Multiple Chronic Conditions External Partner Meeting
Friday, November 21, 2008
9:00 a.m. to 12:30 p.m.
Room 505A, Hubert H. Humphrey Building
200 Independence Avenue, SW
Washington, DC 20201
Speakers
- Dr. Joxel Garcia, MBA; Assistant Secretary for Health; Department of Health and Human Services (DHHS)
- Gerard Anderson, PhD; Professor and Director; Center for Hospital Finance and Management; Department of Health Policy and Management; Johns Hopkins University
- Dr. Chad Boult, MPH, MBA; Professor and Director; Lipitz Center for Integrated Health Care; Department of Health Policy and Management; Johns Hopkins Bloomberg School of Public Health
- Mary D. Naylor, PhD, FAAN, RN; Professor and Director; NewCourtland Center for Transitions and Health; University of Pennsylvania School of Nursing
- Dr. Mona Sarfaty, Research Assistant Professor, Department of Health Policy, Jefferson Medical College
- Julie L. Lewis, Assistant Director for Health Policy, Center for Health Policy Research, The Dartmouth Institute for Health Policy and Clinical Practice
- Christine Vogeli, PhD; Institute for Health Policy; Massachusetts General Hospital
Dr. Anand Parekh
Dr. Parekh noted that the number of Americans who suffer with multiple chronic conditions (MCCs) is increasing, particularly as the population ages. The MCC Work Group was convened so that the Office of Public Health and Science (OPHS) could learn more about the MCC population, what is being done to improve patients’ health status, and what more could be done.
He said that the work group has worked diligently over the last several months looking at existing programs and initiatives that focus on the MCC population. The purpose of this day’s meeting, said Dr. Parekh, is to hear from national experts in the field and learn more about:
- Who has multiple chronic conditions?
- What is their health status?
- What interventions have been shown to improve their health status?
- How can/should the public health and health care systems change to improve the health status of those with MCCs?
Dr. Parekh then introduced Assistant Secretary for Health Joxel Garcia, highlighting Dr. Garcia’s experience as deputy director of the Pan American Health Organization and commissioner of the Connecticut Department of Public Health.
Dr. Garcia related his experience of accompanying visiting nurses on their rounds and witnessing first hand some of the challenges posed to the health care by MCC patients. Many people had multiple diagnoses and filled prescriptions at two or three pharmacies. For many patients, pharmacies were the most frequent point of contact with the health care system.
Dr. Garcia expressed hope that the day’s meeting would illuminate not only advances in the science of MCC, but progress from diagnosis to multiple treatments. “We aren’t looking for the problems,” he said. “We know all the problems already. We’re looking for answers and solutions.”
He noted with optimism that his meeting with the incoming Obama Administration’s transition team included a discussion of MCCs, indicating that the fruit of the working group’s efforts will remain of interest in the new regime.
Dr. Parekh then requested that working group members introduce themselves.
Panel #1
What do we know and not know about the health status, quality, and costs of care for individuals with multiple chronic conditions?
Gerard Anderson, PhD; Professor and Director;
Center for Hospital Finance and Management;
Department of Health Policy and Management;
Johns Hopkins University (JHU)
Accompanying Documents:
Chronic Conditions; Chronic Conditions: Making the
Case for Ongoing Care
Dr. Anderson began by stating that “we’re going to be talking about numbers, but it’s really about people.” According to the World Health Organization:
- Chronic diseases are now the major cause of death and disability worldwide.
- Non-communicable conditions (NCDs) including cardiovascular diseases (CVD), diabetes, obesity, cancer, and respiratory diseases now account for 59 percent of the 57 million deaths that occur annually worldwide.
- More importantly, NCDs account for 46 percent of the global burden of disease.
NCDs are not limited to high-income countries like the United States and Canada:
- 80 percent of NCD deaths occur in low- and middle-income countries.
- Poor people in poor countries have greater difficulty preventing and treating NCDs than rich people in poor countries.
- Low-income countries are dependent on international aid agencies (USAID, Gates Foundation, etc.), but these organizations are not addressing the tremendous need.
Example: The burden of disease in Bangladesh.
- 46 percent is due to NCDs.
- 46 percent is due to CDs.
- 8 percent is due to injuries.
Despite these statistics, there are virtually no programs in Bangladesh to prevent or treat NCDs.
NCDs in the United States
The United States has the highest level of NCDs in the world. This is not due to a longer lifespan, since U.S. life expectancy is below many industrialized countries. Obesity is one of the main factors.
- The number of Americans with NCDs is increasing rapidly, according to the Rand Corp. The total was 118 million in 1995; it is projected to be 171 million by 2030.
- Hypertension is the most common NCD across all ages.
- One in four Americans has two or more chronic conditions.
- Two thirds of children with chronic disease have only one; one third of children with chronic disease have two or more.
- Most adults with NCDs have more than one condition. Almost all seniors with NCDs have more than one condition.
NCDs and Health Care Costs
- People with NCDs accounted for 85 percent of all health care spending in 2004, according to the Medical Expenditure Panel Survey. This percentage has been increasing over the years with each survey.
- Average annual per capita spending on health care increases with the number of chronic conditions, ranging from $1000/year for one condition to $16,000/year for eight or more conditions. Many health care programs lack catastrophic caps to protect multiple NCD patients.
- People with multiple NCDs fill more prescriptions and spend more on drugs. The average annual number of prescriptions filled by people with a single condition is 7.5 at a cost of $476; those with five or more conditions fill 57.1 prescriptions and spend $3,799. Harmful drug interactions are common because not all of the patients’ doctors are aware of the drugs being prescribed by other practitioners.
Research and the Health Care System
A Johns Hopkins survey of practicing physicians found that they do not believe that their clinical training has prepared them to:
- Coordinate in-home and community services (66 percent).
- Educate patients with chronic conditions (66 percent).
- Manage the psychological and social aspects of chronic care (64 percent).
- Provide effective nutritional guidance (63 percent).
- Manage chronic pain (63 percent).
The medical education system needs to change because it is not oriented towards multiple NCDs. This reorientation must begin with research. The Medicare system works for acute illnesses and single conditions, but about 40 percent of the Medicare population has not been included in clinical trials, and these are generally the patients with two more chronic conditions. Published clinical trial results cannot provide proper guidance for the treatment of patients with multiple NCDs because these people have not been included in the trials. This lack of evidence-based guidance results in a lot of variation in medical practices.
The more chronic conditions a patient has, the more doctors he or she sees, often resulting in poor care coordination. This lack of coordination in turn increases the chances for inappropriate hospitalization, which climbs dramatically depending on the number of chronic conditions.
One quarter of individuals with a chronic condition—39 million people—also have limitations for daily living activities. Most of those 39 million people have multiple NCDs.
Discussion
Dr. Anderson and work group members participated in a question/answer and discussion session on his presentation:
Alzheimer's. ARC recognizes Alzheimer’s as a chronic condition. JHU researchers originally tried to rank chronic conditions in order of their significance and did not get far. In looking at financial factors, for example, the dramatic rate at which medical expenses climb based on the number of NCDs is more significant than the costs of individual conditions.
Medical Recordkeeping. A case controlled study failed to show that electronic records cut costs, but Dr. Anderson said that this result is due to the difficulty of targeting patients with multiple NCDs. The Centers for Medicare and Medicaid Services (CMS) have had trouble identifying such people to enroll them in studies and target them in programs. Electronic recordkeeping probably will not have that much of an impact on addressing people with a single NCD, according to Dr. Anderson, but there is cost saving potential with multiple conditions. As far as using technology to overcome fragmented care and multiple prescriptions, there are a few examples of success in large integrated health care systems that can imbed relevant information in their sophisticated recordkeeping software.
NCD Treatment Success in Developing Countries. Most treatment successes are region-specific because regions are not communicating with each other. Brazil, for example, has developed good exercise programs, and a smoking cessation program has succeeded in Brazil, Mexico, and Uruguay. An exception to this regionalization is a nutrition program that has succeeded in numerous low- and middle-income nations, an important result as the high fat/junk food culture and urbanization continue to spread across the globe. Obesity and poor nutrition will continue to be growing problems and researchers are working with USAID, the World Bank, and other organizations to address the situation.
HHS Wish List for Incoming Administration:
- Research - Get patients with multiple NCDs into clinical trials sponsored by the National Institutes of Health and other agencies. A research base is needed to demonstrate effective care.
- Education - Engage CMS, the Health Resources and Services Administration (HRSA) and others to help provide better training for care coordinators and move medical education into the 21st century so that practitioners are taught to identify and treat patients with multiple NCDs.
- Payment – Medicare and Medicaid should pay for treatment coordination, which requires quality metrics. Risk adjustments under managed care do not take multiple NCDs into account and these conditions are underpaid.
Medical Education. The young generation of medical students is receptive to learning about NCDs. There is a real foundation of knowledge, but it has not been disseminated the way that it should have been. Geriatrics is a strength of JHU, and those in the geriatric field already know quite a bit about treating NCDs. U.S. medicine is dominated by specialists, however, and the trick will be to get the urologists, endocrinologists, and others involved.
Dr. Parekh interjected that half of the patient population with multiple NCDs is under 65 years of age, so it is not just the elderly who need attention.
International Involvement. Lack of awareness is the biggest obstacle to international organizations’ involvement in addressing multiple NCDs. The next largest obstacle is identifying specific programs that organizations can implement—if they allocate funding to NCDs what is the first thing that they should address? Nutrition? Exercise? Medications? As an example of a funding disparity, only two percent of deaths in Eastern Europe and Eurasia are due to childbirth, yet the problem attracts significant funding.
Integrated Data. Many diseases are backed by powerful foundations, but multiple NCDs lack such strong advocacy. There is good information on specific chronic diseases, but data is not collected on multiple NCDs. Researchers have started recognizing multiple NCDs’ prevalence and know that at least 25 percent of Americans have multiple NCDs. But there is little good data on topics such as the most prevalent patterns and treatment outcomes. Even specialists frequently lack an understanding of how other conditions affect their field—how diabetes interacts with cardiovascular disease, for example. Clinical trials are needed for people with multiple NCDs.
Panel #2
What interventions have been shown to be beneficial or cost-effective in improving or maintaining the health status of individuals with multiple chronic conditions?
Dr. Chad Boult, MPH, MBA;
Professor and Director;
Lipitz Center for Integrated Health Care;
Department of Health Policy and Management;
Johns Hopkins Bloomberg School of Public Health
Accompanying Document:
Successful Models of Chronic Care
Reducing the burden of chronic diseases involves:
- Primary prevention.
- Improving health care for people who have these diseases.
Chronic care in America is: provider-centric, idiosyncratic, fragmented, uncoordinated, and expensive.
Chronic care in America needs to be: patient-centered, evidence-based, comprehensive, coordinated, and affordable.
American chronic care ranked last among six developed nations in the Commonwealth Report 2007. The cost of care in the United States was two to three times greater than costs in each of the other five nations, and the value (quality/cost) was the lowest.
A lot of work has been done to create alternative chronic care models that narrow the “quality chasm” and/or contain costs. “Diffusibility” is a key factor—not only should a model work well, it should work widely across the health care system.
Methods of Research to Find Successful Models of Care
- Search the scientific literature for models of care that have produced positive results (improved the clinical outcomes, lowered costs, or accomplished both).
- Tabulate the evidence for promising models.
- Classify the strength of the evidence.
- Rate the successful models’ diffusibility (an expert panel rated each model based on six criteria for diffusibility).
- Explore policies to promote use of models.
Results
Ten successful models were selected then:
- Categorized according to care setting (community, transitional, institutional).
- Presented according to how they improved outcomes and contained costs.
- Rated according to diffusibility potential.
Successful potentially diffusible primary care models included:
- Advanced-practice nurse (APN)-physician team (for dementia patients).
- Interdisciplinary teams for patients with congestive heart failure.
- Guided care for multi-morbid patients.
- Transitional care.
- APN-physician dyads (for nursing home patients).
Potentially diffusible models that support primary care included:
- Care management (for congestive heart failure).
- Pharmaceutical care (pharmacist counsels patients taking multiple drugs).
- Self-management training.
- Proactive rehabilitation (reaching older people who are having trouble living in their homes before an event takes away their independence).
- Caregiver support/education (typically for family members of MCC patients).
Federal Policy Options
Medicare
- Care management and shared savings payments to “medical homes.”
- Payments to pharmacists, nurses, and rehab therapists who supplement primary care.
- Payments to organizations that provide nurse-based transitional care from hospital to home.
CMS is conducting a demonstration under which the agency will pay primary care providers to assemble a medical team to coordinate care.
Administration on Aging
Fund state and area agencies on aging to provide:
- Self-management courses accessed as a community service rather than a medical service.
- Caregiver education/support. This community service is quite effective but is not available to most people.
Accompanying Document:
Guided Care: A New Model of Care for Older Americans with Chronic Conditions
In moving from a model of acute care to one of guided care, we must take the system that we have and make it relevant to MCC patients, said Dr. Boult.
The Guided Care Model
GCN = guided care nurse, a specially trained registered nurse in a primary physician’s office. GCNs collaborate with physicians in caring for 50-60 high-risk older patients with chronic conditions and complex health care needs. The GCN’s efforts are supported by electronic health records.
Guided Care Nurse Activities
- Assess needs and preferences in a comprehensive two-hour process covering a patient’s environment and family.
- Create an evidence-based “care guide,” which is an action plan for the patient to take home. It can include what to eat, red flags for health concerns, when to call the doctor, etc.
- Monitor patients proactively. Go over the action plan including medications, activities, and how well the patient is following guidance. Ensure maximum adherence to the plan and identify emerging problems. When adherence obstacles crop up, conduct motivational interviewing in order to change the patient’s behavior by making adherence part of his or her own goals. For example, use the fact that a patient wants to keep up with her grandchildren and remain in her own home as motivations for her to stick to her medication and exercise regimes.
- Support chronic disease self management.
- Communicate with providers in emergency departments, hospitals, specialty clinics, rehab facilities, home care agencies, hospice programs, and social service agencies. Give providers a two-page summary of plan. When this occurs, other providers tend to follow the guide.
- Smooth transitions between sites of care. This is one of the most important parts of the model. The GCN provides the transition plan to the in-patient hospital team, prepares the patient’s family for the transition, checks on the medications dispensed by the hospital, makes sure that the patient can use any new equipment issued by the hospital, coordinates with the visiting nurse, and communicates all information to the primary care doctor. Currently most doctors do not know that a hospitalization occurred, let alone the details.
The model is too expensive to provide guided care to everyone age 65 or over. CMS reviewed a year’s worth of claims and discovered that the 25% highest risk patients accounted for 80% of Medicare spending. A pilot test targeting 150 such high-risk patients was conducted between October 2003 and September 2004 in a primary care practice in Baltimore with two internists and a GCN. The outcomes were a higher quality of care with 23 percent lower costs.
Randomized Trial
JHU researchers are now two years into a 30-month randomized trial of the Guided Care Model that will collect data until June 2009 if fully funded. The trial:
- Includes 904 high-risk older patients of 49 community-based primary care physicians practicing in 14 teams in the Baltimore-Washington, DC area.
- Assigns physician/patient teams randomly to Guided Care or “usual” care.
- Will have measured outcomes at six, 18, and 30 months.
Some outcomes so far:
- Guided Care patients are twice as likely to assess their care at the highest level.
- A year into the trial, the satisfaction level of all Guided Care physicians was up and the level for all “usual” care physicians was down.
- The net savings on costs of care per caseload (55 patients) was $75,000.
- Anecdotal evidence points to Guided Care outcomes as improving as practices gain experience with the model.
- A Guided Care implementation manual will come out in January.
- More information is available at www.guidedcare.org.
Mary D. Naylor, PhD, RN,
Marian S. Ware Professor in Gerontology and Director,
New Courtland Center for Transitions and Health, University of Pennsylvania School of Nursing
Accompanying Document:
Evidence-Based Best Practices for Multiple Chronic
Conditions
University of Pennsylvania researchers are trying to identify opportunities for evidence-based practices in treating MCC patients. Dr. Naylor presented a transitional care model (TCM) and a chronic care model (CCM) that embody such practices. She singled out the circumstances of one patient—Mr. Jenkins—along with his wife and their family—as an example of a situation that could benefit from evidence-based models for MCC care.
MCC patients experience an ebb and flow between acute and regular health care systems. Mr. Jenkins, who is a 76 year-old alumnus of a prestigious business school:
- Has a history of seven chronic conditions.
- Sees six specialists—his primary care physician has retired.
- Takes 12 prescribed medications daily and is coping with dietary restrictions.
- Has had three recent hospitalizations for heart failure and diabetes.
- Has health problems that are increasingly interfering with his lifestyle.
- Has a wife showing signs of cognitive impairment.
Mr. Jenkins has serious challenges in coping with his situation even thought he is health literate and has strong financial and family support. Less fortunate MCC patients are even more challenged.
Transitional Care
Dr. Naylor defined transitional care as a range of time-limited services that complement primary/chronic care and are designed to ensure health care continuity and avoid preventable poor outcomes among at-risk populations as they move from one level of care to another, among multiple providers, and/or across settings.
The Case for Transitional Care
- High rates of medical errors.
- Serious unmet needs.
- Poor satisfaction with care.
- High rates of preventable readmissions—patients transitioning out of the acute phase often get in trouble once they are at home.
- Tremendous human and cost burdens—the economic burden is often highlighted, but the human impact is frequently underestimated.
The transitional care model includes:
- Screening – Over the last 20 years, the percentage of those screened who meet the risk criteria for a poor discharge from the hospital has increased from 20 percent to 40-45 percent.
- Engaging the elder and the caregiver - the transitional care nurse (TCN) begins to work right away after the emergency screening. The TCN focuses on trying to do a good job with the patient while he or she is hospitalized in order to prevent adverse events.
- Managing symptoms - TCNs spend a lot of time managing symptoms, including working with the pharmacist to streamline medication.
- Educating and promoting self-management - Experience has shown that it is unrealistic to assume that a patient can automatically take on the responsibility of managing his or her care if providers have not been doing a good job at it. An important aspect is educating a patient to know when he or she is getting into trouble and to seek help.
- Collaborating.
- Assuring continuity.
- Coordinating care.
- Maintaining relationships.
Unique features of TCM:
Care is delivered and coordinated…
…by the same TCN. The nurse follows the patient through the entire event. Rather
than focus on handoffs, the TCN focuses on hospital events to prevent adverse
outcomes.
…across settings.
…seven days per week for a mean of two months.
…using evidence-based protocol.
…with a focus on long-term outcomes.
Findings from Randomized Clinical Trials
Trials were funded by the National Institutes of Health, the National Institute of Nursing Research, and the National Institute on Aging (1990-2010). NIH has spent $10 million so far and is in its 5th clinical trial. Across all random controlled trials TCM has consistently:
- Increased the length of time to the first re-hospitalization.
- Decreased total re-hospitalizations for all causes.
- Increased patient satisfaction.
- Improved physical function and quality of life (results from most recently completed trial only).
- Decreased total health care costs by $5000.
Barriers to TCM’s Wide Application
The fundamental barrier to TCM is cultural. The model challenges the traditional medical culture, practice, and care-giving models/roles of acute care in favor of collaboration and quality of management. Other challenges include:
- Lack of Medicare reimbursement.
- Lack of defined quality and financial incentives for collaboration across settings of care.
Translating TCM into Practice
In order to make the science on TCM more widely known, the University of Pennsylvania research team has formed partnerships with Aetna Corporation and Kaiser Permanente to test “real world” applications of research-based TCM for high-risk elders. Progress to date includes:
- TCM proposed as part of Aetna’s 2009 Strategic Plan.
- Continued data collection by Kaiser.
- Adoption of TCM by the University of Pennsylvania Health System and reimbursement of TCM by Blue Cross and Aetna for their members.
Economic Factors
- Total health care savings include hospital, physician, and visiting nurse services.
- Savings do not account for medications and out-of-pocket costs.
- Total benefits of TCM to the entire Medicare and/or chronically ill populations (e.g. quality-adjusted life years) have not been calculated. If the health care system does better at avoiding preventable poor outcomes it can focus instead on risk reduction and prevention.
The Chronic Care Model
Many studies have tested one or more dimensions of the chronic care model (CCM), but on patients with only one condition.
The Case for Chronic Care
- Lack of holistic oversight and monitoring - solutions must be more comprehensive than just medical management of people with MCCs and deal with factors such as depression and cultural components.
- Inadequate transfer of information.
- Poor communication among patients, family caregivers, and providers.
- Limited engagement of patients in care.
- Inefficient care—overuse, misuse, and under use of services; ineffective use of health care providers.
Mr. Jenkins needs both TCM and CCM. A strategy of focusing on prevention is helpful, but as the population ages and people acquire more that one condition, something more will need to be in place. TCM is an important complement to CCM. The health care system needs to get both in place by the time Boomers begin placing heavy demand.
Discussion
Drs. Boult and Naylor and work group members participated in a question/answer and discussion session on the presentations:
Identifying Patients. Dr. Naylor explained that university researchers developed risk screens. Aetna already had a risk screen. Researchers made a business case that TCM could add value because a screening profile identifies appropriate candidates. Researchers shared their screening tool, and the medical community integrated it into workflow processes.
Engaging Physicians. In the case of Aetna, nurses reached out to physicians to make sure that they understood the TCM process, including identifying the two or three issues that would likely result in an acute event. With Kaiser Permanente, hospital-based nurses identified patients at admission. These nurses reached out to everyone involved in patient care in the hospital and in the home after discharge. Dr. Naylor said that “physicians love it. They develop trust and confidence that somebody else is their eyes and ears” to assure proper patient care.
Coordinating Health and Social Services. Dr. Naylor explained that TCNs work with a patient an average of two months. During that time, TCNs identify people in the community, including volunteer patient advocates, who can help. TCNs are trying to position the patient for the long haul and often need to find a primary care provider in addition to the specialists treating the patient in the hospital. The TCN starts with what the patients want to achieve, something that is vital to preventing an acute care episode. Dr. Boult noted that after analyzing claim data, Medicare, Kaiser Permanente and Tricare are ensuring that nurses have a database of all community resources.
He said, however, that health care providers must do a better job of referring patients to support groups and lay leaders. Dr. Naylor added that the health care system did not do a good job early on at coordinating social services, which has a negative impact on outcomes. Now TCNs focus on building partnerships right up front, functioning as both a direct deliverer and coordinator of services. The Alzheimer’s Association is an example of an advocacy group that coordinates with TCNs and has an advisory group to maximize access to social services.
Dr. Parekh asked what evidence exists for engaging community services. Dr. Naylor highlighted their benefit in educating about self care. “When we get to people with multiple chronic conditions, what is reasonable for self-care? We’re not there yet with understanding coordination, so how can we expect the patient to be? We’re teaching people how to identify when they’re running into trouble and who to call about it, and making sure that whoever is called is responsive.”
An audience member emphasized the importance of avoiding the medicalization of social service interventions by not providing them through the medical system. Dr. Boult agreed, saying researchers did not want TCN nurses leading self-management support groups and relied instead on lay people with chronic conditions.
Identifying Successful Interventions. Dr. Naylor pointed out that identifying the core elements of successful interventions can become complicated because patients’ needs change substantially from the beginning to the end of the intervention period. Electronic records can be helpful, but at some level, more data is not the answer. Patients need individual care and assistance learning to manage in their day-to-day environment. Successful intervention is a multi-dimensional process.
Dr. Anderson said that JHU researchers looked at intervention programs that were identified as successful by patients to see what those programs have in common.
Dr. Boult’s approach was identifying interventions that had already been shown to be effective, then combining them in innovative ways. “We already have some information on what works best for whom,” he said.
Patient Values in APN Nursing. Patients’ view of their recovery is key to a successful intervention, said Dr. Boult. APNs ask patients to name their top three 3 goals, with responses varying widely. A typical reply might be “I wish that I could walk to church” or “I want to make it to my granddaughter’s graduation.” Nurses use these goals to develop a plan, they check with the physician to see if the guidelines fit the patient. If a patient has declared that he is not going to stop smoking, for example, the medical team may put off the issue until some trust has developed between practitioners and patient.
Panel #3
What changes in the health care delivery system or the public health system could lead to improvements in the health status of individuals with multiple chronic conditions?
Dr. Mona Sarfaty, FAAFP,
Research Assistant Professor, Department of HealthPolicy, Jefferson Medical College
Accompanying Document:
Improving Outcomes in People with Multiple Chronic Conditions: through changes in medical care delivery,use of community resources (public health), and caremanagement (including self management)
The ranks of primary care physicians are thinning due in large part to burnout, and trainees are shunning the primary care field. As the primary care crisis grows, three promising trends are coalescing: innovations in primary care practice, the quality of care movement and payers’ interest in rewarding improved quality, and the chronic care model. A sign of the changing times is the American Medical Association House of Delegates’ endorsement of paying primary care practitioners to coordinate the full range of a patient’s care by a “medical home.”
The chronic care model, formulated in the 1980’s by Dr. Ed Wagner, sees the future of primary care—and indeed the entire health care system—as depending on its ability to provide high quality patient-centered care to those with chronic conditions. The model is based on varying amounts of qualitative evidence as well as results from random controlled trials. Studies that evaluate the model are available at www.improvingchronicillness.org.
The growing need for chronic care is revealed by statistics reported by the Partnership for Solutions on the average annual number of prescriptions filled by a patient based on the person’s number of chronic illnesses. Patients with one illness filled 10.4 prescriptions annually while those with five chronic ailments fill 49.2 prescriptions (statistics include refills). Another revelatory statistic shows that among adult MCC patients who saw a medical specialist, 22 percent saw doctors who lacked information about the patient’s medical history.
The chronic care model calls on the community and the health system to support the key elements of good chronic care—an informed, activated patient and a prepared, proactive practice team that productively interacts to produce improved outcomes.
Prepared practice team – At the time of the visit, members of the team have relevant patient information, decision support, people, equipment, and the time required to deliver evidence-based clinical management and self-management support.
Informed, activated patient – The patient understands the disease process and realizes his/her role as the daily self manager. Family and caregivers are engaged in the patient’s self-management. The provider is viewed as a guide on the side, not the prime actor.
Recent research ranks the following health service delivery techniques to improve diabetes as most effective: patient education, case management, team changes, clinical education, and self management. The Community Guide from the Centers for Disease Control also recommends disease and case management as effective interventions for diabetes, along with self-management education in community centers and in the patient’s home.
One key to a balanced health care delivery approach for those with chronic conditions is the “patient-centered medical home,” said Dr. Sarfaty. The American Academy of Family Physicians, American Academy of Pediatrics, American College of Physicians, and American Osteopathic Association developed a consensus statement on the elements necessary for a patient-centered medical home. Many prestigious medical organizations have also endorsed the statement, which includes:
- A personal physician.
- A whole person orientation.
- Coordinated and integrated care.
- Safe and high-quality care (e.g. evidence-based medicine, appropriate use of technology, continuous quality control).
- Enhanced access to care, meaning that patients get prompt responses.
- Insurance payments that recognize the added value provided to recipients who have a patient-centered medical home.
National Center for Quality Assurance criteria for a patient-centered home:
- – Access and communication.
- – Patient tracking and registry functions that allow a physician to find records of specific patients and observe what their experience has been over time.
- – Care management that makes staff available to help patients and their families manage the problems of everyday life as they provide self care and navigate the health care system.
- – Evidence-based patient self-management support that helps people determine what actions they need to take to manage their situation.
- – Electronic prescribing that helps eliminate the safety hazards of redundant identifications and conflicting medications that are hard to identify from paper records.
- – Test tracking to eliminate the current problems caused by a large number of MCC patients either receiving redundant tests or not receiving necessary tests.
- – Referral tracking – Ensuring that patients are seen by appropriate practitioners.
- – Performance reporting and improvement – The practice receives reports on quality, conducts studies to assess its own performance, and provides feedback at the practice level, where it is most effective.
- – Advanced electronic communications to interface with the outside world.
Issues that remain to be addressed:
- Although payers are interested and hopeful about the patient-centered medical home as a model for chronic disease care, reimbursement rates are still the main reason that primary care practices are squeezed.
- Intervention and how to implement care must be taught across medical subjects at educational institutions.
Julie Lewis, Health Policy Director, The Dartmouth Institute for Health Policy & Clinical Practice
Accompanying Document:
Aligning Goals and Incentives: Delivery System Reform
Lewis noted that according to Congressional Budget Office statistics, the U.S. regions that spend the most on health care ($10,000-$12,000 per capita) spend up to three times as much as lower spending regions ($4,000-$5,000). High-spending regions do not necessarily represent prosperous areas—some are low-income locations.
High-spending regions:
- Generally do not emphasize evidence-based quality.
- Have patients that are much more likely to have 20+ physicians, making it hard to track aspects of health care.
- Foster the perception in physicians’ of worse communication with patients and greater difficulty in coordinating care.
- Foster the perception in patients of having worse access to primary care.
Key elements that drivethe use of supply-sensitive services include:
- A capacity and payment system that keeps all providers busy
- Clinical judgment based on strong evidence.
- A greater likelihood for provider intervention in gray areas.
Important strategic aims for delivery system reform include:
- Organizational accountability for capacity, cost, and quality—something that the current medical delivery system does not have.
- Engaged patients, informed choices.
- Meaningful measures of system performance.
- The right workforce to lead the change.
Interim steps that are crucial to long-term reform include:
- Local integration through support for interoperable electronic health records.
- Network-level quality measurement, which Lewis categorizes as “much better than physician-level” measures.
A Dartmouth-Brookings collaboration called Accountable Care Organizations (ACOs) aims to foster local accountability for costs, quality, and capacity. A version of the “medical home” function will be included as part of integrated, high-quality, coordinated care. Performance measurement will play a “substantial” role in ACOs, said Lewis, and will occur at the network level with an emphasis on technical quality, care coordination, and patient experience.
ACOs function better, she concluded, when they are part of a “suite” of health care reforms and when medical homes are part of an integrated system. Multiple initiatives within the ACO model produced $800 million in target expenditures and $10 million in shared savings. According to an agreed-upon 80/20 split, $8 million in savings went to providers and $2 million went to payers.
Christine Vogeli, PhD, MGH Institute for Health Policy
Accompanying Document:
Multiple Chronic Conditions: Effective Strategies
Nature of the Problem
- A small fraction of patients are responsible for a large fraction of costs.
- Most high-cost patients have MCCs.
- Outcomes for these patients depend on quality of care.
- Outcomes also depend on patients’ self-management.
Implication: Improved delivery of care plus better self management can improve quality of care and reduce costs.
According to the Institute of Medicine, four factors have converged to drive the burgeoning costs of MCCs:
- An aging population.
- The increasing prevalence of chronic illness.
- Lifestyle and medication regimes and symptom recognition requirements that are so complicated that a large fraction of the population cannot implement them without help.
- A health care system that is not sufficiently pro-active or supportive in assisting MCC patients with self management.
There are three broad types of MCCs:
- Related; share a common basis. Metabolic syndrome, for example, is thought to reflect underlying genetic predispositions.
- Complicating. Heart failure, for example, can be a complication brought on by hypertension.
- Unrelated.
The implications for care and cost are based not only on the number of chronic conditions a person has, said Dr. Vogeli, but also on which combination of conditions increase costs the most. The health care system must offer the most effective disease and patient care management to obtain the best benefits:
- Current strategies for disease management typically focus on a single disease.
- Current care/case management does not adequately address psycho-social issues and access to care.
- It is fair to say that disease management can reduce costs, but probably not enough to do more than reach the break even point.
- Many interventions have never been shown to reduce costs.
- The evidence is mixed but generally positive about whether disease management improves patient outcomes.
- The health gains produced by disease management are more modest than the apparent opportunity (60 percent to 80 percent, not 60 percent to 95 percent).
- Many interventions have never been shown to improve outcomes.
Physicians may not be pleased with the flow to disease management companies of prestige, influence over patient care, and income. Optimal quality would seem to result from informed, activated patients and reorganization of physicians’ practices to better integrate patients’ disease management. A major limitation to disease management is identifying the patients who will benefit the most. Predicting who will be a high-cost case is not straightforward.
Partners HealthCare Connection Program (Boston)
- Developed in 1998 for heart failure patients using nurse practitioners. Now focuses on the most acutely ill patients in the system.
- Population focus is the 45,000 patients with chronic conditions who have had at least one hospitalization or emergency department (ED) visit. Out of these, 10,000 are enrolled in the Connection program.
- The program’s goal: assist high-risk Medicaid and uncompensated patients over the age of 18 in managing their chronic conditions and reduce unnecessary hospital use through telephone coaching.
Program Strategy: Care Management
- 24/7 health coaching over the phone by practice-based registered nurses.
- Health education and social work referrals based on needs evaluations.
- Allowing primary care providers to opt out of the program. Very few physicians opted out universally, although some opted out individual patients.
Key Results
- ED visits decreased 5.1 percent and outpatient visits dropped 7.9 percent. This result occurred when a member of the patient’s healthcare team literally walked down to the ED to talk about his/her condition.
- The program’s patient refusal rate was much lower than for payer-based programs (2 percent versus 30 percent).
- A pilot program is now in place to address medical/psychological conditions (heart failure combined with substance abuse, for example).
- The mean health care cost per patient dropped slightly (from $1,261 to $1,217).
Remaining challenges include:
- Patient identification and stratification – There is never enough data.
- Only 50 percent of high-risk patients discharged from Partners hospitals have a Partners primary care physician.
CMS/Massachusetts General Hospital Care Management Demonstration Project
- Three year demonstration project to determine if case management for high-cost beneficiaries can improve care quality and reduce overall costs under the Medicare fee-for-service program.
- Goal is to return 5 percent savings on net fees.
- Project includes comparison to a matched cohort.
- Twelve case managers help physicians manage 2,500 of their sickest hospital patients. The case managers are practice-based and received customized support for the three or four patients who required acute attention. The work flow includes end of life care.
- To date, enrollment is about 90 percent; after a six-month enrollment period, demonstration enrollment is one percent better than controls.
Particularly challenging for patients with MCCs are legacy systems that have created silos of information. These inhibit patient identification and create information holes. The question also remains about how well MCC patients can be served in a community hospital setting.
Dr. Vogeli concluded with a list of elements for effectively addressing MCC patients:
- An appropriate information system infrastructure.
- Medication management and reconciliation using strategies such as e-prescribing with dosing and interaction checks and clear, concise information for patients.
- Integrating practitioners who serve MCC patients with primary care practices.
- Medical homes with longitudinal medical management.
- Attention given to the critical transition from hospital to home.
- Attention given to dealing with social issues such as medical literacy and detachment from the system.
- Identification of key unrelated but exacerbating conditions, such as patients with co-occurring heart failure and substance abuse.
Discussion
Drs. Sarfaty, Lewis, and Vogeli and work group members participated in a question/answer and discussion session on the presentations:
Medication Guides. A Food and Drug Administration official noted that the agency has an effort underway to reword its medication guides so that they are more helpful when practitioners give explanations and directions to patients. Dr. Naylor noted that such tools are helpful to clinicians, who then transfer key messages to patients. As a general rule, patients do not want to have to go through a lot of information.
Health Literacy. Several topics arose concerning health literacy and patient compliance:
- Patients generally can absorb only three pieces of information at once.
- Doctors are often in a time crunch and have a difficult time translating compliance directions to patients. Nurses can help, but they can also be hampered by a time crunch.
- One helpful idea: write detailed, explicit directions, then back them up with an explanation. Challenge: when a patient cannot read well, teaching health literacy is even more of a problem.
- Another idea: visiting nurses can record patient questions on a tape recorder to ensure that they all get answered.
Patient Involvement. Compliance is more likely when a treatment plan begins with the patients’ needs and empowers individuals to make changes. One patient, for example, balked at taking 15 prescribed medications. His practitioners collaborated on which two or three were the most critical. Patients with MCCs are more challenging, but practitioners still need to start with a person’s own goals and engage that patient in his/her care. People need to think that they have some control. The relationship should be based on partnership, not paternalism.
Adding to Wish List for Obama Administration. Other guest speakers added to Dr. Anderson’s policy wish list for the incoming Administration:
- Advancing nurse education to include more information about patient-centered treatment.
- Advancing the medical home as a valid treatment option. Integrating the medical home concept into the Medicare and Medicaid systems.
- Continuing to focus on treatments that meet evidence-based criteria.
- Continuing to focus on the importance of the transition from hospital to home and how to refine this process. Integrating the transition process into the Medicare and Medicaid systems.
- Continuing to focus on interaction with patients for better self management.
- Increasing recognition of the importance of information system support.
- Creating multi-stakeholder demonstration programs that include the Medicare system. The agency is understaffed.
- Setting a 10-year goal to improve chronic care in the United States. This is a challenge on the scope of putting a man on moon. It requires a high-level sustained commitment, not a program here or there. Providing high quality care should be the driver, not limiting services. It would be essential that patients and their families become active in the care.
- Creating an independent government enterprise to drive fundamental—not incremental—delivery system change. This would require an independent review of evidence-based approaches to chronic illness, which would be the basis for system change. The system would match patient need and savings would be put into health promotion and disease prevention rather than returned to health care providers.
- Including MCC patients in medical trials with special focus on people with five or more chronic conditions.
- Paying practitioners for coordinating care.
Dr. Parekh thanked the speakers and work group members for taking leadership roles in searching for “answers to burning questions” about effectively addressing multiple chronic conditions. He expressed optimism that such leadership will continue through the Federal government’s upcoming transition.
Print 
Download Reader 
HHS Home|HHS News|About HHS
