achieving interoperability for accountable care

Results of a recent survey provide insight into key business intelligence considerations for accountable care organizations as they strive to achieve data interoperability across their provider networks.

In its reporting on the effects of the Affordable Care Act (ACA), the national media has focused largely on the law's extensive reforms of the health insurance market. Little attention, however, has been given to another important feature: the provision establishing the Medicare Shared Savings Program (MSSP) with the intent of promoting the development of accountable care organizations (ACOs). This provision, which generally is well-known only to industry stakeholders, likewise has the potential to dramatically reshape the U.S. healthcare system.

The concept of an ACO as a coordinated network of providers with shared responsibility for providing better care at lower costs is wellestablished, based on the idea of aligning provider financial incentives with patient needs. The MSSP has sparked significant ACO growth in the past several years: There are more than 600 ACOs with a diverse range of underlying payment and care delivery models. Although many ACOs contract with the federal government, approximately one-third are entirely commercial.3

Facing New Challenges

ACOs are at the vanguard of the transition from fee-for-service to value-based care. With new financial incentives and outcome measures at stake, ACO providers require information systems that enable them to apply business intelligence to effectively coordinate care and collect clinical data. Each time a patient transitions to a new provider or provider setting, data need to be collected and analyzed to inform a new set of care decisions. To deliver fully informed care seamlessly, ACOs require business intelligence, which they can obtain by developing the capability to integrate data from disparate sources to construct complete patient histories and generate holistic views of their patient populations.

Integrated data sets allow ACO leaders and clinicians to apply new analytic tools to measure and compare performance. Various forms of health IT, such as electronic health records (EHRs), registries, databases, health information exchange tools, and analytics software, can help an ACO aggregate data and integrate providers and systems across disparate settings into a unified network. The aggregated data then can be used to support clinical decision making, risk stratification, and predictive modeling to support an ACO's efforts to deliver high-quality care safely and efficiently, with minimal cost and optimum use of resources.

These objectives of accountable care, although straightforward in concept, pose many practical challenges. For decades, providers at the institutional, departmental, and practice levels have evolved different processes and systems to meet their specific needs, often without consideration of other stakeholders in the system. Because there has been little, if any, focus across the industry on developing information systems using common standards and protocols for encoding and communicating data, most providers now face enormous obstacles to achieving the interoperability required for care coordination and data aggregation. Different systems in an ACO network often cannot connect to pass information to each other, let alone facilitate the creation of composite data sets for benchmarking and analysis.

A Survey of the Interoperability Problem

When seeking to develop the interoperable systems required to support the collection, sharing, and analysis of data in an ACO, it is necessary to understand what capabilities currently exist. To obtain this understanding, an online survey of ACOs was conducted in July and August 2014 to assess these organizations' capabilities and the challenges of healthcare IT implementation.11 Among the 62 ACOs that responded, survey results demonstrated diverse healthcare IT infrastructure capabilities oriented around population health management. Ninety-five percent of survey respondents cited interop-erability as a key concern.

A preliminary objective of the survey was to understand what technology systems ACOs have in place. The overwhelming majority (86 percent) of ACOs have EHRs. Many also use disease registries (74 percent), data warehouses (68 percent), and clinical decision support systems (58 percent). All of these systems provide important data for managing patient populations.

Through these tools, 95 percent of the surveyed ACOs collect electronic clinical data and post-adjudicated claims data, both of which are critical for understanding quality outcomes and metrics. A smaller percentage of ACOs (63 percent) collect readjudicated administrative, billing, and financial data to support operations and help manage resources and costs.

Collection of data and access to data are different issues. An organization may collect data but not be able to share it with other providers. The survey found that lack of data access was largely viewed as a prohibitive barrier to ACO activities. ACOs need both internal and external data to coordinate care and determine progress toward benchmarks.

Internal sources include the various information systems used by departments and providers within the network, such as labs, radiology, EHRs, and practice management. If an ACO is composed of loosely affiliated providers or the organization has extensive legacy health IT systems in place, these systems may not be interoperable.

External sources include health insurance claims data, community institutional data, and patient-generated data from mobile applications and devices.

A significant challenge is that patients attributed to an ACO are free to receive care at other settings- a common problem known as "network leakage." Thus, to build a complete profile of its patient population, an ACO may need to obtain data from competing organizations.

The following were among the key issues cited by survey respondents.

Data access. Respondents noted the inherent difficulties of acquiring patient data. Not surprisingly, all respondents indicated that lack of access to data from outside their organizations or networks was a significant obstacle. A more surprising finding was that half of responding ACOs also regarded simply gaining access to data inside their organizations as a difficult challenge. In this case, interoperability concerns can make data aggregation difficult. Different systems may use different nomenclatures or protocols to label and transmit data, which need to normalized for the purposes of aggregation. Many health systems must interface with a variety of technology solutions, which increases the complexity of integrating data.

Data quality. Even when data are accessible, respondents are unsure about their ability to use data effectively to support operations. Nearly three-quarters of respondents expressed concerns about data liquidity (i.e., the flow of data throughout a health information system and the availability of data when and where it is needed) and quality, and 88 percent viewed data integration as a challenge. Fewer ACOs based in health systems and independent practice associations had difficulties with data liquidity (66 percent and 60 percent, respectively) than did medical group- and hospital-based ACOs (both too percent).

Perhaps more troubling is the uncertainty about how ACOs are addressing the issue. Health information exchange (HIE) solutions could enable ACOs to share crucial data, yet relatively few ACOs (44 percent) incorporate HIE technology into their health IT infrastructure. Among ACOs that do utilize HIE, about one-third (33 percent) do so through an enterprise solution, which would not usually encompass data from external sources. Even fewer ACOs overall said they participate in the statewide (21 percent) or community-based (17 percent) public HIEs that are more uniquely positioned to aggregate data from across the healthcare ecosystem.

Workforce issues. To achieve data interoperability across their networks, ACOs must invest not only in technology, but also in their workforce. Survey data suggest that as ACOs evolve over time, their staffing needs change. Early-stage ACOs reported having enough staff to collect and analyze data, but the number of ACOs that regard their staffing as sufficient dropped off significantly for those in the intermediate and advanced stages of operation (i.e., those with one to two years of operating experience) before increasing again among mature ACOs (i.e., those operating for more than two years).

Lessons Learned

Given that ACOs represent a first step toward addressing deficiencies of the current healthcare system, it's essential to consider strategies for improving access to and use of data within these organizations. Of the ACOs surveyed, those with 18 or more months of operation reported substantially more advanced capabilities, greater use of data for analytics, and larger performance improvements associated with healthcare IT.

Ina perfect world, entities seeking to form an ACO could implement the necessary health IT components prior to entering into contractual arrange - ments, and clinicians and organizations would have time to familiarize themselves with how the tools work and create new workflows supporting their use. However, real-world data suggest that many ACOs are implementing technologies as they develop their operations. Eighty-eight Mpercent of survey respondents reported that integrating healthcare IT into workflows is a challenge. However, early adopters did have some valuable insights and strategies to share.

Create a shared governance structure to make IT decisions. ACOs involve a number of different stakeholders, systems, and data. Coordinating efforts among these groups requires a sound decision-making process. A shared governance structure and model is critical to ensuring technology and data issues can be resolved in a timely and collaborative manner. Participation early in the process can help break down competitive interests and head off potential integration issues.

Conduct a readiness assessment and gap analysis. It is important to examine the capabilities, strengths, and weaknesses of all the technology and data components that are being used in each participating organization. An effective gap analysis can help groups identify and prepare for technological challenges as they emerge.

Reconfigure the technology infrastructure and processes to support new value-based care delivery protocols. The infrastructure and work processes for most organizations were built around a fee-for service model. It is important for ACOs to establish integrated care delivery processes that match the new clinical integration requirements. Integrated care delivery involves new levels of care coordination, communication, and teambased care. Today, care transitions all too frequently involve a simple referral without any follow-up and with limited communication by the referring clinician. Integrated care necessitates the creation of a feedback loop, so providers can communicate with each other and coordinate better care. Health IT solutions such as secure messaging or clinical portals also can facilitate information sharing and communication between

providers. Technology systems should be built around these new requirements. Systems should have the capability to capture and aggregate data related to clinical outcomes for patient care, as well as population management.

Consider targeting programs around high-risk groups. Some ACOs have had success by stratifying patient populations to identify targeted areas for intervention. For example, diabetes and heart disease conditions have clear evidence-based guidelines. Clinicians can deploy fairly simple analytics tools to identify patients who utilize these healthcare services most extensively and use the guidelines to manage care. Other ACOs have found success applying across-the-board analytics to distinguish what works best. ACOs can use analytics to track and compare performance of service lines in aggregate to identify best practices in treatment. Financial analytics can also help ACOs better manage claims adjudication or identify specific services associated with high costs but poor outcomes.

Develop real-time data-sharing systems. To effectively manage patients at all points in the system, ACOs require technology that supports the real-time sharing of information to all providers in the system, both acute and ambulatory care. In new care coordination models, when all parties involved (e.g., patients, providers, facilities, payers) have real-time access to the same data, providers are able to access critical patient information at the point of care. For example, real-time data sharing could help ensure a primary care provider is notified when a patient is taken to the emergency department (ED), and that the clinicians in the ED have access to tests and notes from the patient's earlier visit to the primary care physician's office.

Ensure privacy and security policies and procedures are in place. Sharing patient health information is the foundation of accountable care. ACO participants can legally share patient health information for treatment, payment, and joint healthcare operations. However, there are fine points that must be navigated in both federal and state privacy and security laws. It is critical, for example, that ACOs have robust policies that ensure compliance with HIPAA, business-associate arrangements, and other federal and state privacy requirements. All patient data should be encrypted and accessible only to parties with proper authorization.

It is important that contractual agreements between providers clarify the policies and procedures related to access to different data sets and systems. Systems that may have been proprietary in the past may now need to link to competing providers to support patient care and population management. ACOs also should have clear policies outlining patient privacy rights regarding use of data, such as research, that falls beyond the treatment, payment, and operations purposes covered by HIPAA. Role-based access and audit trails can help an ACO avoid inappropriate data use.

Assess and address workforce issues expeditiously. Given that workforce issues change over time, mature ACOs will have different needs than start-ups. ACOs should be prepared to adjust their staffing needs as requirements change. For example, when organizations are adopting a new EHR system, there will be a need to significantly increase technology support staff to help clinicians learn new systems. Systems with established EHRs may increase analytics staff to generate aggregate reports, and clinical staff to communicate results and support patient management.

Moreover, as organizations mature, reporting structures will need to evolve. For instance, an early-stage ACO may focus on generating a weekly report of all patients who were readmitted to a hospital to identify areas for improvement or intervention. A more mature ACO may run analytics on its entire patient population to generate predictive risk scores and intervene before a patient is readmitted, which would require staff with more advanced statistical knowledge.

Staff also require continual training on how to use new technologies that are adopted and how to incorporate new treatment protocols into care. Even a technology as ubiquitous as an EHR can be disruptive to a clinician's workflow, forcing the clinician to adapt to a new method of entering data and taking patient histories. Analytics can help identify best practices in treatment and set new protocols for care that may conflict with a clinician's established practices.

Participate in broader interoperability efforts. ACOs should leverage the broader work being done with respect to standards and interoperability. When selecting new technologies, ACOs should pay attention to the standards underlying the system and carefully select products that use nationally recognized standards. Yet technology and innovation are evolving at a rapid pace, and as cloud-based and mobile applications and other new technologies emerge, standards will need to evolve and change. At a minimum, technology experts should keep abreast of new requirements and potential changes. Professionals who want to participate in the selection of specific standards will want to have a seat at the table to ensure national standards are relevant and practical for their systems. Many public and private entities are working tirelessly to refine standards, create consistent specifications for implementation, and develop interoperability solutions. Organizations should seek out these groups and become engaged in these efforts.

One innovative approach to addressing the interoperability hurdle is the development of open and secure application programming interfaces (APIs). By creating common protocols for systems to interact without the need for cumbersome or expensive interfaces, APIs almost certainly will have a critical impact on future interoperability issues. Although APIs in healthcare technology are still largely under development, it is important for ACO administrators to be aware of such solutions to better plan for technology acquisition and future upgrades.

In short, ACOs should participate in conversations about health IT development to help ensure technology regulations do not inhibit the growth of value-based care.

Building a Culture of Transformation

ACOs require an entirely new approach to health care that can be extremely disruptive to long-established practices. The experience of many healthcare providers indicates that legacy systems and fee-for-service processes can be stubbornly difficult to transform. Foremost among the challenges has been the difficulty of integrating disparate care settings and information systems, which poses a significant obstacle to an ACO's population approach to health care. What ultimately is required is a national infrastructure of policies, standards, and rules to facilitate information sharing and use of data for business intelligence. ACOs should strive to foster a culture conducive to collaboration, both within their networks and in the national community at large.

The ability to acquire and properly utilize business intelligence requires collaborative partnerships. For the healthcare industry to transition successfully from fee-for-service to value-based models of care, consensus must be achieved among all parties involved, and reaching this consensus takes efforts at all levels, ranging from legislative and regulatory initiatives to private (e.g., claims and clinical) activities and patient participation.

AT A GLANCE

Based on findings of a recent survey, accountable care organizations should keep eight points in mind as they seek to establish interoperability among their provider constituents:

* Create a shared governance structure to make IT decisions.

* Conduct a readiness assessment and gap analysis.

* Reconfigure the technology infrastructure and processes to support new value-based care delivery protocols.

* Consider targeting programs around high-risk groups.

* Develop real-time data-sharing systems.

* Ensure privacy and security policies and procedures are in place.

* Assess and address workforce issues expeditiously.

* Participate in broader interoperability efforts.

The Goal of Accountable Care: An Overview

Within an accountable care organization (ACO), a group of healthcare providers is accountable for the overall care of an entire group of patients. This type of arrangement represents a radical departure from the episodic approach of the current fee-for-service system, where providers are paid by third parties for the volume of services they provide.

A widely recognized limitation of the fee-forservice system is that providers are financially rewarded simply for delivering additional services, with little consideration given to the quality of care. Although many providers remain strongly committed to delivering high-quality care, the fee-for-service system offers no financial motivation to keep track of patients after an encounter, confer with specialists, follow up on test results, or perform any other of a host of activities that might help to improve a patient's health.

ACOs are intended as a solution to this disconnect between payment and quality. ACOs incorporate novel payment mechanisms (e.g., shared savings and bundled payments) designed to give primary care providers incentives to coordinate care with specialists, hospitals, and post-acute providers to meet quality and cost benchmarks. The financial incentives in an ACO encourage providers to follow patients as they move through the care continuum. Under some ACO contracts, providers' failure to meet measurable quality improvements can result in financial penalties.

A Resource Focused on Interoperability

In 2014, eHealth Initiative launched 2020 Roadmap, a public-private collaborative, producing a series of recommendations for key federal policymakers and the private sector to transform care delivery through data exchange and health IT by focusing on the areas of interoperability, business and clinical motivators, and data access and use. Leaders from ACOs are invited to participate in this effort. For more information about eHealth Initiative's 2020 Roadmap, visit www.ehidc.org.

a. Petersen, M., Paul Gardner, P., Tu, I, Muhlestein, D., Growth and Dispersion of Accountable Care Organizations: June 2014 Update, Leavitt Partners, June 2014.

b. The Landscape of Accountable Care and Connected Health: Results from 2014 National Survey of Accountable Care Organizations, eHealth Initiative and Premier, Inc., Sept. 24, 2014.

About the authors

Jennifer Covich Bordenick, MA, is CEO, eHealth Initiative, Washington, D.C. ([email protected]).

Tracy H. Okubo, PMP, is senior director, HIT Stakeholder Outreach, eHealth Initiative, Washington, D.C. (tracy.okubo@ ehidc.org).

Alex Kontur is manager, research and projects, eHealth Initiative, Washington, D.C. ([email protected]).

Nadeen Siddiqui is a student at the University of Texas-Austin.