Key Strategies for Healthcare in the Cognitive Era
Download these slides to brainstorm, shape, and convey your population health strategies.
Why population health management addresses the care of populations and the engagement of patients across care settings and over time.
How ACOs have strong incentives to improve population health to meet quality goals and reduce costs.
How the PCMH model is essentially holistic primary care, in which physician-led care teams coordinate and manage care, and how they can achieve their full potential by integrating with the medical neighborhood and automate care management.
Where predictive modeling can be used for risk stratification of a population or risk adjustment, and how it is used to classify the population into high-, medium-, and low-risk categories so care teams can deliver appropriate interventions to each group.
Why clinical integration of providers must come before ACOs and population health management.
How automation tools for risk stratification, patient outreach, and care management can leverage the capabilities of care managers and enable physicians and all care-team members to work at the top of their licenses.
Why a healthcare organization’s data infrastructure is critical to its success in population health management, and how infrastructures must be comprehensive, scalable, and flexible for years to come.
How social and behavioral factors play a giant role in health, to an even greater extent than health care proper, and explain the variances in individual health over time.
How cognitive computing can become a dependable and essential assistant to physicians and to organizations engaged in population health management.
Why patient engagement is vital to quality improvement, better patient outcomes, and population health management, even being called the “holy grail” of population health management.
The $3.2 trillion health-care industry, as conventional wisdom has it, is a big ship to turn around. But, employers, consumers and government can no longer afford health-care costs that, while growing more slowly than in past years, have reached stratospheric levels. The fee-for-service payment system that rewards providers for the volume of services has been implicated in the high cost of health care. So, with a concerted push from payers, the industry is in the midst of a rapidly accelerating shift from fee for service to various forms of “pay for value.”
The Centers for Medicare and Medicaid Services (CMS) has already taken a number of steps in its transition to value-based payments. To start with, the Medicare Shared Savings Program (MSSP) is rewarding accountable care organizations (ACOs) that create savings and meet quality goals. While most of the 400-plus ACOs participating in this program are taking only upside risk today, many of them will have to accept downside risk as well, starting in 2018, if they choose to renew their MSSP contracts. CMS also has placed a small portion of hospitals’ Medicare revenue at risk for achieving cost and quality goals, and it began applying a similar pay-for- performance program to physicians in 2015.
By the end of 2018, half of Medicare payments are expected to go to alternative payment mechanisms (APMs) such as ACOs, patient-centered medical homes (PCMHs), and bundled payments. Further, the new law that replaces the sustainable growth rate (SGR) with a different Medicare payment approach gives physicians involved in APMs a 5 percent annual bonus from 2019 to 2024.
Private payers are moving in tandem with CMS. In March 2014, Anthem BlueCross BlueShield, one of the nation’s largest health insurers, said that it had tied a third of its commercial reimbursement to pay-for-value quality programs. UnitedHealth Group said it was expanding its incentive programs, with a goal of offering at least half of its network physicians the ability to earn bonuses for value, quality, and efficiency within a few years. Aetna is paying incentives to practices that have achieved PCMH recognition and is working with scores of provider groups and health systems to create accountable care organizations (ACOs).
About half of the 700-plus ACOs have contracts with private payers. Most of these contracts are based on shared savings rather than capitation. But 45 percent of private-payer agreements include downside risk, meaning that providers are at financial risk if their health-care spending exceeds their budgets.
What all of this means is that health-care providers can no longer avoid the reality that their current business models are obsolete. As they transition to the new care- delivery methods, they must stop basing business decisions on how their clinicians and facilities can produce additional and ever-more-costly billable services. Those services and facilities have been profit centers up to now; but in the new world of value-based reimbursement and financial risk, they are becoming cost centers.
The fulcrum of profitability in this new world is maintaining or improving patients’ health and delivering good outcomes. The only proven way to achieve those goals is to manage population health effectively and efficiently. To do that, health-care organizations need advanced health IT, including analytics and automation tools that enable them to transform their work processes.
Changing the Mindset
Except for group-model HMOs such as Kaiser Permanente and Group Health Cooperative, some large groups and IPAs in California, and a few health-care systems in other states, health-care providers are not well positioned for population health management (PHM). While many health-care organizations are creating new structures to prepare for value-based reimbursement, health care is still oriented to fee-for-service. Physician practices still organize care around office visits, and hospitals focus on acute care within their four walls.
One recent study found that physician practices of all sizes increased their use of evidence-based care management processes from 2006 to 2013. But, by the end of that period, even large groups used fewer than half of the recommended processes for chronic disease management, on average.
The concept of caring for entire patient populations on a continuous basis, whether or not individual patients present for care, is only gradually seeping into the consciousness of health-care managers and providers. And it is still difficult for many provider organizations to accept the idea that filling beds and appointment slots is less important than ensuring that all patients receive recommended preventive and chronic care.
To transform themselves, organizations must reduce two kinds of waste: first, the avoidable tests, procedures, hospital admissions and readmissions that lead to high costs for employers and consumers, and second, the internal waste that inflates the cost of care delivery. The reorganization of care processes can address both kinds of waste simultaneously by improving the quality and efficiency of care.
Organizations that go down this path need to adopt consistent policies and procedures, starting with a common set of clinical protocols. They have to form care teams that can coordinate care for every patient, tailoring their approach to the individual’s health risks and conditions. They must restructure work flows so that each member of the care team is working up to the limit of his or her training and skill sets. And they must use their care managers as efficiently as possible so they can provide appropriate support to all patients who need help.
Electronic health records (EHRs), which have become widespread in the past few years, are essential to any PHM strategy. But EHRs are not designed to support PHM. While they can supply much of the data required to track and monitor patients’ health and identify care gaps, they must be combined with claims data to provide a broad view of population health and to track individual patients across care settings. Moreover, providers need electronic registries to identify care gaps and provide the near-real-time data required to intervene with subgroups of patients efficiently and in a timely manner. Although some EHRs include such registries, they’re not as comprehensive, flexible, or usable as those available from third party developers.
The IT infrastructure for PHM must also include applications that automate the routine, repetitive work of care management. These automation tools offer several advantages: First, they can lower the cost of care management by taking over time-consuming chart research and outreach work. Second, they free up care managers to devote personal attention to high-risk patients who urgently need their help. Third, they allow providers to do pre-visit planning and post-visit follow-up on a consistent basis. Fourth, they can bring noncompliant patients back in touch with their personal physicians. And fifth, these tools enable organizations to quickly scale up their care management efforts so they can continuously care for all of the patients in their population.
Most important, the combination of these tools offers a mechanism for engaging patients in their own health care. Without patient engagement, population health management is impossible.
The rise of accountable care organizations (ACOs) in recent years reflects the concurrent emergence of value-based reimbursement and financial-risk contracts. Composed of physicians and hospitals that are committed to lowering costs and improving quality, ACOs must be able to deliver high-quality care within a budget. Strategies such as admitting patients to lower-cost hospitals and de-emphasizing expensive tests can help them do this in the short term; but in the long term, ACOs will have to manage population health well to be successful.
The patient-centered medical home (PCMH)—a holistic approach to primary care—is considered an essential building block of ACOs. The National Committee for Quality Assurance (NCQA) has given medical home recognition to more than 10,000 practices, comprised of over 48,000 providers, and the number of PCMHs is growing rapidly.
The growth of patient-centered medical homes bodes well for the transformation of health-care through ACOs. But to coordinate care effectively across care settings, the primary care physicians who have built medical homes must gain the cooperation of specialists, hospitals and other health-care players in the medical neighborhood.
This might seem like a no-brainer at a time when health-care organizations are trying to prepare for value-based payments. But during this transitional period, when most specialists and hospitals still depend to a large extent on volume-based reimbursement, it’s not easy for primary care doctors to persuade them that their future success depends on working with medical homes to coordinate care and reduce costs. Employed physicians will follow organizational directives to some extent, but at least half of physicians are still in independent practices.
Some health-care organizations, including hospital systems and IPAs, have formed clinically integrated networks (CINs) that facilitate the collaboration of providers across care settings and business boundaries. These networks, which depend on health IT for communications and data sharing, can connect providers who otherwise might not collaborate with one another. Whether CINs will spread and help organize medical neighborhoods remains to be seen, but that’s what they promise to do. If they succeed, they can provide the foundation of effective ACOs.
As mentioned earlier, patient engagement is the sine qua non of population health management. At a population-wide level, this is about ensuring that patients take good care of themselves and comply with doctors’ recommendations for proper preventive and chronic care and better health behavior. Fully half of disease prevention is up to patients, including their diet, exercise, and smoking behavior.
But acute care, especially in hospitals and ambulatory surgery centers, also requires patient engagement for optimal outcomes. Not only must patients prepare for procedures, but they must also follow their post-discharge care plans and take the medications that have been prescribed for them.
It is unrealistic to expect some patients—especially those who are very ill, elderly, or poorly educated—to comply fully with their discharge instructions. Even if they want to, they may be unable to comply because they can’t afford their medications or can’t get an appointment with their primary care physician. So in a world of accountable care and value-based payments, providers must learn how to help these patients and get them involved in their own care after they leave the hospital.
The government’s Meaningful Use EHR incentive program encourages providers to engage patients in several ways. In stage 2 of Meaningful Use, eligible professionals (EPs) and eligible hospitals must share records with patients online. EPs must also communicate with patients via secure online messaging and must send them reminders about preventive care. The final rule for stage 3 requires that patient-generated data be incorporated into the EHR for more than 5 percent of all patients seen by an EP or discharged from an eligible hospital or emergency department during a reporting period. These may include a variety of inputs, including data from home health monitors, wearable devices, and patient-reported outcome data.
Meanwhile, technology offers several other avenues for engaging patients. Automated patient outreach programs can message patients by e-mail, text or phone, reminding them to make appointments with their doctor for needed preventive and chronic care. Text messaging has been shown to help certain kinds of patients, such as people with diabetes and pregnant women, take better care of their health. Educational materials tailored to individual patients can be made available online. And personal health records (PHRs), which can be used to download, store and transmit health records, can help patients keep track of their care plans and medications and communicate with their providers.
Recent advances in home telemonitoring and telehealth using mobile devices provide not only the opportunity for remote consultations, but also generate data that can keep providers and care managers apprised of patients’ conditions between visits. Of course, these new data streams could easily overwhelm providers; the information must be carefully screened so that caregivers see only relevant data.
A roadmap for population health management
No two health-care organizations are exactly the same or operate in the same environment with the same population. Nevertheless, as this book makes clear, provider organizations can follow a common roadmap that will take them close to where they want to go.
That map begins with the risk stratification of their population to identify which patients have the greatest health risks (and therefore, pose financial risks to the organization). Health-risk assessments administered to patients and analytics applied to clinical and claims data enable organizations to classify their populations.
Organizations should also reengineer their work processes, using Lean/Six Sigma methods where possible, and then apply automation tools to make those processes more efficient. The first step is to automate patient outreach, applying clinical protocols to registries that are either standalone or part of EHRs. These registries can be used to launch outbound messages to patients who have care gaps.
Another type of automation involves running reports on registry data to classify patients by subgroups. Care managers can then design campaigns to improve the health of specific subgroups, such as patients with type 2 diabetes and hypertension.
Many organizations make a mistake when they try to perform care management manually. They end up hiring a large number of nurses to search medical records and call patients individually. This is not only a waste of time and money, but it also wastes the skills of these highly trained professionals.
Another common error is a failure to do pre-visit planning and post-visit follow-up. With the help of automation, it’s relatively easy to find out what patients should do before office visits, such as getting tests done so that physicians can see the results. Similarly, tracking patient compliance after visits is not a big chore if the practice uses its EHR for automated tracking of orders for tests and referrals. Providers can find out whether patients filled prescriptions by getting online medication histories from Surescripts.
With the help of some health plans, a growing number of providers are using claims data to identify their patients’ care gaps and do predictive modeling. Claims data shows all services that a plan member received from any provider, but it is out of date and contains a lot of errors. By feeding a registry with integrated EHR and claims data and applying analytics to this information, organizations can identify care gaps and high-risk patients in near-real time. Moreover, that actionable information can be supplied to providers at the point of care and to care managers as they plan their work and prioritize their cases. That allows them to intervene proactively with the patients who need help the most.
What this book is about
As the foregoing comments suggest, this book draws connections among the new care delivery models, the components of population health management, and the types of health IT that are required to support those components. Two key concepts tie all of this together: 1) Advanced analytics must be applied to comprehensive data to understand all the factors involved in population health and to make the data actionable; and 2) PHM requires a high degree of automation to reach everyone in a population, engage those patients in self-care, and maximize the chance that they will receive the proper preventive, chronic and acute care.
In the course of explaining how to do this, we also describe how health-care organizations are transforming themselves to manage population health and prepare for value-based reimbursement. The ACO, PCMH and CIN models have already been discussed, and the advent of bundled payments will also have a major impact on hospital and post-acute care. But at its core, the transition to accountable care focuses on care teams that take responsibility for managing and coordinating the services provided to individual patients. These care teams must also engage patients in caring for themselves and improving their health behavior. And as care teams become more sophisticated, many of them will use Lean thinking to continuously improve their own work processes.
The book is laid out in three sections that progress from the general to the particular aspects of population health management. Section 1, entitled “New Delivery Models,” first explains what PHM is and why it’s important. Ensuing chapters cover ACOs, patient-centered medical homes, and clinically integrated networks, which are the favored vehicles for PHM.
Section 2, “How to Get There,” discusses the health IT infrastructure that PHM requires, starting with the impact of the government’s “meaningful use” program on EHR development and adoption. Other chapters in this section address the extended data infrastructure needed to support PHM, predictive modeling applications, and the return on investment in automation and analytic solutions designed for PHM.
Section 3, “Implementing Change,” describes how organizations can use health IT to manage population health. This begins with the basics of care coordination and moves on to advanced methods of care management that utilize Lean thinking. Following a chapter on overall methods of patient engagement, we discuss post-discharge automation, which is another way to involve patients in their own care.
The final pair of chapters covers the new trends that are becoming important in population health management. Chapter 13 explains why social determinants of health (SDH) are now on the agenda of many organizations involved in PHM and shows the best way to make a difference in this area. Chapter 14 describes the strides that cognitive computing is making in health care and explains why it is integral to the future of PHM.
While this book is intended for health-care executives and policy experts, anyone who is interested in health care can learn something from its exploration of the major issues that are stirring health care today. In the end, the momentous changes going on in health care will affect all of us.
Health care in the U.S. does not function as an effective system for a variety of well-documented reasons, as I pointed out in my introduction to the textbook Population Health: Creating a Culture of Wellness. With a strong push from the federal government, as well as private payers, the U.S. health care industry is slowly pivoting toward value-based reimbursement. This transition to “income for outcome” will incentivize health care providers to pay more attention to non-visit care and will induce health care organizations to start managing the health of their patient populations, not just their health care. Only by doing so can they hope to reduce costs and improve quality enough to succeed financially under the new payment models. Moreover, to manage care properly, disparate health care providers and institutions will have to cooperate with each other to build a real health care system.
The premise of this book, contained in its title, is that population health management (PHM) cannot succeed unless physicians, their care teams, extended care networks, and community resources align with each other. It makes a whole lot of sense for doctors to play a leading role in population health management. Outside of friends and family, consumers trust physicians more than any other health care constituency, and certainly more than insurance or drug companies. The doctor-patient relationship is the key to patient engagement, which can lead to improved medication adherence, evidence-based guideline compliance, and lasting, sustainable health behavior change.
Written by the experienced team at IBM Watson Health, the book focuses sharply on the practical mechanics of how healthcare organizations can transition to population health management. While generous dollops of theory are also provided, the most germane parts of the book describe the practice of this new model (to most practitioners) of health care delivery.
For example, consider the chapters about consumer engagement and the social determinants of health. Individual engagement is a prerequisite of population health management; without it, people are less likely to make the lifestyle changes required to prevent or reduce the impact of chronic illnesses. But to get individuals engaged and support them in self-care management, health care providers must also be aware of the social determinants of each person’s health. As noted in Chapter 13, clinical health care accounts for only 10%-25% of the variations in individual health over time. Health care’s influence on the length of quality life would be markedly augmented, however, if it were combined with efforts to improve the social, economic, emotional, and physical environmental factors that contribute to health.
Research supports the need for population health management to extend beyond health care. Collaboration among providers on care coordination will ultimately need to incorporate social services, behavioral health, job placement and advancement, housing, and possibly spiritual and other community resources.
During my tenure in many population health leadership roles, including being the Global Medical Leader of GE and Corporate Medical Director for Truven Health Analytics, I learned the importance of paying attention to all health determinants, especially in engaging non-compliant individuals. Often their resistance can be overcome by enlisting other domains for assistance. Many people are more inclined to manage their own care, for example, after establishing a home to live in and finding a steady job.
Another theme that runs through the book is the need to build a health IT infrastructure that can support population health management. Initially, many health care delivery systems assumed that they could simply rely on their EHR vendor to give them everything they needed. But experience has demonstrated that the EHR is only a starting point. It requires additional advanced data sources, cognitive analytics, secure, cloud-based platforms and mobile capabilities to establish a robust population health solution.
Moreover, PHM requires the ability to aggregate, normalize and analyze data from many different sources. Individuals receive their care across multiple care settings; many different care providers function inside a variety of delivery systems; and members of accountable care organizations (ACOs) utilize many different EHRs and patient portals. To connect these providers and health records requires the pursuit of interoperability, which is still largely lacking in health IT systems. These systems also lack the kind of clinical decision support and automation tools needed to facilitate care management and to make it efficient and effective. Longitudinal care coordination will ultimately require a personal health record that includes data from all care settings in which the patient has been treated.
The book’s final chapter takes an in-depth look at the exciting developments in cognitive computing, which has the potential to take health care and PHM to a whole new level. A next-generation big data approach, cognitive computing can help health care organizations understand their populations better by providing insights into factors such as demographics, geographical location, behavioral health, transportation, lifestyle choices, consumer purchases, and socioeconomic status.
Cognitive computing can also search the medical literature in seconds, can use natural language processing to convert unstructured data into structured data, can improve predictive modeling, and can provide the analytic power to help physicians understand the genomic information about the people on their panel. This is what doctors will need in 21st century medicine. On their own, they will never be able to absorb more than a tiny fraction of the 1 million new published articles that come out each year. And with the emergence of genomics, proteomics, and microbiomics data, analytics will be an essential part of everyday medicine. Doctors will require significant learning – indeed, cognitive – computing resources to determine the relevance of all this data so they can provide highly personalized care, precision medicine, and the most appropriate care pathway for each person.
The most valuable lesson you will gain from this book, however, is that health care providers can impact the health status of the communities they serve. By doing so, they can improve the health of individuals and can also contribute to their performance as workers and as family and community members. The potential impact is enormous. This can lead not only to a higher quality of life for individuals, but also to enhanced productivity for employers and greater prosperity for communities.
Ray Fabius, MD
Former Chief Medical Officer, Truven Health Analytics, and GE Global Medical Leader
The future of population health management (PHM) will be tied to the rise of cognitive computing, which uses massively parallel processing and artificial intelligence to convert unstructured data into structured data, search the medical literature, and find connections among myriad types of data. Clinicians can collaborate with cognitive computing systems, which learn from experience, to improve health care
Including social and economic factors as well as physical environmental factors, SDH accounts for much more of the variations in individual health than health care does.
Readmissions, which affect nearly a fifth of Medicare patients discharged from the hospital, are more numerous than they should be because of the fragmentation of our healthcare system. Multiple government programs have been established to address this problem.
Patient engagement is essential to improving health outcomes and is therefore an integral part of population health
management. Visits to physicians alone are not enough to sustain patient engagement, but the physician‐patient relationship is critical to success in this area.
Healthcare reform hasn’t solved the major problems of our system with cost, quality and access. To do that, we’ll need to achieve the Triple Aim, including finding a way to manage population health efficiently.
In 2010, the Affordable Care Act authorized a Medicare Shared Savings Program (MSSP) for accountable care organizations (ACOs), and the Medicare Access and CHIP Reauthorization Act (MACRA) of 2015 will continue to advance ACO models. Private payers are also contracting with ACOs. To succeed, ACOs must learn how to manage population health effectively.
The patient‐centered medical home (PCMH) movement is growing rapidly, with support from both private insurers and the government. While medical homes can’t achieve their full potential until they integrate with the medical neighborhood and start automating care management, new financial incentives will support PCMH evolution.
Clinical integration across providers and sites of care must come before ACOs and population health management. But most organizations lack the key components of clinical integration, including a robust IT infrastructure.
The government’s electronic health records (EHR) incentive program is designed to transform healthcare delivery and dovetails with other healthcare reform initiatives.
The use of data warehouses in combination with analytic tools falls short in the context of population health management (PHM), which requires comprehensive, scalable, and flexible health IT. To address the volume, velocity, and variety of data needed to support the manifold components of PHM, a big data solution is required. But this approach must be combined with other forms of IT to optimize care management, care coordination, and patient
"Dr. Richard Hodach’s book addresses the most critical elements for providers as they transform themselves to provide population health management. This book is a must read for anyone interested in population health."
"The key message of this timely book is that healthcare teams must adopt automation to deliver on the promise of population heath management. As my own organization has discovered, technology allows healthcare organizations to scale up quickly to prepare for value-based care. In this book are the practical strategies your organizations can embrace today."
“This book is a real gem! Health care transformation to improve health, quality, and cost (the Triple Aim) requires new payment for value, new delivery systems founded on patient centered medical homes, and new information systems to support care of populations. Dr. Hodach provides spot-on vision of how all three must work together, along with a detailed roadmap for success.”
“Dr. Hodach brings a wealth of knowledge and experience in the field of population health management. His keen insights into US healthcare transformation underscores the need for each healthcare system to have a well thought out and deliberate population health strategy.”
“While there are many books on this subject, the clarity of someone well-trained and experienced provides insight into the real, dysfunctional working of the healthcare delivery system while providing useful guidance on working within the rapidly evolving dynamic is has become.”
Use this infographic companion deck to share visual overviews of key topics in population health management. The 14 slides in this PowerPoint deck were created to be used together as a presentation, or separately as individual slides. Download these slides to brainstorm, shape, and convey your population health strategies.