Provider-Based Population Health

The goal of population health management (PHM) is to keep a population as healthy as possible. It is well known that about 5 percent of patients’ use 50 percent of healthcare resources and 10 percent use 70 percent of those resources. So, while PHM’s goal is to take good care of every person, organizations must pay special attention to these high-risk patients in order to lower population health costs to the maximum extent possible. Many organizations that are beginning to manage population health are focusing on the top 5-10% as a beginning strategy.

But keep in mind that this is the “tip of the iceberg”. As was learned by the disease management industry and emphasized in a recent publication by Ian Duncan, this could be a mistake. Below the waterline, so to speak are other cohorts of individuals with varying degrees of health risk and cost that will be the next individuals in the top 5-10%.

Beyond High-Risk Patients

So while PHM focuses partly on the high-risk patients who generate the majority of health costs, it has to systematically address the preventive and chronic care needs of every patient. Because the distribution of health risks changes over time, the objective is to modify the factors that make people sick or exacerbate their illnesses.

According to the Healthcare Risk Adjustment and Predictive Modeling by Ian Duncan, the implication of the analysis for predictive modeling may be seen in the transition of members between cost groups.

The first point to note is that the overall cost distribution remains relatively stable. Healthy members count for 69.5 percent of all members in the baseline year and almost 67.6 percent in the subsequent year. Within the Low cost category (67.6 percent), however, 85 percent of members (57.4 percent of the total 67.6 percent) were previously healthy, and 15 percent regressed from Chronic or Catastrophic categories to Healthy. Conversely, Catastrophic cost members were 1.8 percent of the population in the baseline year and 2.2 percent in the subsequent year. Fewer than one-third of these members (0.6percent of the total 2.2 percent) were previously catastrophic.

Importance of Intervention for All Risk Levels

This and other research has demonstrated that there is considerable movement between cohorts. Understanding that and being able to effectively and efficiently intervene is crucial. Or said another way, managing, outreaching and providing care and education to what is beneath the waterline is critical to success.

Leveraging Technology and Automation

Such an approach requires the use of automation. Not only are there not enough providers and care managers to manage every patient continuously, but PHM also involves a large number of routine tasks that do not have to be performed by human beings. Bringing modern information technology to bear on these tasks saves time and money and makes PHM economically feasible. Automation also allows organizations to better assess population needs and stratify populations based on geography, health status and utilization, and demographics.

So in the end:

• You have to accept that there is a roll for managing the complex patients with the highest costs at the top but if you don’t manage the patients across the population, you will fail in managing cost across the whole population.
• Being able to stratify the population by health risk and clinical care gaps is critical to be able to mange with limited resources.
• You need both a high-performance team as well as technology automated wherever possible to achieve the goals of population health management.

Download Phytel’s white paper on Population Health Management.

Recently, Physicians Practice published an article, “Top 4 ACO Considerations for Physicians.” It posed this question to readers: “Confused about how ‘accountable care’ will affect your practice? We’re here to help.”

Overall, this was a great article with comprehensive points detailing the challenges that physicians and practice administrators need to consider for an ACO model. However, I had some concerns over the prominence the article gave to using an EHR to mine population data. Specifically in point #2, the author briefly mentioned “Additional Responsibilities,” stating that staff needs to take on extra responsibilities for outreach to manage a population etc.

ACOs Will Benefit from Automation

While it’s true that EHR systems have the data, they are specifically designed to assist physicians to care for individual patients, not for managing a population or large patient panels, as ACOs will require. It is crucial that physicians and administrators understand the drastically different workflows and capabilities needed beyond the raw data, and that without proper automation, these “additional responsibilities” cannot be scaled to effectively impact the population and have the potential to break a practice’s back.

For ACOs to be effective, practices will need to closely track a patient’s care history to identify and meet their care needs. This includes monitoring a patient’s status between episodes of care so the practice can intervene proactively, give patients appropriate support, and engage them in their own care. More and more physician groups are using electronic registries and patient outreach programs to assist them in these efforts. By using the registry data from EHRs, these programs can send automated phone, e-mail or text messages to patients, telling them to make an appointment with their physician. Such tools enable physicians to practice at the top of their license and relieve their care teams from being overwhelmed with the responsibilities care management of large populations entail.

Engaging Patients, Managing Care

In effective ACOs, automation will be key to engaging the patient and managing their care, allowing practices to:

• Use registries to track the health status and care gaps of all patients
• Use proactive outreach to notify patients when they need care
• Manage more patients at different levels of risk
• Automate case management and transitions of care workflows
• Implement educational and operational improvement processes

The transition to ACOs and other emerging models of care is certainly top of mind for physicians right now so the more education and communication we can have on these issues, the quicker we’ll reap the rewards of better population health.

As a veteran of the disease management industry, and most recently, spending several years assisting provider organizations to manage their patient populations, I fully appreciate the problems inherent in traditional employer-based and health plan-sponsored disease management programs.

However, like Al Lewis, I also disagree with the statement that disease management is dead. Although some models have failed in the past, disease management models continue to evolve and are being transformed and incorporated into new physician-led population health models such as the Patient-Centered Medical Home.

Further, as health systems assume more financial risk and move towards Accountable Care Organizations, they will take on responsibility for managing entire populations and will be using population health tools and services that evolved from the original disease management principles and fundamentals.

I would like to emphasize several points that strongly indicate that disease management is alive and well.

Provider-Led Population Health

Today, there are a number of marketplace drivers such as Meaningful Use, Pay-for-Performance, PCMH and ACO and, not coincidentally, that share an underlying objective of provider-led population health. Population health means responsibility for everyone in the population: both those who are active health seekers and those who are not.

Achieving the IHI Triple Aim will require providers to provide patient-centered, personal, coordinated care across their populations. Provider-led population health will demand new tools and automation to be successful and will resemble those introduced and pioneered in employer and health plan-sponsored disease management programs (e.g., define population, identify care opportunities, stratify by risk, engage patients in care, proactively manage care, and continuously measure and monitor outcomes).

NCQA and Disease Management

I see no evidence that NCQA was in any way responsible for the demise of disease management. NCQA should be praised for their pioneering work on population health, quality reporting, disease management programs and HEDIS measures. They are now on the leading edge in the development of Patient-Centered Medical Home (PCMH) and the Accountable Care Organization (ACO) standards. Again, disease management principles and population health management are embraced in the new standards that NCQA has developed. The PCMH is widely endorsed and adopted. There are now 3302 NCQA-endorsed practices across the country and the number continues to grow.

Demonstration Projects

The demonstration projects from 2006 showing equivocal outcomes and financial results for disease management programs are dated and not a part of the newer provider-led delivery models. More recent demonstrations around PCMH, multi-payer pilots, and ACOs have shown successful health outcomes. Key components from the disease management industry and programs were incorporated into these new demonstrations.

• PCMH Demonstrations showed positive results which were attributed to dedicated care managers and data-driven analytics, principles used in disease management programs. (Fields, Leshen, and Patel) (Takach).
• Payers publishing positive results around ACOs are using all of the principles established in the disease management programs but in a new environment, featuring collaboration between payer and physician. (Higgins et all).

Implementation

One of the biggest issues in the early disease management programs was that the physicians and care teams were not involved in their implementation. Many of these programs were designed very well but were focused predominantly on patient education and behavior change. Doctors became very disappointed and distrustful and cited many reasons to not accept these programs as valid.

• Patients were getting enrolled into programs without their provider’s permission. Patients were getting information or education that was coming from the employer or health plan, rather than their physician.
• Results were coming back to physicians based on paid claims and were not timely or comprehensive, raising questions of accuracy. These reports used data that was old and not as rich as could be found in their offices, such as lab results and other biometric indicators.
• Doctors were frustrated by the multitude of payers offering DM programs to segments of their patient population and expecting doctors to follow different rules or incentives for different patients.

Doctor Involvement

Doctor involvement, not disease management, was one of the principle issues that needed to be changed. New legislation, payment reform/quality incentives, movement toward value-based accountability emphasizing the other marketplace drivers mentioned above will continue to embrace providers around improving quality and value. Provider organizations are being given the responsibility to manage their entire patient panel. To do so, providers will use the tools built and pioneered by the principles within disease management programs:

• Patients will be engaged proactively by their physicians and care teams, providing them with self-management tools and behavior change support to prevent disease and complications, and coordinating their care when other services are needed.
• Reports will use data from EMR systems, which are more timely and accurate. More importantly, physicians recognize that data in these reports are related to them and their patients, and they accept the information as opportunities to improve.
• Physicians and physician leaders are now beginning to think in terms of “my population”, which may include multiple payers, but one set of evidence-based standards and financial goals.

Conclusion

So, in the end, disease management is becoming transformed and deployed into new models, refining some aspects and putting the provider in the leadership role to achieve new thresholds of success.

References

• Fields, D., E. Leshen, K. Patel. 2010. Driving Quality Gains and Cost Savings Through Adoption of Medical Homes. Health Affairs. 29(5):819–26.
• Health Affairs, 30, no.7 (2011):1325-1334; Medical Homes Show Promising Results Reinventing Medicaid: State Innovations To Qualify And Pay For Patient-Centered Mary Takach
• Health Affairs, 30, no.9 (2011):1718-1727; Between Health Plans And Providers: Early Lessons From Accountable Care Models In The Private Sector: Partnerships; Aparna Higgins, Kristin Stewart, Kirstin Dawson and Carmella Bocchino

Until recently, a hospital’s responsibility for care may have ended when a patient was discharged. While health systems have used a variety of techniques to reduce readmissions, they have experienced mixed results. But new government incentives, plus a rising awareness of the need to improve patient safety, are placing an increased emphasis on discharge planning and post-acute care.

New CMS Regulations Begin October 2012

Front and center are the Centers for Medicare and Medicaid’s (CMS’) new regulations on preventable readmissions. Starting Oct. 1, 2012, hospitals with “excessive” readmissions—situations in which that number of patients readmitted to a hospital is significantly higher than expected—will lose a percentage of their Medicare reimbursement across the board. In FY 2013, the decrease can be up to one percent of reimbursement, rising to two percent in 2014 and three percent in 2015.

In the first year of this program, CMS will examine 30-day readmission rates for patients with heart failure, acute myocardial infarction, and pneumonia—three of the leading conditions for which patients are readmitted. Beginning in FY 2015, CMS may also scrutinize chronic obstructive pulmonary disorder and several cardiac and vascular surgical procedures.

Lowering Readmissions through Partnership for Patients

CMS has also launched other programs that might contribute to lower readmission rates. To begin with, the agency plans to spend $500 million—or half of the $1 billion earmarked in the Affordable Care Act for improving patient safety—to help hospitals and their community partners decrease readmissions over a five-year period ending in 2016. Through the government-sponsored Partnership for Patients, CMS will pay these “community-based organizations” a set amount per discharge for managing Medicare beneficiaries at high risk for readmission.

Other Incentives: Payment Bundling and ACOs

Two other CMS initiatives authorized by the health reform law are worth considering: payment bundling and accountable care organizations ACOs). Under CMS’ recently announced plan for its bundling demonstration, providers may choose among four different options. One option includes all care provided from admission to the hospital to 30 or 90 days after discharge. Another would cover only post-acute care for up to 30 days.

In both scenarios, providers would be paid on a fee-for-service basis, adjusted retrospectively for variance from a budgeted amount. While neither option penalizes providers for readmissions, both encourage improvements in the quality of post-acute care, which should reduce the number of readmissions.

Shared-Savings Program for ACOs

Finally, next year CMS will launch its shared-savings program for ACOs, which are groups of hospitals and doctors that are committed to raising the quality and lowering the cost of care. To receive financial rewards from CMS, these organizations will have to save money, which will give them a strong incentive to cut readmissions.

Nevertheless, it will be difficult for healthcare organizations to decrease readmissions significantly in our fragmented, uncoordinated healthcare system. While most of the levers of improvement are known, reengineering inpatient processes and engaging patients and outpatient providers remains challenging.

Leveraging Technology to Auomate the Post-Acute-Care Process

Fortunately, new applications of health information technology now offer inexpensive ways to automate post-acute-care processes. These solutions can raise the effectiveness of care managers, improve the communications between inpatient and outpatient providers, and make it easier for patients and caregivers to absorb and apply the knowledge required for self-management of complex conditions.

A recent consensus report by The Commonwealth Fund emphasizes the role of health IT in care coordination, while asserting that today’s information technology is inadequate to the purpose.

“Anchoring the electronic health record (EHR) in the traditional visit-based care delivery model limits the potential of the medical home to generate paradigm-shifting care delivery transformation and the positive outcomes it promises…Health IT requires new functional capabilities, such as multiple team member access and permissions, care management workflow support, integrated personal health records, registry functionalities, clinical decision support, measurement of quality and efficiency, and robust reporting.”

The bulk of the technologies required to achieve these goals are already available. Among the reasons they’re not being properly deployed in most cases are these:

• Electronic health records are not designed to do population health management or care coordination.
• Registries tend to be focused on patients with particular conditions, rather than entire populations.
• Care management workflow support is still a relatively new concept, but one that more and more groups are embracing.

Group practices that are trying to transform themselves also have to manage a variety of other issues, including cultural barriers to change, potential infrastructure issues, and limited reimbursement for non-visit and non-physician care.

Key Building Blocks

Technology tools are also being used to improve communications across care settings, especially during transitions of care. These range from traditional point-to-point interfaces to physician and patient web portals to secure electronic messaging. Even computer faxing has its place as the industry moves from paper to electronic documentation of care.

The automated stratification of patients into different risk categories is also important to groups seeking to do population health management. For example, UNC Healthcare uses a health risk assessment (HRA) survey to find out how sick each of its patients with diabetes is. Then it uses an advanced patient registry and evidence-based algorithms to drive team-based care for each of those patients, depending on the severity of his or her condition.

Many practices use electronic registries to supplement their EHRs. These registries compile lists of subpopulations that need particular kinds of preventive and chronic care, such as annual mammograms for women over 40 or HbA1c tests at particular intervals for diabetic patients. The continuously updated data in the registries comes from EHRs, practice management systems, or a combination of the two. Evidence-based clinical protocols, which can be customized by physician practices, trigger alerts in the registries. When a registry is linked to an outbound messaging system, patients are notified by automated telephone, e-mail or text messages to contact their physician for an appointment. Some registries can also send actionable data to care teams prior to patient visits.

Care Coordination Leverages IT

The overall lesson to be drawn from the efforts to improve care coordination will require the use of information technology. The identification of patients with particular conditions, health risk assessments, the ability to send care gap alerts to providers, the care management of chronically ill patients, tailored patient education, and persistent reminders to patients to get the care they need—all of these interventions require some degree of automation to be performed in a timely, consistent, cost-effective manner.

The recent advances in health IT and further developments in this vital field will continue to support and enhance care coordination as it expands across the spectrum of care. Assuming that payment methods support coordinated care, we can look forward to a proliferation of new IT tools that will help turn the dream of affordable, high-quality healthcare for all into a reality.

As we all know, the meaningful use requirements of the HITECH Act are designed to facilitate quality improvement, better care coordination and population health management.

However, it may not be widely known that physicians will typically need to use supplemental information technologies along with their EHRs to show meaningful use consistently over the next five years.

While EHRs are good at point-of-care interactions and decision support, physician practices and hospitals will need to acquire certain additional technologies to address quality improvement and manage population health. These may include electronic registries, multiple outreach and communications methods, software that can calculate the metrics required for quality reporting, and solutions that extend the reach of the provider and care team to help keep patients engaged in their care.

While the stage 2 and 3 requirements have yet to be drawn up, the final rule for stage 1 already shows that HHS remains on course to deploy meaningful use as a lever to get physicians to use EHRs for quality improvement and population health management.

To meet stage 1 meaningful use requirements for population health management, supplemental technologies may be required. For example, the patient reminders in stage 1 require condition identification first and then gaps for the condition which is performed by the EMR technology. Hundreds and sometimes thousands of patients may need to be contacted. This can consume many hours of staff time, calling or mailing these individuals. Technology can take this same list and automatically generate phone call, email messages, and text messages, analyze the response rate and resend the message. This activity is continuously being performed in the background rather than a one-time action taken on a condition list and can provide reports to the team on a regular basis to track effectiveness.

To meet meaningful use requirements during all stages, healthcare organizations and providers will be seeking the ability to generate lists of patients with specific chronic conditions or preventive-care needs, collect and report quality data, and generate condition-specific educational materials for patients. To do this manually would be a burden on care organizations. The solution lies in finding a way to automatically identify, engage, and track patients who need preventive and chronic care services.

What all of these methodologies have in common is that they automate the work of monitoring, educating and maintaining contact with the patient population that meaningful use will require. Especially at a time when primary-care providers are in short supply and stretched thin, it is essential to provide this level of automation so that the routine, repetitive work can be done in the background, rather than taking up the valuable time of doctors and nurses.

Information on the care gaps of specific patients should be automatically generated and provided to care coordinators and care managers within practices. These clinical staffers can then use this information to prepare doctors and nurses for patient visits. Between visits, they can use the population health improvement technology to make sure that patients get their needs addressed and come back for follow-ups. The technology solution does the heavy lifting, increasing care managers’ productivity and allowing practices to do more with fewer personnel.

To attest that a physician has gathered data on at least six of the 38 quality measures, practices will have to identify the numerator and the denominator on each metric. For example, if smoking cessation advice is the measure, an organization must be able to identify the number of smokers in the practice and what percentage of those patients received physician counseling.

By combining EHRs with these automated approaches, physicians can show meaningful use, qualify for medical home certification, obtain pay for performance incentives, and prepare themselves for the value-based reimbursement systems that are down the road. At the same time, these adjunctive technologies enable physicians to gather the quality data they will need to report to Medicare and private payers in an automated manner. And by giving care teams real-time data on the services that patients need when they’re in the office, these methods empower physicians and other clinicians to improve quality and engage in productive conversations with patients about how they can maintain or restore their health.

It’s clear that the meaningful use requirements of the HITECH Act can be met using EHR and supplemental technologies to keep patients engaged and coordinated in their care.

Dr. Richard Hodach is Chief Medical Officer at Phytel.