Case Analysis: Systems Acquisition

The enactment of the Affordable Care Act (ACA) brought about sweeping legislation intended to reduce the numbers of uninsured and make health care accessible to all Americans. It also ushered in an era in which changing reimbursement and care delivery models are driving providers from the current fragmented system focused on volume-based services to an outcomes orientation. As a result, the health care system now taking shape is one in which value-based payment models financially reward patient-centered, coordinated, accountable care.

Against this backdrop, providers' increasing use of evidence-based medicine and growing capabilities in managing volumes of clinical evidence through sophisticated health IT systems will mean that treatments can be tailored for the individual and interventions can be made earlier to keep patients well. Furthermore, patient engagement is fast becoming a critical component in the care process, particularly in the area of population health management (PHM).

Health care providers' interest in improving population health appears to be increasing because of the sudden ubiquity of the phrase, because many are participating in accountable care organizations (ACOs), and because even hospitals not participating in an ACO increasingly have incentives to reduce their number of potentially unavoidable admissions, readmissions, and emergency department visits (Casalino, Erb, Joshi, & Shortell, 2015).

In this chapter we'll not only seek a common understanding of PHM but also explore how the advent of shared accountability financial arrangements between providers and purchasers of care has created significant focus on PHM. We'll also review the core processes associated with accountable care and examine the strategic IT investments and data management capabilities required to support population health management and enable a successful transition from volume-based to value-based care.

PHM: Key to Success Although the ACO model is still new and evolving, approximately 750 ACOs are in operation today, covering some 23.5 million lives under Medicare, Medicaid, and private insurers. Although not all ACOs have demonstrated success in delivering better health outcomes at a lower cost, many have achieved promising results (Houston & McGinnis, 2016). As such, significant ACO growth is expected. In fact, it is predicted that upward of 105 million people will be covered by an ACO by 2020 (Leavitt Partners, 2015).

Similarly, although the industry's move to value-based payment is also in its early stages, value-based contracts are expected to substantially increase throughout the next decade. CMS has a stated goal that 50 percent of Medicare payments will be tied to alternative payment models by the end of 2018 (US DHHS, 2015). In fact, the projected impact of MACRA, which we discussed in Chapter One, on the adoption of value-based payment models is expected to rival the impact of Meaningful Use on adoption of EHRs. In addition, the substantial payment reform activity at the federal level is paralleled by private insurers' efforts to support value-based payment and new models of care. For example, Aetna expects that 75 percent of its contracts will be value-based by 2020 (Jaspen, 2015).

These trends will accelerate the demand for services and technology that enable health systems and other organizations (health plans, Medicaid, community-based organizations, employers, and so forth) to jointly manage the health and care of populations—either as an ACO or in an ACO-like fashion. Although diverse, these organizations will all have a common need to improve operational efficiency, drive better patient outcomes while reducing the overall cost of care, and effectively engage consumers in managing their health and care.

Although the new reimbursement system is still taking shape, it's clear that population health management will become a required core competency for provider organizations in a post fee-for-service payment environment (Institute for Health Technology Transformation, 2012). Understanding Population Health Management Population health as a concept first appeared in 2003 when David Kindig and Greg Stoddart (2003) defined it as “the health outcomes of a group of individuals, including the distribution of such outcomes within the group” (p. 380).

It is important to note that medical care is only one of many factors that affect those outcomes. Other factors include public health interventions; aspects of the social environment (income, education, employment, social support, and culture); the physical environment (urban design, clean air and water); genetics; and individual behavior (Institute for Health Technology Transformation, 2012). “Improving the health of populations” was later identified as one element in the Institute for Healthcare Improvement's triple aim for improving the US health care system, along with improving the individual experience of care and reducing the per capita cost of care (Berwick, Nolan & Whittington, 2008, p. 759).

Today, population health management comprises the proactive application of strategies and interventions to defined groups of individuals (e.g., diabetics, cancer patients with tumor regrowth, the elderly with multiple comorbidities) to improve the health of individuals within the group at the lowest cost. PHM interventions are designed to maintain and improve people's health across the full continuum of care—from low-risk, healthy individuals to high-risk individuals with one or more chronic conditions (Felt-Lisk & Higgins, 2011). PHM also seeks to minimize the need for expensive encounters with the health care system, such as emergency department visits, hospitalizations, imaging tests, and procedures. This not only lowers costs but also redefines health care as an activity that encompasses far more than sick care, because it systematically addresses the preventative and chronic care needs of every patient—not just high-risk patients who generate the majority of health care costs (Institute for Health Technology Transformation, 2012).

Although population health can also mean the health of the entire population in a geographic area, the population health efforts most health systems and ACOs are undertaking are aimed at providing better preventive and medical care for the “population” of patients “attributed” to their organizations by Medicare, Medicaid, or private health insurers (Casalino et al., 2015). New Care Delivery and Payment Models: The Link to PHM

As we know, historically, there has been a lack of accountability for the total care of patients, the outcomes of their treatment, and the efficiency with which health resources are used. The fact that health care services are paid primarily on a fee-for-service basis has contributed to the fragmentation and lack of accountability. Fee-for-service emphasizes the provision of health services by individual hospitals or providers rather than care that is coordinated across providers to address the patient's needs. Providers are rewarded for volume and for conducting procedures that are often more complex, when simpler, lower-cost, better methods may be more appropriate (Guterman & Drake, 2010).

Value-based care is emerging as a solution to address rising health care costs, clinical inefficiency and duplication of services, and to make it easier for people to get the appropriate care they need. As the federal government continues to test and implement several new payment models designed to achieve optimal health outcomes at a sustainable cost, commercial insurers are also partnering with health care providers in various arrangements that similarly seek to reward value rather than volume of services.

As discussed in Chapter One, two popular models of delivery system reform are the patient-centered medical home (PCMH) and the ACO. The PCMH emphasizes the central role of primary care and care coordination, with the vision that every person should have the opportunity to easily access high-quality primary care in a place that is familiar and knowledgeable about his or her health care needs and choices. The ACO emphasizes the urgent need to think beyond patients to populations, providing a vision for increased accountability for performance and spending across the health care system (Patient-Centered Primary Care Collaborative, 2011). Both models rely on health care organizations and physicians providing coordinated and integrated care in an evidence-based, cost-effective way. This, of course, has significant implications for an organization's ability to manage information effectively.

In conjunction with new models of care are new or modified forms of payment for health care services, which are being piloted in various communities around the nation. These include bundled payments, pay for performance, shared savings programs, capitation or global payment, and episode-of-care payments.

Bundled payments may take different forms such as making a single payment for hospital and physician services instead of separate payments, bundling payments for inpatient and post-acute care, or paying based on diagnosis instead of treatment. Bundled payments are often applied to surgical procedures such as hip replacements. Pay-for-performance (P4P) programs reward hospitals, physician practices, and other providers with financial and nonfinancial incentives based on performance on select measures. These performance measures can cover various aspects of health care delivery: clinical quality and safety, efficiency, patient experience, and health information technology adoption. Most P4P programs, however, are still a bonus to a fee-for-service model (Miller, 2011). An integral part of the ACA, shared savings programs are intended to reward providers by paying them a bonus that is explicitly connected to the amount by which they reduce the total cost of care compared to expected levels.

Capitation or global payment places full risk with the provider organization; the provider is responsible for the costs of all care that a patient receives. An episode-of-care payment system would pay the provider organization a single payment for all of the services associated with a hospitalization or other episode of acute care, such as a heart attack, including inpatient and post-acute care (Miller, 2011).

The revised payments associated with these programs signal the federal government's most all-encompassing effort thus far to distribute risk and hold providers financially accountable for the quality of care they deliver. Although an in-depth discussion of these and other proposed payment reform systems is beyond the scope of this book, the following resources can provide a wealth of detailed information on health care payment reform initiatives: Centers for Medicaid & Medicare Services (www.CMS.gov) Healthcare Financial Management Association (www.hfma.org) American College of Healthcare Executives (www.ache.org) Progress to Date: PCMHs Growing support for the PCMH has arisen across the vast majority of the US health care delivery system to include commercial insurance plans, multiple employers, state Medicaid programs, numerous federal agencies, the Department of Defense, hundreds of safety net clinics, and thousands of small and large clinical practices nationwide (Grundy, Hacker, Langner, Nielsen, & Zema, 2012). Private and public payer initiatives together have grown from eighteen states in 2009 to forty-four states in 2013, and they now cover almost twenty-one million patients. These heterogeneous initiatives overall are becoming larger, paying higher fees, and engaging in more risk sharing with practices (NCQA, 2015).

Because the patient-centered medical home is foundational to ACOs—with ACOs often described as the “medical neighborhood”—the PCMH is likely to gain even greater prominence as ACOs continue to develop in the marketplace (Grundy et al., 2012). Moreover, a growing body of scientific evidence shows that PCMHs are saving money by reducing hospital and emergency department visits, mitigating health disparities, and improving patient outcomes. Examples of specific outcomes achieved by various PCMHs include the following:

Lower Medicare spending More effective care management and optimized use of health care services Improved care management and preventative screenings for cardiovascular and diabetes patients Reduced socioeconomic disparities in cancer screening (NCQA, 2015) Additionally, more than nine thousand primary care practices and forty-three thousand clinicians (doctors and nurse practitioners) across the country have earned the PCMH designation from the National Committee for Quality Assurance (NCQA), the nation's largest credentialing organization. The designation is earned by demonstrating achievement of goals related to accessible, coordinated, and patient-centered care (Olivero, 2015).

Progress to Date: ACOs

In the value-based care world, ACOs are expected to play a leadership role in improving population health—whether participating in contracts with Medicare, Medicaid, or managed care organizations (MCOs) or health plans. These arrangements are often complex and may differ widely, including elements such as governance requirements, payment structures, quality metrics, reporting requirements, and data sharing (Houston & McGinnis, 2016).

Several different ACO models, including the Pioneer ACO program and the Medicare Shared Savings Program (MSSP), are testing and evaluating various risk-sharing agreements. In December 2011, CMS signed agreements with thirty-two organizations to participate in the Pioneer ACO model, designed to show how particular ACO payment arrangements can best improve care and generate savings for Medicare. As of May 1, 2016, there are nine Pioneer ACOs participating in the model for a fifth and final performance year (CY2016). The MSSP is a key component of the Medicare delivery system reform initiatives included in the Affordable Care Act and is designed to facilitate coordination and cooperation among providers to improve the quality of care for Medicare fee-for-service (FFS) beneficiaries and reduce unnecessary costs. Eligible providers, hospitals, and suppliers may participate in the MSSP by creating or participating in an ACO.

Although there has been considerable debate among policymakers as to the success of the ACO model, some of these ACOs are already reporting positive results for improving patient outcomes and controlling costs, as shown in Table 4.1 (Houston & McGinnis, 2016). Table 4.1 Key attributes and broad results of current ACO models

Source: R. Houston and T. McGinnis. January 2016. “Accountable Care Organizations: Looking Back and Moving Forward.” Center for Health Care Strategies. Used with permission.

Medicare Attribute MSSP Pioneer ACO Commercial ACOs Medicaid ACOs ACO prevalence 333 ACOs in 47 states 18 ACOs in 8 states 528 commercial contracts 66 ACOs in 9 active state-based programs Key model features Shared savings payment methodology 33 quality metrics Designed for large hospital systems Shared savings system with higher risk and reward potential than MSSP Same 33 quality metrics as MSSP Often independent contracts between ACOs and MCOs Many feature narrow provider networks. Various approaches to payment including shared savings and capitation Various approaches to quality measurement Results to date CMS has reported results for different cohorts of MSSP ACOs based on start date, which have shown significant savings, but it is difficult to aggregate these results, though only 26% of ACOs received shared savings payments ACOs consistently improved on 27 of 33 quality metrics. Increases in patient satisfaction relative to patients not enrolled in ACOs $304 million in savings over three years ACOS consistently improved on 28 of 33 quality metrics.

Increases in patient satisfaction relative to patients not enrolled in ACOs Began with 32 participants; 14 have left program Not many publicly reported results available across programs due to proprietary information and difficulty comparing results CO, MN, and VT have collectively reported $129.9 million in savings. ED visits in OR decreased by 22%. ACO Challenges Now with years of observation and learnings to draw from, several key challenges facing ACOs have been identified, including difficulties working across organizational boundaries, building the requisite infrastructure for effective data sharing, and truly engaging patients in the care process. One of the more notable challenges currently being worked on is the alignment and consolidation of myriad quality measures being used in public and private programs.

Effective quality measures are imperative to accountability in organized systems of care, especially when performance affects the ability of the provider to share in savings or determines whether a provider avoids penalties or receives bonus payments (Bipartisan Policy Center, 2015). However, the notion of “measurement fatigue” and the increasing administrative burden it places on providers is a legitimate concern (Buelt, Nichols, Nielsen, & Patel, 2016). Another challenge with quality metrics is that although they tend to capture performance on specific outcomes, such as lower avoidable readmissions, or processes, such as screening for depression, they may not accurately measure the overall health of the patient, making it difficult to assess the true impact and efficacy of ACO arrangements (Houston & McGinnis, 2016). Implications for Health Care Leaders Through the combination of changing health care business models and payment mechanisms, we are witnessing transformational change in the nature of health care delivery. It is evolving from one of reactive care with fragmented accountability and a dependence on full beds to a model of health management, care that extends over time and place and rewards for efficiency and quality. This transformation poses potent challenges for providers and has enormous implications for today's health care leaders, particularly by placing greater emphasis on these issues:

Keeping patients well and managing and preventing disease Establishing more efficient organization and utilization of care teams and venues of care Creating a care culture that is comfortable with change and ongoing automation Engaging patients in managing their care and overall health Ensuring the most cost-effective care is provided and that clinical processes are streamlined and follow the best evidence More specifically, accountable care and the move to population health management will require industry perspectives and health care delivery practices to shift from

Care providers working independently to collaborative teams of providers Treating individuals when they get sick to keeping groups of people healthy Emphasizing volumes to emphasizing outcomes Maximizing the use of resources and assets to applying appropriate levels of care at the right place

Offering care at centralized facilities to providing care at sites convenient to patients Treating all patients the same to customizing health care for each patient Avoiding the sickest chronically ill patients to providing special chronic care services Being responsible for those who seek services to being responsible for the needs of the community Putting forth best efforts to becoming high-reliability organizations (Glaser, 2012b) Additionally, accountability will bring new performance and utilization risks to providers as the focus shifts from optimizing business unit performance to optimizing network performance. At the same time, instead of maximizing the profitability of care, organizations will increase the volume of desired bundled episodes while controlling costs. At an operational level, organizations must change their structure as well as workflows to implement PHM and adopt new types of automation tools and reporting. This will require setting clear goals, the active participation of leadership—including physician leaders, an assessment of technology requirements, and an effective rollout strategy (Institute for Health Technology Transformation, 2012).

Health IT clearly plays a vital role in the success of new models of care and payment reform and should be an integral part of the organization's planning process. Whether participating in an ACO or not, all health care organizations should be thinking about building a population health management strategy and addressing related gaps in their information technology (IT) capabilities. Minimally, this would include acquiring the capabilities and tools to do the following: Know, characterize, and predict the health trajectory that will happen within a population. Engage members, families, and care providers to take action. Manage outcomes to improve health and care. Accountable Care Core Processes Accountable care frameworks are based on risk and reward, with providers and organizations agreeing to share the financial risk for a population in return for the opportunity to access rewards on meeting health care quality and cost goals. ACOs are responsible for tracking and measuring specific quality metrics to indicate that patient outcomes are improving or evidence-based processes are being used. Some, but not necessarily all, metrics may be tied directly to the payment methodology, meaning that performance on these metrics will trigger either a quality incentive (such as an increased percentage of shared savings) or a disincentive (such as not receiving any shared savings) (Houston & McGinnis, 2016).

To accomplish the goals of PHM, a provider must deliver proactive preventive and chronic care to its attributed patient population. As such, the care team must maintain regular contact with patients and support their efforts to manage their own health. At the same time, care managers must closely monitor high-risk patients to prevent them from deteriorating or developing complications. The use of evidence-based protocols to diagnose and treat patients in a consistent, cost-effective manner is also central to PHM efforts. In many respects, success in population health management depends largely on a provider's ability to manage several core processes in an accountable care environment. We'll review these core processes in the next sections. Identifying, Assessing, Stratifying, and Selecting Target Populations

To manage population health effectively, an organization must be able to track and monitor the health of individual patients, while also stratifying its population into subgroups that require particular services at specified intervals. ACOs typically stratify their patient population by common care needs, conditions, and expenditure levels and then deploy tailored interventions based on these characteristics (Houston & McGinnis, 2016). For example, a high-risk pregnancy may require more frequent interventions (office visits, fetal heart monitoring, etc.) than standard prenatal care warrants.

Stratification also involves the ability to identify a patient or cohort at risk for a negative health event (e.g., myocardial infarction, stroke, mental health crisis) or preventable health care utilization (e.g., surgical procedure or hospitalization) (Gibson, Hunt, Knudson, Powell, Whittington, & Wozney, 2015). The Agency for Healthcare Research and Quality (AHRQ) describes another method of stratification as being able to identify subpopulations of patients who might benefit from additional services. Examples of these groups include patients needing reminders for preventive care or tests, patients overdue for care or not meeting management goals, patients who have failed to receive follow-up after being sent reminders, and patients who might benefit from discussion of risk reduction (Institute for Health Technology Transformation, 2012).

Although there are numerous ways to identify and segment patients, having the ability to identify risk, alert appropriate stakeholders, and intervene in the care process at the right time is a key component of population health management.

Providing High-Quality Care and Care Management Interventions across the Continuum A key tenet of accountable care is to ensure that the health and wellness of a population is managed, the most cost-effective care is provided, clinical processes are streamlined and follow the best evidence, the necessary reporting is in place, and payments and reimbursement are appropriate. Although this is an obvious goal for all providers, ACOs must facilitate cross-continuum medical management of patients for active episodes and acute disease processes or for any patient outside of the defined goals of a target population. An ACO must demonstrate, in a variety of ways, its commitment to being patient centered and to engaging patients in their care and overall health.

To effectively care for populations, care management involves the patient-centered management and coordination of care events and activities in multiple care settings by one or more providers (e.g., fine-tuning coordination among care team members, identifying care gaps and situations requiring additional interventions, as well as managing care transitions). For example, research indicates that poorly executed transitions of care between different locations (e.g., from hospital to primary care) are associated with increased risks of adverse medication events, hospital readmissions, and higher health care costs. Determining which transitions present the greatest risks and targeting care management services to patients undergoing those transitions should conserve resources and lead to better cost and quality outcomes (AHRQ, 2015).

Additionally, lack of follow-up care after hospital discharge can result in complications, worsening of patients' conditions, and a higher chance of readmission (Nielsen & Shaljian, 2013). Therefore, another example of a care management intervention is ensuring that hospitals notify primary care practices when patients are discharged and that primary care teams follow up with patients shortly thereafter.

The overall aim of care management is to manage the most complex patients through the health care system, as well as managing the overall health of a select population (e.g., diabetics and elderly), taking their preferences and overall situation into consideration. Care management ensures that all patients from the lowest risk level to high-risk “super users” receive care at the right time, in the right place, and in a manner best suited for the patient. This requires proactive care, communication, education, and outreach.

Managing Contracts and Financial Performance Under new payment models, proactively understanding patient coverage and financial responsibility will be more critical than ever. Financial teams must have a solid handle on estimating reimbursement and associated payment distributions, carrying out predictive modeling for reimbursement contracts, measuring performance against contracts and predicting profitability, as well as integrating with other key processes to share information.

For example, profit maximization under a shared savings-risk model requires a shift away from revenue-focused strategies to cost-containment strategies (Houston & McGinnis, 2016). To effectively manage costs, health care executives will need tools and data to support different types of financial modeling, such as modeling the implications of moving patient care to settings other than the hospital or physician's office. ACOs will also need actuarial cost and utilization predictors to effectively manage the care of a defined population.

These changes represent a significant cultural shift for provider organizations that must be prepared to handle a complex mix of public and private sector payment mechanisms.

Measuring, Predicting, and Improving Performance Data analytics is an integral part of PHM. ACOs typically measure quality and outcomes data against national guidelines or peer groups, and they seek to demonstrate longitudinal improvements. They might also measure costs, utilization, and patient experience on a population-wide basis, and they may use these reports as the basis for quality reporting to payers and other outside entities.

With payment so tightly linked to quality and outcomes, predicting, monitoring, and measuring system performance in key areas becomes paramount in an accountable care environment. Under value-based payment programs, there will be real ramifications for poor care and rewards for improved care. In fact, even low-performing areas can qualify for high payments if they demonstrate year-over-year improvement.

Therefore, providers must have the ability to forecast which patients are likely to become high-risk so they can intervene before a patient's condition worsens. They must also understand in real time if they are complying with a certain set of measures and monitor their continual performance. For example, ACOs will want to measure the effectiveness of care protocols, such as exercise compliance, for a population of diabetic patients. Surgical services providers will need to understand the costs and quality of proposed procedure bundles. Understanding what works and what does not is key to ensuring reimbursements, controlling costs, and, most important, providing the best care for patients (Glaser, 2012a).

Equally important is retrospective monitoring—finding out what didn't happen and why. For example, if a care provider failed to respond to an alert in a timely fashion or deviated from a given standard of care process, they can use these data to determine if new care interventions are necessary or if they need to alter an individual's plan of care. Likewise, knowing that a patient failed to keep an appointment or was unexpectedly seen in the emergency room will enable the care team to engage patients in new ways to better manage chronic disease. With providers facing penalties for readmission, it will be more important than ever to understand if it's the treatment that failed, the discharge plan that failed, or the patient who did not follow through on the post-discharge plan (Chopra & Glaser, 2013). Preparation and Automation Is Key Overall, the accountable care movement demands that providers be more focused and aggressive in managing their organization and their patients. Among other challenges, changes in reimbursement will require providers to predict which patients will need extra care, more intensively engage and manage high-risk patients, model the financial implications of delivering sub-par care, assess the performance of core organizational processes such as transitions of care, determine conformance to medical evidence, and report quality measures to purchasers of care.

The long-term success of the transition to value-based payment models and PHM relies largely on health care providers investing in the IT tools and infrastructure—as well as acquiring the data management and analysis expertise—needed to automate and support these core processes. In addition, as with any IT endeavor, expertise in change management and workflow redesign is also a core requirement.

Even for providers that may not be participating in an ACO, building the organizational and IT competencies to support accountable care is critical to staying competitive. Organizations that fail to develop and demonstrate accountable care capabilities may not fulfill their obligations to the community they serve—in fact, they may not survive.

Yet, organizations embracing the transformation from traditional fee-for-service to value-based PHM are finding significant gaps in their IT capabilities (Gibson et al., 2015). In the following section we examine the core IT building blocks and capabilities necessary to support accountable care and the move to PHM.

Data, Analytics, and Health IT Capabilities and Tools

As more providers and health systems evolve into ACOs, they are becoming increasingly aware of what it takes to manage care from a population health perspective. As we know, this includes establishing new partner networks, targeting populations, aligning providers and contracts, developing cross-continuum protocols for care management, and enabling efficient data sharing. It's All about the Data For a PHM program to be effective there is a critical need to focus on the data and information that will increasingly power clinical decisions. This includes aggregating and normalizing clinical data, claims data, administrative data, and self-reported patient data to create a holistic view of the patients within a health care network. These data enable the network to identify populations of patients whose conditions can be managed through evidence-based care plans that are coordinated across care settings.

For example, the risk of progression from glucose intolerance to diabetes mellitus can be influenced by diet and exercise. Individuals within this “rising risk” population are at different stages of readiness to change and consequently at different stages of modifiable risk. Having this insight enables providers to offer services at the appropriate level and time (AHRQ, 2015).

However, for many organizations, obtaining population health data can be difficult because it must be collected and organized from many disparate sources (e.g., laboratory information systems, EHRs, practice management systems, and home-monitoring devices). Data types that require aggregation and normalization include labs, radiology reports, medications, vital signs, diagnoses, demographic information, and more. Returning to our diabetes example, although a diabetic's blood glucose result is discrete data that can be found in an EHR, the results of the same patient's foot or eye exam may be found only in text format within a practice management system.

Data management for PHM purposes is also challenging because there's no guarantee the various IT systems talk to each other, and each provider and health plan may have a different system for patient identification and provider attribution. An important first step in connecting patient data across different care settings is to establish master patient indices (Glaser & Salzberg, 2011). Patient indices can serve as a crosswalk among the different medical record numbers and identifiers that may be used by various provider organizations to correctly identify patients. In addition, a record locator service may be used to determine which patient records exist for a member and where the source data is located. The key concept behind having a record locator service is that a patient's health information is housed on computers at the various sites of his or her care and this information is queried and aggregated from these sites at the time of a request.

Beyond the EHR: Core PHM Solution Components Although a certified EHR certainly provides the necessary foundation for effectively responding to new payment models, population health requires a range of IT applications, PHM solution components, and analytical capabilities. In fact, early adopters of PHM solutions are already seeing the need for next-generation capabilities to support the following transitions:

From management of the sickest patients to management of all patients Static risk categorization to risk categorization that follows a patient's evolving risk Focus on a single disease or condition based on simple data values and events to a focus on multi-disease or condition using evidence-based care plans “List” generation with significant manual work for care managers to significant process automation Loosely connected care “actors” to a care team that includes the patient and family Retrospective analysis to concurrent analysis (Glaser, 2016a) As organizations look to enhance their population health management strategies, they should make investments that enable the IT platform to do the following:

Collect data from multiple, disparate sources in near–real time, including any EHR, devices used in the home and at work, and other data sources, such as pharmacy benefit managers or insurance claims. Support organizations in not only aggregating but also transforming and reconciling data to establish a longitudinal record for each individual within a population. Identify and stratify populations to pinpoint gaps in care, enabling providers to act on information and match the right care programs to the right individuals (Glaser, 2016a). In addition to having an EHR that spans the continuum of care, providers pursuing PHM might invest in a PHM platform that sits above the EHR and other sources of data and must be EHR agnostic. In general, the following key technologies will enable the core accountable care processes. Revenue Cycle Systems and Contract Management Applications One could argue that the revenue cycle system forms the foundation of a provider's response to accountable care and payment reform. As the reimbursement environment becomes more complex, revenue cycle systems must evolve to support payments based on quality and performance, requiring new capabilities such as these:

Aggregating charges to form bundles and episodes, with the aggregation logic enabling different groupings for different payers Managing the distribution of payment for a bundle to the physicians, hospitals, and non-acute facilities that delivered the care Streamlining transitions between disparate reimbursement methodologies and contracts when billing and collecting Providing tools for retrospective analysis of clinical and administrative data to identify areas for improving the quality of care and reducing the cost of care delivered These new capabilities must complement routine activities such as registering patients, scheduling appointments, and administering patient billing.

Care Management Systems Used by care managers and discussed previously, care management systems enable proactive surveillance, automation, coordination, and facilitation of services for many different subpopulations across the care continuum. Specific capabilities might include helping to facilitate transitions of care more efficiently, use of automated campaigns (e-mail, text, phone) to

better manage high-risk patients, and supporting care teams in delivering evidence-based interventions to reduce high-cost utilization.

According to time-motion studies published in the journal Population Health Management by Prevea Health, automation of routine care management tasks enables care managers to manage two to three times as many patients as they can with manual methods (Handmaker & Hart, 2015). Rules Engines and Workflow Engines Processes that are efficient, predictable, and robust enable an organization to thrive in an accountable care environment. Workflow and rules engines can monitor process performance, alerting staff members to missed steps, sequence issues, or delays.

Workflow engines specialize in executing a business process, not just decisions made at a discrete point in time. The technology can greatly assist in clinical decision making by not only presenting clinicians with alerts and reminders, such as a rules engine, but also by encouraging teamwork in clinical decisions, assisting with the time management and task allocation in process delivery, stating changes in patient or operational conditions, and creating behind-the-scenes automation of process steps.

In a value-based purchasing world where each core measure needs to be associated with what's happening today, performance improvement interventions must occur in real time—that is, while the patient is still in the acute care cycle. Therefore, sophisticated IT tools such as workflow and rules engines that push information to the front lines, guiding decisions at the point of highest possible impact, will be required.

Data Warehouse, Analytics, and Business Intelligence Analytics will facilitate proactive management of key performance metrics, because accountable care creates a greater need to assess care quality and costs, examine variations in practice, and compare outcomes.

An enterprise data warehouse will fuel a wide range of analytic needs and provide intelligence to enable continual care process improvement initiatives. For example, it will be imperative that an organization can compare a hypertensive patient's total cost of care relative to its peers and national benchmarks, and perhaps even more important, predict if those costs will significantly increase because of comorbidities, complications, or gaps in care.

Applied to the data in registries or warehouses, predictive analytics tools can also help caregivers identify patients who are likely to present in the ER or be readmitted so they can tailor appropriate interventions and avoid penalties for excessive readmissions.

Although most providers lack experience with the tools and techniques associated with advanced data analysis, the application of business intelligence (BI) in health care will become the platform on which the organization not only monitors performance but also makes critical decisions to uncover new revenue opportunities, reduce costs, reallocate resources, and

improve care quality and operational efficiency. Thus, enhancing an organization's competency in data analytics and BI will become essential for success in population health management. Health Information Exchange (HIE) Essential to successful implementation of new models of care and payment reform is the exchange of clinical and administration information among different health care entities and between providers and patients. Although there has been some success in the regional health information exchange (HIE) movement, much of the focus now is on HIE capabilities at the integrated delivery system or ACO level. This enables providers to obtain a composite clinical picture of the patient regardless of where that patient was seen. By participating in an HIE or sharing health information, a number of potential important benefits may be realized:

Serves as a building block for improved patient care, quality, and safety Makes relevant health care information readily available when and where it is needed Provides the means to reduce duplication of services that can lead to reduced health care costs Enables automation of administrative tasks Provides governance and management over the data exchange process Facilitates achievement of meaningful use requirements (HIMSS, 2010) The concept of HIE is not new. For nearly two decades organizations and collaborators have tried to facilitate HIE, but unfortunately a number of HIE initiatives have failed to be sustainable over the long term (Vest & Gamm, 2010). The HITECH Act placed renewed interest in the success of HIE by providing incentive payments to eligible providers for Meaningful Use of electronic health records, which includes having the ability to exchange information electronically with others in order to have a comprehensive view of the patient's health and care (Rudin, Salzberg, Szolovitis, Volk, Simon, & Bates, 2011). However, despite investment at the national, state, and local levels, the increase in HIE utilization remains modest.

In fact, a recent survey of organizations facilitating health information exchange found that 30 percent of hospitals and 10 percent of ambulatory practices now participate in one of the 119 operational health information exchange efforts across the United States (Adler-Milstein, Bates, & Jha, 2013). Although this is substantial growth from prior surveys, the researchers also found that 74 percent of HIE efforts report struggling to develop a sustainable business model. These findings suggest that despite progress, there is a substantial risk that many current efforts to promote health information exchange will fail when public funds supporting these initiatives are depleted. Adding to the challenge, HIE efforts have struggled to engage payers, and only 40 percent of HIE efforts in the country have one or more payers providing financial support (Adler-Milstein, Cross, & Lin, 2016).

Still, there is reason to remain optimistic, with more recent data showing that hospitals' rates of electronically exchanging laboratory results, radiology reports, clinical care summaries, or medication lists with ambulatory care providers or hospitals outside their organization has doubled since 2008 (see Figure 4.1). Moreover, this exchange has significantly increased annually since 2011 (Henry, Patel, Pylypchuk, & Searcy, 2016).

Figure 4.1 Percent of nonfederal acute care hospitals that electronically exchanged laboratory results, radiology reports, clinical care summaries, or medication lists with ambulatory care providers or hospitals outside their organization: 2008–2015

Source: Henry, Patel, Pylypchuk, and Searcy (2016).

Although there is still significant progress to be made to improve the use of exchanged information and to address barriers to interoperability, HIE is critically important to the success of care transformation efforts nationwide. Thus, the industry must continue its efforts toward achieving sustainable HIE approaches to ensure that the massive national investment in health IT throughout the past decade delivers its intended return—higher-quality care, improved outcomes, and lower cost.

Registries and Scorecards Serving as a kind of central database for PHM, registries can be used for patient monitoring, care gap assessment, point-of-care reminders, care management, and public health and quality reporting, among other uses. By integrating clinical, financial, and operational data across disparate sources into a single chronic condition and wellness registry solution, data can be normalized and turned into meaningful, actionable information.

For example, registries and scorecards enable providers to identify, score, and predict risks of individuals or populations to allow targeted interventions to be implemented. When applied to a population, the registry can show, for example, how all of a particular provider's patients with type 2 diabetes are doing, which diabetic patients are out of control, or how well an entire organization is treating patients with that condition (Nielsen & Shaljian, 2013). Longitudinal Record and Care Plan As we know, even if a provider is diligently capturing patient information in an EHR, the data are valuable only in the world of collaborative, accountable care if the information can be integrated with patient data from other sources and harmonized to produce a single, consolidated record at the member level. The longitudinal record presents a complete picture of the patient's medical history in an organized, coherent view.

Serving as the sister solution to the longitudinal record, a longitudinal care plan provides a consolidated, normalized view of indicators to be monitored, events due to happen, and actions to be taken to ensure that a patient maintains and improves his or her level of health.

Patient Engagement Tools Medical interventions that occur solely through office-based patient-provider interactions will no longer provide the level of monitoring and scrutiny needed to manage the health of individuals and populations. As such, providers must continue to harness the power of technology to engage patients in their care via tools such as home-monitoring devices, patient portals, and personal health records (PHRs), as well as through the use of social media, texting, and e-mail.

Portals and PHRs

Although patient portal use is still considered modest at best, given later-stage meaningful-use requirements and the anticipated benefits of patient engagement in the value-based care world, many providers are ramping up their portal efforts and seeing adoption rates well above 20 percent (Buckley, 2015). Another recent study predicts that PHR adoption will exceed 75 percent by 2020, an optimistic projection that outpaces the PHR goals set under the Meaningful Use incentive program (Ford, Hesse, & Huerta, 2016). These consumer-centric technologies are designed to help patients and consumers better manage their own health and care, securely communicate with providers, pay bills, obtain test results, view doctors' notes, refill prescriptions, schedule appointments, and so on. Perspective The HIE Lessons Despite the fact that the environment for building, creating, and developing an HIE organization has never been better, the concerns about long-term sustainability and the impact and value of exchanging health information persist. The National eHealth Collaborative (NeHC) conducted a comprehensive study of twelve fully operationally HIEs across the nation to find out from their leaders what factors have led to their success (NeHC, 2011). In-depth structured interviews were conducted with senior executives representing the business, clinical, and technical areas of each HIE. The key critical success factors these leaders identified in sustaining an HIE are as follows:

Aligning stakeholders with HIE priorities in an intensive and ongoing effort. Create a shared vision that all stakeholders can embrace and that serves as the cornerstone to success. Foster an environment that is built on trust and that promotes learning and resolves differences when they arise. Make ongoing and effective stakeholder engagement a priority. Establishing and maintaining consistent brand identity and role as a trusted, neutral entity dedicated to protecting the interests of participants. Data use and data integrity are two critical elements. The culture, policies, and procedures regarding the use of data must ensure that no entity will gain competitive advantage at the expense of others. Consent and security policies must meet the requirements of various stakeholders and regions or states. The HIE infrastructure must ensure that patient data are accurate, reliable, and trustworthy. Ensuring alignment with vision in making strategic choices. Assess the stakeholders' alignment with the initiative and congruence with the vision before deciding to pursue them. Regardless of how promising a source of funding may have initially appeared, some HIEs chose not to pursue it because the funding source did not have the full support of all stakeholders. Considering structural characteristics and dynamics of the HIE market. The geographic location, composition of stakeholders, and resource capabilities are all factors to consider. Understanding clinical workflow and managing change. The implementation of an HIE requires that clinicians and administrative staff members understand the impact of HIE applications on workflow and identify opportunities to improve efficiencies. Different business models, governance structures, and strategies may be used to create value for the HIE participants.

Source: NeHC (2011).

Some patient portals and PHRs are integrated into a provider's existing website, and others are extensions of the organization's EHR system. For example, New York-Presbyterian (NYP) Hospital's award-winning patient portal, myNYP.org, was built to expand on its existing EHR. Use of the portal led to a 42 percent increase of appointments scheduled using myNYP.org, and it lowered the no-show rated from 20 percent to 12 percent over a period of six months after it was made available in January 2012 (Glaser, 2013). Additional applications of the same appointment-alert technology can provide customized patient education material and personalized reminders to patients who fit a specific clinical profile, such as patients who missed an immunization. Social Media Additionally, with one-third of consumers using online forums and social media sites such as Facebook, Twitter, and YouTube for health-related matters (PwC, 2012), many providers are actively engaged in using social media to communicate with patients and disseminate information on everything from emergency department wait times to new clinical offerings and research endeavors. They might also use social media channels to provide useful links to self-management tools and invitations to chronic care management programs. In fact, nearly 95 percent of hospitals have a Facebook page and just over 50 percent have a Twitter account (Griffis et al., 2014).

Automated Messaging Similar to social media, the use of automated messaging tools (via text, e-mail, or phone) can be equally beneficial in urging patients to schedule necessary appointments, fill their prescriptions, and comply with discharge orders. For example, one study showed that diabetic and hypertensive patients were two to three times more likely to attend a chronic care visit if successfully contacted using automated provider communications (Nielsen & Shaljian, 2013). Perspective Top Tips for Portal Adoption Given the modest adoption rates of PHRs and patient portals to date, research firm KLAS asked providers what best practices for patient portal adoption they would pass along to other providers trying to improve their rates. The following are their suggestions:

Educate patients. “What contributes to adoption is educating our patients about the portal, helping them sign up, and encouraging them to use it. But education is key. Patients have embraced the portal and use it for much of our communication, bill pay, results review, and more.”

Educate patients—again and again. “We ask patients on the phone whether they have signed up for the portal, and at their appointments we check to see whether they have filled things out on the portal. Then the medical assistants who greet the patients ask whether they have put their information on the portal. We promote the portal five or six times. On their way out, the doctors tell the patients that they are going to send their results to the portal.”

Educate staff members as if they were patients.

“The patients get inundated and get tired of hearing it, but it was the kickoff that got everybody in the practice used to pushing the portal. We also made everyone here register on the portal to see what the patients would go through and so we could make changes and adjustments to fit our needs. It is an ongoing process, and we try to do contests every quarter. That is what contributes to our success, and it is pretty impressive.”

Give patients a reason to use the portal. “We are apparently doing something right in encouraging patients to come to our portal. They come to the portal to fill out the patient history and the medication list. I think that is because of the way our front desk staff members make new-patient appointments and the way they present the portal to the patients. They tell them that we can give them less waiting time when they come in if they get on the portal. We have an aggressive sign-up process. We give patients a Chromebook in the waiting room and help them sign up for the portal right away. We have a similar process in the ED and inpatient areas. We try to push as much content to the portal as possible.”

Talk to your vendor and physicians. “We drove adoption from the top down. In our initial phase, the adoption didn't go well because we thought we knew what we were doing and could do it ourselves. We went back and listened to Medfusion. We took the portal to the doctors who understand technology. They came back from a CMS meeting and said we had to do the portal. They said we might not like it, but we have to do it.”

Hold your vendor accountable. “When we started to deploy Empower in our ambulatory area, we hit challenges and barriers with the physician group. The physicians really wanted to yank the product out; they didn't want anything to do with it. They were beyond frustrated. We worked with MEDSEEK and the physicians, and in the last year and a half, we went from having a handful of patients on the portal to having sixty-five thousand. We were finally able to leverage the solution in the ambulatory space after we made changes to the product and the interface. There were deal breakers in how the product looked and felt from a patient perspective, and we worked through those.” PHM is most effective when a symbiotic relationship exists between human interventions and automation tools. Patient engagement tools and outreach programs enable providers to correspond with each person in their patient populations, with the goal of raising the percentages of patients receiving the recommended care as reflected in the quality measures payers use to evaluate provider and health system performance. More important, such programs assist providers in keeping patients as healthy as possible for as long as possible, a core tenant of PHM.

Telemedicine and Telehealth The growing use of telemedicine can make patient interactions more convenient, expand geographic horizons particularly where needed medical specialists are few in number, and make care more accessible to those with mobility issues.

Perspective Five Reasons to “Like” Consumers' Use of Social Media With an abundance of patient-generated health information now available through online patient communities, social media can play a vital role in improving our understanding of disease and accelerating new approaches to treatment. Consider the following ways patient and consumer use of social media is benefiting health care.

Creates a Sense of Community For those seeking emotional support and tips for coping with a disease, social media delivers on many fronts. It can enable the formation of communities regardless of member locations and enable members to communicate asynchronously.

Sites such as PatientsLikeMe and Inspire provide virtual medical communities focused on chronic diseases where patients can discuss their conditions, track key health information, share side effects of medications and therapies, and bond with others as they chronicle the highs and lows of their health care journeys.

In fact, a 2014 survey of PatientsLikeMe members found that the vast majority of adult social media users with health conditions embrace the idea of sharing their health information online if it helps clinicians improve care, assists other patients, or advances medical research.

Users of online health communities also frequently cite as reasons for their membership the accountability the sites provide them in managing their own health and reaching their health-related goals, as well as the motivation, support, and advice they receive from others. Online communities can also lessen the feeling of isolation that often accompanies those with rare conditions or parents with a critically ill child. Delivers New Clinical Research Insights As more and more patients use social media to track their health conditions and actively participate in their care, there is a greater opportunity to use this real-world data to better inform new treatments and treatment decisions, enhance symptom management, and ultimately improve outcomes.

For example, in analyzing the results of observational data housed on PatientsLikeMe, researchers found that lithium therapy had no impact on ALS disease progression, which was later confirmed by subsequent randomized trials (Chretien & Kind, 2013).

Although PatientsLikeMe began as a social network enabling people to crowdsource the collective wisdom of others, it has developed into a powerful analytical platform for clinicians and researchers. In fact, the network is quite transparent with its members about how it makes money—by sharing the information members provide about their experience with diseases and selling it to their partners (companies that are developing or selling products to patients). This may include drugs, devices, equipment insurance, or medical services.

In addition to helping patients find and take advantage of clinical trials, health care social networks also provide an opportunity for participant-led research, in which members initiate new fields of study. For instance, Inspire members with spontaneous coronary artery dissection (SCAD) persuaded researchers at the Mayo Clinic to launch new research about their condition, which led to the creation of a SCAD registry, a key step in the further study of this rare disease (Tweet, Gulati, Aase, & Hayes, 2011). Indeed, there is tremendous potential for online patient communities to contribute to the notion of a continuously learning health system.

Builds Awareness of Cause-Related Issues or Personal Health Care Crises Social media can also serve as the birthplace for beneficial social movements, as well as hubs for galvanizing emotional and financial support for a personal health care crisis.

The ALS Ice Bucket Challenge is a terrific example of social media's power to deliver on the fund-raising aspect of the campaign and on the equally important goal of helping the public become more aware of ALS and efforts to find a cure.

The simple act of pouring ice on one's head, capturing it on video, and calling out another person to do the same spread across social media channels like wildfire. With everyone from schoolchildren to celebrities getting in on the act, the ALS Association raised $115 million in 2014, a staggering increase from its $23.5 million intake in 2013 (ALS Association, 2015).

On a smaller scale, sites such as GoFundMe and My Cancer Circle can help keep family and friends abreast of a loved one's illness and treatment status, provide tools to coordinate meal deliveries and rides to medical appointments, as well as enable financial contributions to help offset personal health care expenses. Provides Assistance with Treatment, Physician, or Hospital Selection Although physician rating sites have been around for many years, social media has given health care consumers a more active voice and an ever-present tool set for broadcasting opinions on all things health care–related—from physicians and hospitals to medications, devices, and insurance plans.

Like it or not, social media is proving to be a vehicle that can help scale positive and negative attitudes about one's health care experience at Internet speed. In fact, a 2012 survey by Demi & Cooper Advertising and DC Interactive found that 41 percent of people said social media would affect their choice of a specific doctor, hospital, or medical facility.

Of course, the downside here is that the negative opinions of a vocal minority could cause unjust reputation management issues for providers.

With the viewpoints of those in online social networks playing such a key role in influencing health care decisions, providers ought to ensure they are optimizing their social media channels and actively participating in helping consumers share positive opinions online.

Complements Traditional Approaches to Measuring Patient Satisfaction

Beyond just randomly monitoring opinions shared on social media, savvy providers may want to turn to social media to supplement their traditional means of capturing patient satisfaction and feedback on inpatient experience.

In fact, researchers at Boston Children's Hospital conducted a study to determine if Twitter could provide a reasonable form of complementary quality measurement, given the real-time nature of tweets. The team amassed unsolicited knowledge (versus data gleaned from very targeted survey questions) about what pleased or angered consumers by collecting more than 400,000 tweets directed at the Twitter handles of nearly 2,400 US hospitals between 2012 and 2013 (Ulrich, 2015).

Although certainly no replacement for patient satisfaction surveys, according to the researchers the data are suggestive and provide proof of principle that Twitter and the right analytical tools may provide a valuable means for complementing standard approaches to measuring quality. Moreover, the ability to correlate social media data points such as tweets with actual outcomes measures (e.g., patient length of stay in the emergency department or readmission rates) provides an interesting avenue for further exploration. The American Telemedicine Association defines telemedicine or telehealth as exchanging medical information via electronic communications to improve a patient's clinical health status. Health care providers are embracing telemedicine because they see it as an efficient and cost-effective way to deliver quality care and improve patient satisfaction (Glaser, 2015a). Today's telehealth framework spans the continuum of care and can include services such as the following:

Telepsychiatry Remote image interpretation (teleradiology, teledermatology) e-Visits or televisits between providers and their patients Video visits for semi-urgent care Clinician-to-clinician consultations Critical care (virtual ICU, telestroke) Remote monitoring of a patient with a chronic disease Cybersurgery or telesurgery Let's take a closer look at some of the more popular applications of telemedicine and telehealth. Two-way interactive video-conferencing or other web-based technologies can be used when a face-to-face consultation is necessary. In addition, a number of peripheral devices can be linked to computers to aid in interactive examination. For example, a stethoscope can be linked to a computer, enabling the consulting physician to hear the patient's heartbeat from a distance. Electronic monitoring of physiological vital signs can be done through electronic intensive care unit (eICU) patient-monitoring systems, and telesurgery can enable a surgeon in one location to remotely control a robotic arm to perform surgery in another location.

Telehealth is also being used to capture and monitor data from patients at home. Examples include monitoring patient blood sugar levels through glucometers attached to cell phones and conducting teledermatology visits with the aid of cell phone cameras.

According to the American Hospital Association (AHA), 52 percent of hospitals used some form of telehealth in 2013, and another 10 percent were beginning to implement such services (AHA, 2015). Its growth potential is also notable. Business information provider IHS predicts the US telehealth market will grow from $240 million in revenue in 2013 to $1.9 billion in 2018—an annual growth rate of more than 50 percent (EY, 2014).

In addition to the growing demand for access and convenience, the need for telemedicine is driven by other factors such as the following:

Significant increase in the US population Shortage of licensed health care professionals Increasing incidence of chronic diseases Need for efficient care of the elderly, homebound, and physically challenged patients Lack of specialists and health facilities in rural areas and in many urban areas Avoidance of adverse events, injuries, and illnesses that can occur within the health care system These factors become increasingly important as new health care delivery and payment models evolve and providers are challenged to better manage chronic diseases, avoid readmissions, improve quality, and remove low acuity care from high-cost venues. As we know, the long-term benefits of population health programs are predicated in large part on managing high-cost, chronically ill patient populations more effectively. Furthermore, the rapid deployment of high deductible health plans, which make consumers more conscious and accountable for their health care consumption and spending, has added to the pressure on providers to provide low-cost, convenient options.

Despite all its promise, several major barriers must be addressed if telemedicine is to be used more widely and become available. Concerns about provider acceptance, interstate licensure, overall confidentiality and liability, data standards, and lack of universal reimbursement for telemedicine services from public and private payers are among the complex and evolving issues affecting the widespread use of telemedicine. Furthermore, its cost-effectiveness has yet to be fully demonstrated.

Nonetheless, the barriers are beginning to erode under mounting pressure from all health care constituents. Licensure portability will further ease the barriers to accessing services, whereas regulatory and payment policy changes in support of telehealth are widely expected in the coming years. For instance, on the private payer side, telemedicine use has been bolstered by a growing number of states enacting parity laws, which require health insurers to treat telehealth services the same way they would in-person services.

Transitioning from the Record to the Plan As we reviewed in this chapter, the profound changes in reimbursement and care models are altering the structure of care provision, requiring providers to make investments in a comprehensive IT portfolio—beyond the EHR—to support PHM and enable the core processes

associated with accountable care. These changing business and payment models are leading not only to significant changes in organization and practice but also to changes in the fundamental nature and design of the EHR itself. These changes can be characterized as a transition from the electronic health record to the electronic health plan (Glaser, 2015b).

The EHR does not disappear as a result of this shift. We will still need traditional EHR capabilities: providers need to review a radiology report and document a patient's history and the care delivered. Problems must be recorded and medications reconciled. However, the strategic emphasis will move to technologies and applications that assist the care team (including the patient) in developing and managing the longitudinal, cross-venue health plan and assessing the outcomes of that plan. For example, evidence-based pathways and decision-support logic have been embedded into EHRs to guide provider decisions according to a plan based on patient condition. EHRs can now include or be enhanced by the specific PHM technologies we discussed that enable the organization to understand its aggregate performance in undertaking disease-specific plans for multiple patients.

Provider organizations will not thrive in an era of health reform because they have a superb and interoperable EHR. They will thrive because the care they deliver consistently follows a plan designed to ensure desired outcomes. The EHR must evolve so it focuses on individual patients' care plans—the steps required to maintain or create health.

Every patient's EHR should clearly display the master care plan—a long-term care plan to maintain health integrated with short-term plans for transient conditions. The EHR should be organized according to this master plan: it should highlight the steps needed to recover or maintain health, list the expectations of every caregiver the patient interacts with, and include tools such as decision support and a library of standard care plans. Interoperability is a necessity, because various providers must be able to use the plan-based EHR.

Care Plan Attributes The care health plan has attributes that need to be present to ensure health and should be based on some fundamental ideas.

First, all people have a foundational plan. If the person is a healthy young man, the plan may be simple: establishing health behaviors such as exercise. If the person is a middle-aged man with high cholesterol and sleep apnea, the plan may be annual physicals, statins, a CPAP machine, and a periodic colonoscopy. If a person is frail and elderly with multiple chronic diseases, the plan may be merging the care for each chronic condition, ensuring proper diet, and providing transportation for clinic visits.

Second, plans are a combination of medical care strategies with goals to maintain health (such as losing weight) along with public health campaigns (such as immunizations).

Third, on top of foundational plans there may be transient plans. For the patient undergoing a hip replacement there is a time-bounded plan beginning with presurgery testing and ending when rehabilitation has been completed. A patient undergoing a bad case of the flu has a time-bounded plan. Fourth, people who have a common plan are members of the same population. These populations may be all patients undergoing a coronary artery bypass graft in a hospital, all patients with a certain chronic disease, or all patients at high risk of coronary artery disease. Moreover, a particular person may be a member of multiple populations at the same time.

Fifth, risk is the likelihood that the plan will not be followed or will not result in desired outcomes. A patient motivated to manage his or her blood pressure has a lower risk than a patient who is not motivated. A frail person with multiple chronic diseases is at greater risk that the plans will not keep him or her out of the hospital than a person whose health is generally good despite having multiple chronic diseases.

Sixth, not all care will be amenable to a predefined patient plan. Life-threatening trauma, diseases of mysterious origin, sudden complications—all require skilled caregivers to make the best decisions possible at the moment.

Seventh, plans should be based on the evidence of best care and health practices. And the effectiveness of a plan should be measurable, either in terms of plan steps being completed or desired outcomes being achieved (Glaser, 2015a).

The Plan-Centric EHR The EHR needs to evolve into plan-centric applications. Among others, these applications will have several key characteristics.

A Library of Plans That Cover a Wide Range of Situations This library will include, for instance, plans for managing hypertension, removing an appendix, losing weight, and treating cervical cancer. There will be variations in plans that reflect variations in patient circumstances and preferences, for example, plans that depend on whether the patient is a well-managed diabetic or plans that reflect the slower surgical recovery time of an elderly person.

A master plan will combine, for example, the patient's asthma, hysterectomy, depression, and weight-reduction plans into a single plan. These algorithms will identify conflicts and redundancies among the plans and highlight the care steps that optimize a patient's health for all plans. For example, if each of the five plans has six care steps, the algorithms can determine which steps are the most important.

Team-Based The master plan will cover the steps to be carried out by a patient's primary care provider, specialists, nurse practitioners, pharmacists, case managers, and the patient. Each team

member can see the master plan and his or her specific portion of the plan. Team members can assign tasks to each other (Glaser, 2015a).

Business Models in Other Industries Major changes in an industry's business model invariably lead to major changes in the focus and form of the core applications used by that industry. For example, financial services, retailers, and music distributors, along with many other industries, have also experienced massive shifts in their business models.

Several decades ago, financial deregulation enabled banks to offer brokerage services. The business model of many banks shifted from banking (offering mortgages as well as checking and savings accounts) to wealth management. As banks shifted from transaction-oriented services to services that optimized a customer's financial assets, their core applications broadened to include an additional set of transactions (buying and selling stocks) and new services (financial advisory services).

Prior to the web, most retailers' business models focused on establishing a brand, offering an appropriate set of well-priced products, and building attractive stores in convenient locations. The web enabled retailers to gather significantly richer data about a customer's buying patterns and interests (and to use real-time logic to guide purchasing decisions). Retailers' core applications broadened to include well-designed e-commerce sites and analytics of customer behavior.

In both examples, even though there was a significant shift in the business model, applications needed for the previous model continued to be necessary. Banks still had to handle savings account and mortgage payment transactions. Retailers still needed to manage inventory. And advances in these legacy applications—expanding inventory breadth and reducing inventory-carrying costs—continue to be important. In each case, a critical new set of applications were added to the legacy applications. Often, these new applications were more important than legacy applications.

The business model changes in health care will lead to a shift from applications focused on the patient's record to applications focused on the patient's plan for health. This evolution in the nature of the EHR is a key component to achieving success in population health management. Summary As the health care industry continues its transition from a fragmented, volume-based system toward one that embraces the notion of patient-centered, accountable care driven by value-based payment models, providers must consider what new relationships, processes, and IT assets and skills will be required to succeed—particularly when it comes to managing the health and care of attributed populations.

By implementing a PHM strategy, organizations have enormous opportunity to use data and analytics to improve inefficiency and waste, thereby reducing costs, and monitor adherence to evidence-based protocols to drive better outcomes. Several PCMHs and ACOs are already

showing promising performance in the emerging world of value-based payment and population health management.

In addition to having a robust EHR, organizations looking to enhance their PHM strategies should consider several key solution components. PHM technologies can help providers stratify and select target populations, identify gaps in care, predict outcomes and apply early interventions, and actively engage patients in their care. Moreover, they can enable an organization to understand its aggregate performance in undertaking disease-specific plans for multiple patients and better manage contracts and financial performance.

Additionally, because value-based payment is based on conformance to chronic disease protocols, providers must have the ability to aggregate and normalize real-time, accurate, cross-continuum data from disparate sources illustrating how well the data conform to those protocols. As we know, many hospitals and health systems do not operate from a position of excess revenue, and as outcomes become increasingly tied to the reimbursement stream, it will become critical that providers can rely on their data and IT tools to detect and remedy variations in care.

Population health management solutions are intended to complement—not replace—the traditional EHR. They represent a shift from applications focused on documenting the patient's record of care to applications focused on developing the patient's plan for health.