary-care physicians who make up of the ACO. The most straightforward way to manage these very different capabilities would be for hospitals to simply employ physicians, but as discussed there are historical, legal, strategic, and sociological obstacles to achieving this goal. Simply purchasing physician practices, as many hospitals, and PHOs did during the 1990s, will not do the trick, but it may not be necessary for ACOs to employ all their primary-care physicians. Some organizations appear to be able to incorporate a significant number of nonemployed physicians into ACO-like arrangements and this offers some hope for expanding the range of hospital–physician coordination. A critical element in these organizations is to build legitimacy among independent physicians by making them part of the governance of the organization.

Incorporating specialists into the ACO will be challenging because specialists are not required to limit themselves to a single ACO. The economic model of referrals suggests that ACOs can reduce referrals by introducing training and computer-assisted decision support that make it easier for generalists to substitute their own decisions for those of the specialists. It may, for example, be better to train primary-care physicians to treat rashes and acne rather than sending every case of rash or acne to a dermatologist. However, the vast explosion in medical knowledge implies that there are limits to the substitution of generalist for specialist care. In this case, it may be that efficiently managing referrals to specialists will entail bringing some specialists into the ACO. Keeping these specialists fully occupied will also exert upward pressure on the optimal scale of ACOs.

Given their size, it is likely that free-riding issues will cause ACOs to operate with under-powered incentives, i.e., with incentives that are too weak, by themselves, to elicit meaningful changes in behavior. From this perspective it is helpful to think of the ACO's incentive problem as analogous to the provision of effort when effort is a public good. The experimental literature on public goods provision suggests that the effects of incentives on public good provision depend critically on the ‘meaning’ agents give to the incentive. Well-designed incentives should communicate that they are intended to achieve a socially beneficial outcome rather than threatening individual autonomy or sense of justice. Extending this logic to the case of intrinsically motivated physicians; managing ACOs likely involves paying careful attention to assigning meaning to the payments, but it is unclear if this meaning is more easily constructed within conventional employment relationships or within hybrid organizations in which doctors participate under looser arrangements. Given the medical profession's long history of battling to preserve its status as an autonomous and learned profession, low-powered incentives in ACOs built on a hybrid organizational form might be workable. However, conventional organizations may have greater opportunities to train, screen, and socialize for physicians who might respond well to low-powered incentives.

To the extent that successful ACO's have organizational capabilities that rely on training, screening, socialization, and constructing the ‘meaning’ of incentives they likely also involve relational contracts. Relational contracts are based on informal trusting arrangements whose credibility is enforced by the continuing value of the relationship between parties. The great advantage of relational contracts for ACOs is that they can complement more formal relationships such as those involved in pay for performance. Incentives that would be under powered in the sense of a principal–agent model may be quite a bit more effective if performance this period determined the continuation of a valuable ongoing relationship. Relational contracts can also be used to reduce some of the distortions of high-powered formal incentives.

Taken together, our analysis suggests that as a policy intervention, ACOs are likely to have the biggest effect where care is already integrated. Advocates of ACOs know this and see ACOs as emerging from five different practice arrangements. In order of ease of implementation these are: integrated delivery systems that combine insurance, hospitals, and physicians; multispecialty group practices; PHOs; IPAs, and virtual physician organizations.

List of Acronyms and Abbreviations

ACO

Accountable Care Organization

Association of Cancer Online Resources

Adverse drug events

Agency for Healthcare Research and Quality

Application Programming Interface

Clinical Decision Support

Comprehensive Health Enhancement Support System

Confidence interval

Electronic health record

Health Information National Trends Survey

Health information technology

Health Information Technology for Economic and Clinical Health

Health maintenance organization

International Business Machines

Institute of Medicine (now referred to as National Academy of Medicine)

Light emitting diode

Morbidity and Mortality

National Cancer Institute

National Oceanic and Atmospheric Administration

Odds ratio

President’s Council of Advisors on Science and Technology

Primary care physician

Physician Data Query

Patient reported outcome

Surveillance, Epidemiology, and End Results

US Preventive Services Taskforce

Veteran’s Administration

20.3.2 Coevolution of Payment Reform and HIT

The ACO model is viewed currently as the preferred approach to achieving the triple aim, namely, improve individual and population health, reduce incremental costs, and improve health care quality. Recently, CMS launched this approach to specialty care under the authority of the Affordable Care Act with the Oncology Care Model as one key area [41]. Given that historically oncology practices have been responsible for dispensing chemotherapeutic agents, CMS is seeking to engage oncology practices in innovative experiments testing new payment models for achieving higher performance for enhanced chemotherapy administration. The major aims for this model are to enhance the quality and coordination of oncology care while simultaneously reducing the costs to Medicare. Private payers are encouraged to participate with Medicare to broaden incentives for care transformation to occur at the practice level.

Central to the ultimate success of this approach is the ability of practitioners to use the most current evidence at the point of care and to engage in shared decision making with patients. Both of these processes require the effective use of HIT that is well designed with provider and patient input, easily accessible, and fits seamlessly into the clinical encounter. Many of the chapters in the book highlight the ways in which HIT can optimize evidence-based oncology care delivered by providers, and enable meaningful shared decision making between patients and providers. Despite the many challenges that remain, the future for cancer care is hopeful. The final section below illustrates the many ways in which HIT and informatics can help the various stakeholders achieve a higher quality health care system.

Accountable Care Organizations

Another provision of the ACA is the creation of Accountable Care Organizations (ACOs) which are groups of doctors, hospitals, and other healthcare providers that volunteer to provide coordinated high-quality care to their Medicare patients [24]. The Shared Savings Program from CMS incentivizes ACOs to strive to provide patients with coordinated care at the “right time” and “right place” to avoid unnecessary duplication of services and help prevent medical errors. The Shared Savings Program requires ACOs to meet quality metrics in four areas. In 2013, the metrics include 33 quality measures in the four areas of patient and caregiver experience, care coordination and patient safety, preventative health, and at-risk populations (including patients with diabetes, hypertension, ischemic vascular disease, heart failure, and coronary artery disease) [25]. When an ACO succeeds in meeting the quality metrics, it receives back a portion of the Medicare savings.

Emerging Dominance

With the focus of the ACA on creating accountable care organizations (ACOs), we have seen significant consolidation among providers and between providers and payers across the country. This consolidation has led to significant dominance among these organizations in geographic markets, where there may be only three or four choices of healthcare provider systems. At this time a competitive market does not exist in healthcare services like it does in financial services, where no US bank has more than a 10% market share. The lack of market dominance and power in banking, forces banks to cooperate and share information. With dominant market positions emerging across all healthcare markets, providers are less willing to share information, and they view sharing as a competitive threat rather than an advantage.

The Importance of Patient Engagement in Care Delivery and Health Management

As providers continue to experiment with new healthcare delivery models, from Accountable Care Organizations to Patient Centered Medical Homes, the concepts of patient-centered care and patient engagement have become increasingly important in recent years. This interest reinforces the increasingly important roles that patients and families are assuming as more active and empowered consumers of healthcare services (Rozenblum et al., 2015b). One of the most prominent spotlights on patient-centered care appeared in the 2002 Institute of Medicine (now the National Academy of Medicine) report, “Crossing the Quality Chasm: A New Health System for the 21st Century,” which highlighted the importance of incorporating patients’ needs and perspectives into care delivery (Institute of Medicine, 2001; Berwick, 2002). Triggered in part by this report, patient-centered care and patient engagement have become a central focus in the national healthcare discussion within the United States and elsewhere (Institute of Medicine, 2001; Jha et al., 2008; US Government, 2009). As a result, meaningful collaboration with patients and families and their active participation in the care process and decision making are now considered key elements of healthcare quality and delivery.

Although most models of patient-centered care place an emphasis on the importance of the patient’s and family’s engagement in their care, there is a lack of consistency in how the concepts of patient and family engagement are defined (Gallivan et al., 2012). The term “patient engagement” is sometimes used interchangeably with “patient empowerment,” “patient partnership,” and “patient activation.” While these terms are indeed related, they are not synonymous. Some definitions of patient engagement focus on individuals’ behavior relative to their health care, while others focus on the relationship between patients and healthcare providers (Carman et al., 2013). Recently, a group of researchers from four institutions in the United States defined patient and family engagement as “an active partnership between health professionals and patients and families working at every level of the healthcare system to improve health and the quality, safety, and delivery of health care. Arenas for such engagement include, but are not limited to participation in direct care, communication of patient values and goals, and transformation of care processes to promote and protect individual respect and dignity” (Brown et al., 2015). This definition further elaborates that, “Patient and family engagement comprises five core concepts: collaboration; respect and dignity; activation and participation; information sharing; and decision making” (Brown et al., 2015). In this chapter, we have chosen to use the term “patient engagement” to denote a broader concept that includes patient empowerment, patient partnership, and patient activation.

Part of the impetus for supporting patient-centered care and patient engagement initiatives is the growing body of evidence that they often lead to greater patient satisfaction, improved clinical outcomes, health service efficiency, and improved health-related business metrics (Carman et al., 2013; Manary et al., 2012; Glickman et al., 2010; Isaac et al., 2010). In particular, studies have indicated that patient engagement and shared decision making leads to improvement in self-management and treatment adherence (Hibbard et al., 2007, 2013; Mosen et al., 2007; Greene and Hibbard, 2012; Remmers et al., 2009). Other benefits associated with patient engagement have been found in the form of more efficient health services utilization of diagnostic tests, referrals, emergency department visits, and hospital attendance (Greene and Hibbard, 2012; Remmers et al., 2009). Consistent with this idea, higher levels of patient activation have been shown to be correlated with lower predicted costs per head (Hibbard et al., 2013).

Over the last decade, many healthcare organizations have developed approaches to enhance patient engagement in care delivery. Yet, despite these efforts, difficulties in transforming their organizational culture from provider-focused to patient-centric have left many provider organizations short of achieving meaningful patient engagement and high patient experience scores (Jha et al., 2008; Hibbard and Cunningham, 2008; Bates and Wells, 2012; Rozenblum et al., 2015a). There are number of potential challenges that might explain why many healthcare organizations are not achieving this goal, such as change management, consistency in practice, resource management, healthcare professional engagement and buy-in, and staff education and training (Aboumatar et al., 2015). Consequently, healthcare organizations need to develop more effective approaches to engage patients as a strategic priority.

Questions for Discussion

1.

Compare and contrast the various forms of HIE. Which form might be most advantageous for an Accountable Care Organization? Which form might be best for a public health agency?

Which lesson from the CHINs and LHIIs is most important to modern HIE initiatives?

Do the requirements for HIE in “meaningful use” program make building a value proposition for HIE easier or more difficult? Why?

Under what conditions would it make sense to use a patient-controlled form of HIE as opposed to a federated or centralized approach?

How do the definitions of HIE vary around the world? Why might HIE be implemented differently in a country outside the United States?

Background and Need for Change

We started this book in its initial chapters (see particularly Chapter 2) with the need for changes in healthcare delivery, and especially the need for predictive analytics in medicine. We restate that need for change and point to the fact that change is happening already.

To reiterate, in a recent online guide (AMA, 2013, p. 2) for medical doctors who are considering joining ACOs (Accountable Care Organizations), the American Medical Association (AMA) set out the business case for why the existing medical fee for service model must change in the United States. In that document it was said:

“In 2008, health care expenditures in the US exceeded $2.3 trillion with costs per resident at $7,631 per year” (Henry J. Kaiser Family Foundation, n.d., cited in AMA, 2013). One other source gave the amount as $7,538 (Henry J. Kaiser Family Foundation, 2011, Exhibit 4A).

“In 2009, the percentage of gross domestic product (GDP) spent on health care was 17.3 percent. In 2008, it was 16.2 percent, making the increase to 17.3 percent in 2009 the largest one-year increase since 1960 (Truffer et al., 2010, cited in AMA, 2013, p. 2).

“The country closest to the United States in health care expenditures is Germany, where 11.1 percent of its GDP is spent on health care” (Truffer et al., 2010 cited in AMA, 2013).

The United States has poor health outcomes compared with other developed countries: the effectiveness of healthcare delivery, despite the high costs, is far less than in other advanced economies. One of the primary needs to help correct this is the application of predictive analytics and decisioning, as eloquently stated by Dr. David Dimas of the University of California, Irvine (personal communications, 2012 and 2013):

The Importance of Predictive Analytics in Healthcare: Current practice in healthcare is often based on more traditional statistical analysis (p-value), which can lead to treating a person as a “mean” of a population. An individual’s demographics, health history, comorbid conditions and genetics may cause him to react differently to a particular drug or treatment. As a result, predictive analytics applied to medical data can help develop treatments that are more in tune with the individual patient.

(David Dimas, PhD).

Dr. Dimas, trained as an engineer, and seeing the value of predictive analytics, established one of the first graduate extension Predictive Analytics Certificate Programs in the USA. A course in Predictive Analytics for Healthcare & Medical Research will be added to this program, projected for 2014. The need for people to be trained in predictive analytics is so great that this program has a waiting list of students – and the students are not only the “young twenties” generation; there are also MD and PhD holders and seasoned IT professionals with 20 years of experience taking this program. This speaks to the perceived need in our society for people trained in this field.

Impact of the Changing Consolidated Health Care Landscape

The implementation of the ACA likely will serve to further accelerate consolidation in local markets because it includes provisions that encourage hospitals and physicians to integrate to facilitate the formation of Accountable Care Organizations (ACOs). ACOs provide the full spectrum of inpatient and outpatient services. By giving physicians and hospitals joint responsibility for patient care, the ACO model fosters alignment and perhaps encourages more consolidation to create a larger, more diversified patient base.

The expansion of coverage, as provided for in the ACA, is to be financed in part by a slowdown in Medicare payment rate updates relative to predicted trends, which may lead to further increases in prices charged by hospitals to private insurers. In addition to the provisions included in the ACA, market consolidation also is being driven by a confluence of other forces, including decreasing reimbursement both by government and commercial payers, changing attitudes of physicians specifically related to employment models, building additional leverage with payers, and the potential to reduce costs. Strategy&, formerly Booz and Company, predicts that at least 1000 of the 5000 existing hospitals could look for merger opportunities in the next five to seven years. Whether mergers and acquisitions can reduce costs and improve quality, or whether they simply result in increased costs, remains the major question.

The future of medical care likely resides in large, integrated health systems that have the ability to manage population health and take on risk in collaborative arrangements with payers. The case for hospital consolidation has shifted away from the revenue side of the equation to improved efficiency and reduction of expenses, though the emphasis has shifted for many systems away from inpatient beds to risk products, clinical integration, and large outpatient centers. No longer is the centrality of medical care the inpatient setting; clearly the shift is not only to the outpatient setting, but more and more is moving into the home. Enlightened health systems and academic health centers are preparing for this reality and recognizing that margins, generated mainly from the inpatient setting, are under severe pressure and thus are looking for ways to deliver efficient care in the most appropriate setting in order to reduce cost.