Welcome to Value-Based Care, Healthcare Delivery. This is Lecture c. In this lecture, we describe how accountable care organizations, or ACOs, are evaluated and performing. We also introduce potential legal, ethical, and privacy issues for ACOs. The objectives for this lecture, Accountable Care Organizations Lecture c, are to describe the metrics used to evaluate ACO performance. Discuss how ACOs are performing. Describe potential legal and ethical issues facing ACOs, and identify privacy issues in ACO models of care. To summarize our discussion of ACOs, it is becoming increasingly apparent that the provision of high quality care involves coordinated longitudinal care. And that means accountability and performance assessment at the individual provider level are no longer sufficient. Especially when gaps in care are the result of poor coordination of care. ACOs will need new processes and tools and individuals with new skills to coordinate and improve care. Individual providers often don't have the capital resources to invest, nor do they have enough responsibility for the whole patient to look at the big picture with regard to technology investment within the entire system. By connecting physicians to a larger group, they may be better able to connect and coordinate on behalf of their patients. Although these new models of care are just beginning, we need to give some consideration to how we can measure their impact. One approach is to use a logic model, which is often used for program evaluation as well as planning. By documenting the context, environment, and inputs into the program, one can anticipate the intermediate outcomes and ultimately the impact of the program. There are many ways to assess the impact of ACOs. Fisher employed the logic model concept to analyze ACO impact. The figure on the screen provides a model for understanding different aspects of ACO performance. The ultimate measure of success of the ACO approach will be assessing the three impact areas noted in the right-hand column in the figure. Improved access and patient experience, improved health and functioning of members and reduced costs of care, these reflect what is known as the triple aim. The second column to the right lists three important intermediate outcomes of ACOs. The degree of health information technology achieved in ACOs, the improvement in care processes and the degree of care integration achieved by the ACO. To achieve the desired impact and outcomes of ACOs, it is becoming increasingly evident that ACOs will need to do the things identified in the middle column of the figure. That is, implement health information technology and health information exchange across programs and across the different entities that make up the ACO. Achieve robust data sharing by providers and with payers, and develop the infrastructure to support public reporting on quality and spending performance. This section will focus on the potential ways ACO performance will be measured. In this case, we will describe how ACOs will set spending targets and quality of care benchmarks. Recall that these two categories of measurement are central to the values-based concept of the ACO. This is different from traditional payment systems and incentives that had been tied to the quantity or volume of care provided. In most cases, ACO's spending performance is focused on measures of the total cost of care for a defined set of patients. In other cases, ACO's set spending targets related to a reduction in some measure, such as the per member per month growth rate in spending. The spending targets are risk adjusted, which means that they are adjusted based on the chronic conditions and other health measures of the patients served by physicians in the ACO. We are just now beginning to see how ACOs are performing. A study published in 2014 looked at the largest ACO program, The Medicare Shared Savings Program ,and found mixed results among 220 ACOs. About half of the ACOs met quality benchmarks and kept spending below targets. 29 of these ACOs saved more than 2% and were eligible for sharing savings. The ACOs received $126 million in savings and generated $128 million in total savings for Medicare. Slightly more ACOs, 60, did not meet their quality benchmarks or did not reduce spending enough to share in savings. These ACOs were able to keep total spending under the target level. In subsequent years the results have been fairly consistent, with overall spending reduced and, for the most part, quality indicators improving. There are also mixed results in the Pioneer ACO Program. In this program, there are savings generated and in some cases shared. For instance, in year 1, the program generated $147 million in total program savings. In year 2 of this program, $96 million in total program savings was generated and 11 ACOs received bonuses. However, 3 of the ACOs faced penalties. In each year, pioneer ACOs have left the program, most moving to the MSSP. The quality of care measures used by ACOs in two of the Medicate ACO models include 33 nationally recognized quality measures in the four domains of patient/caregiver experience, care coordination/patient safety, at-risk populations and preventive care. We will now briefly present the measures used to evaluate ACO quality of care performance. For all measures, more detailed descriptions can be found in the documentation Provided by CMS. The Patient/Caregiver Experience domain includes eight indicators that are all collected through patient and caregiver surveys. The measures were developed by the Clinician and Group Consumer Assessment of Health Care Providers and Systems, also called CG CAHPS. ACOs must use CMS approved vendors to administer the surveys. There are ten measures in the Care Coordination and Patient Safety domain. This domain includes some of the most complex measures that will require sophisticated systems for measurement, calculation, and reporting. One example is risk-standardized all conditions re-admissions to hospitals. Re-admission following acute care hospitalizations are costly, and often preventable. Medicare reports that almost 20% of Medicare beneficiaries were readmitted within 30 days of discharge. Others have estimated that reducing national Medicare readmission rates to the level found in top-performing regions could save Medicare $1.9 billion annually. Working together across multiple providers, and focusing on local populations, ACOs should be able to reduce the number of re-admissions for all conditions. Another example in this domain is the percent of primary care physicians in the ACO, who successfully meet meaningful use requirements. Medicare focuses here, because health information technology has been shown to improve the quality of care and the ability of the ACO to coordinate care. The At-risk Population domain includes eight measures focused on ACO patients in the following at risk populations, coronary artery disease, diabetes, heart failure, hypertension, ischemic vascular disease, and mental health. In each at-risk population, there are measures of various patient-care processes and outcomes. These measures are intended to keep ACOs focused on improving the clinical care of patients with these diseases. They are collected through claims data. The final domain of ACO quality measures focuses on preventive care. The measures include things like immunizations and also screenings and followup plans for conditions, such as obesity, body mass index, tobacco use, high blood pressure, and clinical depression. These measures are typically identified through data collected in clinical information systems. The challenges of managing care in an ACO cannot be met without a significant investment in information technology. The increase in adoption of EHR that was facilitated by the HITECH Act of 2009 has paved the way for the kind of technology need for ACOs. In an effort to meet the meaningful use criteria, many providers began to use technology for clinical decision support, health information exchange, and patient engagement. Clinical decision support focused on the quality indicators. Exchange of information across the ACO and engagement of patients through patient portals or other means are all needed to support care management, quality improvement and care coordination and data sharing both among providers and between providers and patients. Since ACOs must also monitor and report on their metrics across a population of patients and across multiple provider entities, a means of aggregating data and facilitating reporting is needed. A data warehouse that ideally combines both cost and clinical data can facilitate that reporting. In addition, there are specialized applications like patient registries, for instance, that would keep track of all patients with a given condition. As well as specialized population health applications that facilitate the monitoring and caring for a population. The American Hospital Association identified several potential legal and regulatory issues associated with ACOs. The heart of the matter is how ACOs develop and share financial incentives and rewards. ACO arrangements could potentially be in violation of several federal laws, including the Medicare ban or self-referrals, fraud and abuse statutes, the Anti-Kickbacks statute, and the Civil Monetary Penalty Law. In addition, nonprofit hospitals will need to make sure ACO participation does not make them out of compliance with the Internal Revenue Service guidelines for nonprofit institutions. As ACOs have developed, many of these issues have been raised and successfully overcome. ACOs may also raise ethical issues for providers. DeCamp and colleagues have outlined some of these issues, including one issue is balancing health care providers' need to achieve cost savings with patients' freedom to choose their health care providers. Hospitals may need to balance the way they establish provider networks and control physician referrals with the rights of patients to choose providers. Focusing on preventing illness may lead to financial conflicts of interest for the organizations. As ACO models develop, quality indicators may give incentives for hospitals to reduce unnecessary readmissions. At the same time, hospitals may still receive more payment with more admissions. These unintended financial effects may lead to difficult decisions for hospital leaders. There is also a risk that in focusing on specific quality measures, less attention will be given to other aspects of quality care that are not part of the required metrics. ACOs will undoubtedly raise new questions about relationships, what to focus on when Medicare requires 33 different quality of care measures, and how benefits, that is savings, are distributed among providers. Strong leadership and governance will be necessary to ensure appropriate responses to these and other ethical issues that may arise as ACOs are implemented fully. DeCamp and colleagues identify specific potential ethical issues facing leaders, clinicians and patients in ACOs. They also describe management strategies to appropriately manage and overcome these ethical challenges. For example, leaders who involve clinicians in the design of shared savings plans may ameliorate the ethical issues of resource allocation and fairness. Similarly, ACOs can monitor clinicians' experiences in the organization, as it relates to potential perceptions of pressure to withhold care, or up-code, in order to improve ACO financial performance. Focused efforts to develop teams of providers, collaboration across primary and specialty care, and communication strategies may also help clinicians manage perceived competition among primary and specialty care. Finally, ACOs can work to use proven patient engagement strategies in the design and operations of ACO programs and services, which would mitigate the conflicts between patient and provider needs. It will be important for ACOs to make sure patient privacy is protected at all times. This will be more challenging than ever as the ACO must share information across many different providers and programs. ACO participants are HIPAA covered entities. That means they must manage protected health information, or PHI, in compliance with HIPAA. ACOs can address potential privacy issues by including privacy and security obligations in all agreements. By updating risk assessment and management plans. By developing a strong breach response plan and by training their workforce on these matters, particularly as they relate to information sharing. If the Health Information Exchange mechanism is managed outside the ACO, the ACO must assure that strong privacy and security protections are in place before agreeing to contract with an outside entity. This concludes Healthcare Delivery Lecture c. In summary, CMS requires ACOs to report on 33 measures in four domains. In terms of assessing how ACOs have performed, there are many measures that can be used. In general, evaluations have found that the overall early performance of ACOs is mixed. Many have achieve both equality and cost savings desired, but some have not. As they develop further, ACOs must work to understand and deal with potential legal, regulatory, ethical, and privacy issues that will arise with this new model of care.
