Coordination (or Bust) for HF Readmission Reduction
Acute decompensated heart failure (HF) is a complex disease state with multiple comorbidities contributing to cardiac dysfunction and poor outcomes. Due to its high prevalence and complexity, HF is the most common reason for rehospitalizations in the U.S., leading to hefty expenses for private payors and the overall healthcare system. Therefore, providers and payors are seeking better management strategies for these patients to reduce preventable readmissions and mortality through improved, protocol-driven coordinated care throughput inpatient and outpatient settings.
These statistics are leading many to question how to better manage the HF patient population, including federal government agencies that have placed greater emphasis on transparency. Specifically, CMS and the Joint Commission have started publicly reporting patient satisfaction, mortality rates and readmission rates, as well as quality measures.
The government has been looking for additional ways to drive down readmissions. As early as 2012, CMS is putting into effect a new readmission policy, where hospitals with a higher rate of risk-adjusted, all-cause rehospitalizations will receive lower reimbursement. Thus, hospitals with low rates of rehospitalizations will receive greater reimbursements and there will be penalties for hospitals with overall higher readmission rates.
"Establishing a policy that would simply not reimburse for HF readmissions within the first 30 days would falsely convey that every rehospitalization for HF is preventable, which is wrong," explains Gregg C. Fonarow, MD, director of Ahmanson-University of California, Los Angeles (UCLA) Cardiomyopathy Center. "Even with the most intensive care and newest therapies, treating heart failure is still challenging, as complications will ensue and gaps can occur. Nevertheless, many HF hospitalizations are preventable and there are substantial opportunities to improve care and outcomes for HF."
Additional national drivers may alter how providers deal with HF readmissions, including the Recovery Audit Contractor programs, bundled payments, care management models, patient-centered medical homes and accountable care organizations (ACOs).
"Terminology is one of the biggest problems in this specialty because heart failure does not directly address the patient's clinical problem," says Mihai Gheorghiade, MD, director of experimental therapeutics at the Center for Cardiovascular Innovation at Northwestern University Feinberg School of Medicine in Chicago. "Therefore, physicians attempt to treat the failure, and do not assess the heart holistically." HF can result from a weak heart muscle itself, but also from abnormal valves or blockages of the coronary arteries that supply the heart muscle with oxygen, he adds, which need to be identified and treated uniquely based on each patient's needs.
For patients hospitalized for HF, Gheorghiade recommends identifying three components: the cause of the HF (i.e., weak heart muscle or coronary artery problems); the initiating mechanism (i.e., high blood pressure or irregular rhythm); and the amplifying mechanism (i.e., lung disease or diabetes) for improved results.
"Currently, providers are treating heart failure without fully understanding the exact cause or contributing causes for HF, which is why they may not attain successful outcomes," he says. "Hospitalization provides a unique opportunity for a more complete diagnostic workup with the presence of all the appropriate modalities and members of the care team with physicians, nurses and pharmacists."
The guidelines recommend that precipitating factors for HF exacerbation be identified, and potentially addressed, including the various comorbid conditions. "Although there are not specific performance measures to address the comorbidities, registries indicate that many hospitals and clinicians are not paying sufficient attention to comorbidities or precipitating causes that, if addressed, could help prevent recurrent hospitalizations," says Fonarow.
Addressing even noncardiac comorbidities in the hospital setting for patients admitted for HF, such as chronic obstructive pulmonary disease, severe renal dysfunction and diabetes, could possibly reduce readmissions. This strategy may be "particularly important" for the 40 to 50 percent of patients with HF who have relatively preserved systolic function, but do not have specific guideline-recommended therapies, wrote Gheorghiade and Peterson (JAMA 2011;305[23]2456-2457).
"The guidelines oftentimes are written as if patients have a single condition, but more typically patients have multiple conditions contributing to their heart failure," says Harlan M. Krumholz, MD, director of the Yale-New Haven Hospital's Center for Outcomes Research and Evaluation in New Haven, Conn. The complexity of these patients is particularly evident in the reasons for readmissions, as only about one in four patients, who are initially admitted to the hospital with HF, are readmitted for the same cause, Krumholz explains.
"We have to place greater emphasis on HF care, but we make a mistake if we do it to the exclusion of other important medical conditions that require attention," Krumholz says. "We also need to be proactive in ensuring a successful transition of care."
He recommends hospitals assess these types of questions: How complex is this individual's medical conditions? How many medications is he or she taking? How many times has the person been hospitalized and for what reasons? Has there been any potential harm incurred during the inpatient stay?
For performance measures, one of the biggest focuses has been on the patient's systolic dysfunction, the use of ACE inhibitors or ARBs, providing discharge instructions and performing medication reconciliation, according to Frederick A. Masoudi, MD, of the cardiology division at the University of Colorado Denver.
Masoudi says another difficulty with treating acute HF is the lack of evidence-based interventions and medications in the acute setting. The most recent example before with the FDA-approved nesiritide (Natrecor, Johnson & Johnson), which was found to have no benefit and lead to hypotension in the ASCEND-HF trial (N Engl J Med 2011;365:32-43).
While some evidence indicates that as the length of hospital stays have shortened, readmission rates have increased, the direct causal relationship is harder to prove.
Yet, the current payment model may be adversely affecting performance measures. "Due to the financial incentives to keep the length of hospital stay short in Medicare's DRG [diagnosis-related group] model, some have speculated that if patients are rushed out of the hospital, it has led to higher readmission rates," Fonarow points out.
Regardless of the payment model, providers need to use the inpatient time to change outcomes. "We need to make the length of time in the hospital as short as it can be, but not any shorter than it should be, particularly due to the risk of infections and weakness with bed rest in the inpatient setting," Krumholz says. "There needs to be a balance."
Masoudi concurs, suggesting the benefit of greater clinical stability and medicine titration achieved with a longer length of hospital stay is a "trade-off" with the risks involved with being in the hospital.
Also, Krumholz does not recommend returning to the days when HF patients were spending seven to 10 days in the hospital. "It isn't necessary anymore," he says. "However, we need to ensure proper support after discharge, which will occur with the coordination between inpatient and outpatient caregivers."
While there is still room for improvement, inpatient HF care might be getting better. Currently, four performance measures are tracked for HF, including: documentation of the assessment of LV systolic function; prescription of ACE inhibitors at hospital discharge; smoking cessation; and discharge planning measures.
"Since these processes have been publicly reported over the last few years, there have been substantial increases in adherence, resulting in less national variation of hospital performance," Masoudi says. "Also, the measures that are publicly reported do not capture the totality of care, which is better achieved through reporting outcomes measures." Despite better adherence to the performance measure, patient outcomes have remained the same (JAMA 2009;302(7):792-794).
One suggestion to decrease readmissions is to establish medical contact with the HF patient within one week after he or she has been discharged from the hospital, which Krumholz says is a "good idea to ensure that the patients are making a positive transition and adhering to treatment recommendations, but that contact doesn't have to be with a physician." However, the time needs to be used to effectively assess the patient.
In a previous, retrospective study of 10,599 patients in Canada, Lee et al found that early collaborative HF care by a cardiologist was associated with increased use of drug therapies, cardiovascular diagnostic tests and better outcomes—including a mortality improvement—compared with follow-up with a primary care physician alone (Circulation 2010;122:1806-1814).
To assess this method, Gheorghiade and colleagues are conducting the randomized IMPROVE HF Bridge trial. "This study will test the hypothesis that the one week post-discharge visit, where all potential targets for therapies are being identified and implemented by an HF specialist/nurse, will improve post-discharge outcomes," Gheorghiade says.
Masoudi notes that attending to the transition of care from the inpatient to the outpatient setting is "fundamental" to reducing HF readmissions and improving patient satisfaction. Also, the underlying goal behind establishing medical contact at one week would begin to keep the inpatient provider engaged in the patient's continuum of care after discharge.
Also, there is value to continuing follow-up beyond 30 days. Fonarow adds there is "no great advantage to patients and the healthcare system if a patient gets readmitted at day 31."
To aid this transition, multiple medical societies are seeking to better inform providers on transitions of care, such as the Hospital to Home initiative, cosponsored by the American College of Cardiology and the Institute for Healthcare Improvement, which publishes best practices on the transition from inpatient to outpatient status for individuals hospitalized with cardiovascular disease.
Also, the American Heart Association's sponsored Get With The Guidelines-Heart Failure program, with more than 600 participating centers, assists hospitals with monitoring and improving HF care and outcomes during various phases of care.
"Since hospitals are only reimbursed when the patient is admitted, they are not driven to follow patient care outside of their walls," Krumholz says. "To improve outcomes and decrease costs, coordinated care would allow for communities of caregivers to become involved with population health, providing an incentive to work together, which is lacking in the current fee-for-service model."
While disease management programs have been effective in reducing mortality and readmission rates in trials, Masoudi notes that they haven't caught on nationally because the incentives are not properly aligned. "Currently, hospitals are not incentivized nor penalized if their patients do not have strong outcomes," he adds. "These new models are very complicated. Although I can only speculate, changing readmission incentives could potentially result in better clinical outcomes and reduced expenditures, especially with chronic conditions like HF."
The trends in transitions in care and newer payment care models are indicative of the growing understanding that care delivery is a "team sport. It's not a single physician providing care to a single patient during a single admission," Masoudi adds.
With the movement toward ACO models, in which hospitals and outpatient care providers will work together and are potentially incentivized to provide higher quality coordinated care with better clinical outcomes, may result in better outcomes for these patients. However, there are some concerns about unintended consequences if hospitals are rewarded for not readmitting patients with no link to long-term quality of care or even 30-day mortality. For instance, Fonarow exemplifies a patient whose HF decompensation has become severe enough that he or she needs to be rehospitalized, but the provider is being financially incentivized to reject that patient, which could expose him or her to excess risk.
Despite some trepidations, Krumholz says that establishing a model of care that emphasizes coordination could particularly benefit HF patients because then you wouldn't just examine hospitalizations, but patients' experiences and episodes of care, which stretches across sites. Rather than just looking at inpatient and outpatient settings, this would assess how the patient is progressing.
The "most successful models for managing HF" includes a multidisciplinary team approach, Fonarow says, in which HF specialists work with the patients' primary care physicians, along with advanced practice nurses, social workers and other specialists to help manage a patient's comorbid conditions. Under this type of "intensive model," Fonarow et al showed that 85 percent of HF patients went without hospitalization over a six-month period (J Am Coll Cardiol 1997; 30:725-732). Also, the estimated savings in hospital readmission costs after subtracting the initial hospital costs for management was $9,800 per patient.
More recently, Phillips et al concluded that if applied on a nationwide basis, multidisciplinary disease management strategies for HF have the potential to prevent 84,000 readmissions, with an estimated reduction in Medicare payments of $424 million annually (JAMA 2004;291:1358-1367).
Krumholz suggests that pay-for-performance initiatives should be geared to a particular benchmark performance of readmission rates without micro-managing how the providers achieve those goals. "As with reimbursing for seven-day follow-up appointment for post-discharge HF patients, we need to ensure that the appointment is productive and didn't just occur," he adds. "This is why it's hard to pay for increments of care, as opposed to outcomes. "But establishing these clinical benchmarks are difficult because the patients are very sick in a varied fashion."
"Unless there is a team-based approach to caring for the patient across the continuum, HF will continue to be associated with substantial morbidity, mortality and expenditures," Fonarow concludes.
The problem
More than one million people are admitted to the inpatient setting for HF annually, according to Centers for Disease Control and Prevention 2006 statistics, and HF is the leading cause of hospitalization for people 65 years and older. Approximately 25 percent of Medicare patients with HF are readmitted within 30 days (N Engl J Med 2009;360[14]:1418-1428), and the rates of hospital readmission within six months range from 25 to 50 percent (J Am Coll Cardiol 2001;38:2101-2113). Despite efforts on the part of healthcare facilities to reduce rehospitalizations, unplanned readmissions cost Medicare $17.4 billion a year (N Engl J Med 2009;360[14]:1418-1428). As problematic, the 60- to 90-day post-discharge mortality rate remains as high as 15 percent (JAMA 2006;296[18]2217-2226).These statistics are leading many to question how to better manage the HF patient population, including federal government agencies that have placed greater emphasis on transparency. Specifically, CMS and the Joint Commission have started publicly reporting patient satisfaction, mortality rates and readmission rates, as well as quality measures.
The government has been looking for additional ways to drive down readmissions. As early as 2012, CMS is putting into effect a new readmission policy, where hospitals with a higher rate of risk-adjusted, all-cause rehospitalizations will receive lower reimbursement. Thus, hospitals with low rates of rehospitalizations will receive greater reimbursements and there will be penalties for hospitals with overall higher readmission rates.
"Establishing a policy that would simply not reimburse for HF readmissions within the first 30 days would falsely convey that every rehospitalization for HF is preventable, which is wrong," explains Gregg C. Fonarow, MD, director of Ahmanson-University of California, Los Angeles (UCLA) Cardiomyopathy Center. "Even with the most intensive care and newest therapies, treating heart failure is still challenging, as complications will ensue and gaps can occur. Nevertheless, many HF hospitalizations are preventable and there are substantial opportunities to improve care and outcomes for HF."
Additional national drivers may alter how providers deal with HF readmissions, including the Recovery Audit Contractor programs, bundled payments, care management models, patient-centered medical homes and accountable care organizations (ACOs).
Treating the whole patient
Acute decompensated HF is "a manifestation of diverse and complex cardiac and noncardiac conditions," leading Gheorghiade and Peterson to classify the disease as "acute HF syndromes," as opposed to a single syndrome (JAMA 2011;305[23]2456-2457).| Observed Hospital-specific 30-Day All-Cause Readmission Rates for Fee-for-service Medicare Beneficiary Discharges After HF Hospitalization | ||||
| Observed Readmission Rate | Overall | 2004 | 2005 | 2006 |
| Mean | 23.1% | 23% | 23.3% | 22.9% |
| Median (50th percentile) | 23.1% | 23.1% | 23.3% | 22.9% |
| 25th percentile | 18.6% | 18.7% | 18.8% | 18.2% |
| 75th percentile | 27.5% | 27.3% | 27.8% | 27.3% |
| Source: Circulation: Heart Failure 2010; 3:97-103 |
For patients hospitalized for HF, Gheorghiade recommends identifying three components: the cause of the HF (i.e., weak heart muscle or coronary artery problems); the initiating mechanism (i.e., high blood pressure or irregular rhythm); and the amplifying mechanism (i.e., lung disease or diabetes) for improved results.
"Currently, providers are treating heart failure without fully understanding the exact cause or contributing causes for HF, which is why they may not attain successful outcomes," he says. "Hospitalization provides a unique opportunity for a more complete diagnostic workup with the presence of all the appropriate modalities and members of the care team with physicians, nurses and pharmacists."
The guidelines recommend that precipitating factors for HF exacerbation be identified, and potentially addressed, including the various comorbid conditions. "Although there are not specific performance measures to address the comorbidities, registries indicate that many hospitals and clinicians are not paying sufficient attention to comorbidities or precipitating causes that, if addressed, could help prevent recurrent hospitalizations," says Fonarow.
Addressing even noncardiac comorbidities in the hospital setting for patients admitted for HF, such as chronic obstructive pulmonary disease, severe renal dysfunction and diabetes, could possibly reduce readmissions. This strategy may be "particularly important" for the 40 to 50 percent of patients with HF who have relatively preserved systolic function, but do not have specific guideline-recommended therapies, wrote Gheorghiade and Peterson (JAMA 2011;305[23]2456-2457).
"The guidelines oftentimes are written as if patients have a single condition, but more typically patients have multiple conditions contributing to their heart failure," says Harlan M. Krumholz, MD, director of the Yale-New Haven Hospital's Center for Outcomes Research and Evaluation in New Haven, Conn. The complexity of these patients is particularly evident in the reasons for readmissions, as only about one in four patients, who are initially admitted to the hospital with HF, are readmitted for the same cause, Krumholz explains.
"We have to place greater emphasis on HF care, but we make a mistake if we do it to the exclusion of other important medical conditions that require attention," Krumholz says. "We also need to be proactive in ensuring a successful transition of care."
He recommends hospitals assess these types of questions: How complex is this individual's medical conditions? How many medications is he or she taking? How many times has the person been hospitalized and for what reasons? Has there been any potential harm incurred during the inpatient stay?
For performance measures, one of the biggest focuses has been on the patient's systolic dysfunction, the use of ACE inhibitors or ARBs, providing discharge instructions and performing medication reconciliation, according to Frederick A. Masoudi, MD, of the cardiology division at the University of Colorado Denver.
Masoudi says another difficulty with treating acute HF is the lack of evidence-based interventions and medications in the acute setting. The most recent example before with the FDA-approved nesiritide (Natrecor, Johnson & Johnson), which was found to have no benefit and lead to hypotension in the ASCEND-HF trial (N Engl J Med 2011;365:32-43).
Longer initial stay = fewer readmissions?
 |
Yet, the current payment model may be adversely affecting performance measures. "Due to the financial incentives to keep the length of hospital stay short in Medicare's DRG [diagnosis-related group] model, some have speculated that if patients are rushed out of the hospital, it has led to higher readmission rates," Fonarow points out.
Regardless of the payment model, providers need to use the inpatient time to change outcomes. "We need to make the length of time in the hospital as short as it can be, but not any shorter than it should be, particularly due to the risk of infections and weakness with bed rest in the inpatient setting," Krumholz says. "There needs to be a balance."
Masoudi concurs, suggesting the benefit of greater clinical stability and medicine titration achieved with a longer length of hospital stay is a "trade-off" with the risks involved with being in the hospital.
Also, Krumholz does not recommend returning to the days when HF patients were spending seven to 10 days in the hospital. "It isn't necessary anymore," he says. "However, we need to ensure proper support after discharge, which will occur with the coordination between inpatient and outpatient caregivers."
While there is still room for improvement, inpatient HF care might be getting better. Currently, four performance measures are tracked for HF, including: documentation of the assessment of LV systolic function; prescription of ACE inhibitors at hospital discharge; smoking cessation; and discharge planning measures.
"Since these processes have been publicly reported over the last few years, there have been substantial increases in adherence, resulting in less national variation of hospital performance," Masoudi says. "Also, the measures that are publicly reported do not capture the totality of care, which is better achieved through reporting outcomes measures." Despite better adherence to the performance measure, patient outcomes have remained the same (JAMA 2009;302(7):792-794).
Post-discharge transition
While public reporting outcomes measures is one way to better track causes of HF readmission, attention is currently being targeted at the discharge and post-discharge processes in the interim.One suggestion to decrease readmissions is to establish medical contact with the HF patient within one week after he or she has been discharged from the hospital, which Krumholz says is a "good idea to ensure that the patients are making a positive transition and adhering to treatment recommendations, but that contact doesn't have to be with a physician." However, the time needs to be used to effectively assess the patient.
In a previous, retrospective study of 10,599 patients in Canada, Lee et al found that early collaborative HF care by a cardiologist was associated with increased use of drug therapies, cardiovascular diagnostic tests and better outcomes—including a mortality improvement—compared with follow-up with a primary care physician alone (Circulation 2010;122:1806-1814).
To assess this method, Gheorghiade and colleagues are conducting the randomized IMPROVE HF Bridge trial. "This study will test the hypothesis that the one week post-discharge visit, where all potential targets for therapies are being identified and implemented by an HF specialist/nurse, will improve post-discharge outcomes," Gheorghiade says.
Masoudi notes that attending to the transition of care from the inpatient to the outpatient setting is "fundamental" to reducing HF readmissions and improving patient satisfaction. Also, the underlying goal behind establishing medical contact at one week would begin to keep the inpatient provider engaged in the patient's continuum of care after discharge.
Also, there is value to continuing follow-up beyond 30 days. Fonarow adds there is "no great advantage to patients and the healthcare system if a patient gets readmitted at day 31."
To aid this transition, multiple medical societies are seeking to better inform providers on transitions of care, such as the Hospital to Home initiative, cosponsored by the American College of Cardiology and the Institute for Healthcare Improvement, which publishes best practices on the transition from inpatient to outpatient status for individuals hospitalized with cardiovascular disease.
Also, the American Heart Association's sponsored Get With The Guidelines-Heart Failure program, with more than 600 participating centers, assists hospitals with monitoring and improving HF care and outcomes during various phases of care.
New models of care to the rescue
The U.S. healthcare payment is being overhauled, and most specialists agree that HF care will benefit."Since hospitals are only reimbursed when the patient is admitted, they are not driven to follow patient care outside of their walls," Krumholz says. "To improve outcomes and decrease costs, coordinated care would allow for communities of caregivers to become involved with population health, providing an incentive to work together, which is lacking in the current fee-for-service model."
While disease management programs have been effective in reducing mortality and readmission rates in trials, Masoudi notes that they haven't caught on nationally because the incentives are not properly aligned. "Currently, hospitals are not incentivized nor penalized if their patients do not have strong outcomes," he adds. "These new models are very complicated. Although I can only speculate, changing readmission incentives could potentially result in better clinical outcomes and reduced expenditures, especially with chronic conditions like HF."
The trends in transitions in care and newer payment care models are indicative of the growing understanding that care delivery is a "team sport. It's not a single physician providing care to a single patient during a single admission," Masoudi adds.
With the movement toward ACO models, in which hospitals and outpatient care providers will work together and are potentially incentivized to provide higher quality coordinated care with better clinical outcomes, may result in better outcomes for these patients. However, there are some concerns about unintended consequences if hospitals are rewarded for not readmitting patients with no link to long-term quality of care or even 30-day mortality. For instance, Fonarow exemplifies a patient whose HF decompensation has become severe enough that he or she needs to be rehospitalized, but the provider is being financially incentivized to reject that patient, which could expose him or her to excess risk.
Despite some trepidations, Krumholz says that establishing a model of care that emphasizes coordination could particularly benefit HF patients because then you wouldn't just examine hospitalizations, but patients' experiences and episodes of care, which stretches across sites. Rather than just looking at inpatient and outpatient settings, this would assess how the patient is progressing.
The "most successful models for managing HF" includes a multidisciplinary team approach, Fonarow says, in which HF specialists work with the patients' primary care physicians, along with advanced practice nurses, social workers and other specialists to help manage a patient's comorbid conditions. Under this type of "intensive model," Fonarow et al showed that 85 percent of HF patients went without hospitalization over a six-month period (J Am Coll Cardiol 1997; 30:725-732). Also, the estimated savings in hospital readmission costs after subtracting the initial hospital costs for management was $9,800 per patient.
More recently, Phillips et al concluded that if applied on a nationwide basis, multidisciplinary disease management strategies for HF have the potential to prevent 84,000 readmissions, with an estimated reduction in Medicare payments of $424 million annually (JAMA 2004;291:1358-1367).
Krumholz suggests that pay-for-performance initiatives should be geared to a particular benchmark performance of readmission rates without micro-managing how the providers achieve those goals. "As with reimbursing for seven-day follow-up appointment for post-discharge HF patients, we need to ensure that the appointment is productive and didn't just occur," he adds. "This is why it's hard to pay for increments of care, as opposed to outcomes. "But establishing these clinical benchmarks are difficult because the patients are very sick in a varied fashion."
"Unless there is a team-based approach to caring for the patient across the continuum, HF will continue to be associated with substantial morbidity, mortality and expenditures," Fonarow concludes.