BACK TO BASICS
Americans are increasingly concerned about the rising costs of health services and health insurance, the impact these costs will have on their access to care and the mounting evidence that patients do not always receive all of the care they need or care delivered without mistakes.
Legislators and regulators at all levels of government are exploring options for increasing quality, reliability and efficiency in our health care system. As health care purchasers look for strategies to both improve the quality of care patients receive and reduce costs, there is increasing interest in examining hospital readmissions.
No one knows better than hospital leaders and the clinicians who care for patients that some admissions could have been avoided if the patients had received the right care at the right time. Other readmissions may be unavoidable due to the natural progression of disease, the complex nature of patients' conditions or a variety of other factors. Not all readmissions are avoidable, and among those that are, many are the manifestation of problems in the broader health care system.
In general, preventing readmissions is a complex, systemwide problem that involves hospitals, physicians and other providers who manage patients' care, as well as patients and their families.
Readmissions can be divided into four distinct categories based on whether the readmission was planned or unplanned and whether the reason for the readmission was related or unrelated to the reason for the initial admission:
Planned, related readmission: An expected or scheduled readmission that is clinically related to the initial admission. For example: A woman is admitted with pre-term labor. The hospital stops the labor and discharges the woman, who is then readmitted the following week to deliver her baby.
Planned, unrelated readmission: An expected or scheduled readmission that is not clinically related to the initial admission. For example: A man is admitted for treatment of a heart attack and, during his admission, it is discovered that he has a tumor in his lung. When his heart function is stabilized, he is discharged and then later readmitted when he is strong enough to withstand the surgical procedure to remove the tumor.
Unplanned, unrelated readmission: An unexpected or unscheduled admission that is not related to the initial admission. For example: A man is admitted for an appendectomy. The hospital performs the procedure, and the man is discharged but then is readmitted the following week with a compound fracture from a car accident.
Unplanned, related readmission: An unexpected or unscheduled admission that is related to the initial admission. For example: A woman is admitted for an emergency appendectomy, discharged and then readmitted the following week for a surgical-site infection.
For the majority of planned readmissions, there are no actions hospitals can or should take to reduce their occurrence. These readmissions are beneficial for the patient and may be beneficial for the health care system as a whole, as in the example where a premature birth was prevented.
Likewise, there are no actions hospitals can take to reduce unplanned, unrelated readmissions because they are not predictable or preventable. However, in the case of some types of unplanned, related readmissions, hospitals may be able to reduce the occurrence of re-hospitalizations. Therefore, public policy efforts to reduce readmissions should focus exclusively on certain types of unplanned, related readmissions. Readmissions that fall into the other three categories should be excluded from consideration.
Even within the category of unplanned, related readmissions, there are many factors that make it inappropriate to hold hospitals accountable for all readmissions, including health care delivery system flaws that are beyond the control of the hospital yet lead to hospital readmissions. In looking at available administrative data, it is impossible to distinguish between patients who have been readmitted due to factors largely within the control of the hospital and those who have been readmitted for other reasons. For example, there are no data that indicate whether an unplanned, related readmission was part of the natural disease progression or due to a missed step in hospital care. The science available to inform policymakers is scant. In fact, most studies have focused on understanding and preventing readmissions for one subset of patients—those with heart failure. Very little is known about the causes for readmissions or potential ways to prevent readmissions among patients with other conditions.
Therefore, not all unplanned, related readmissions may be appropriate for inclusion in a public policy that holds hospitals accountable for reducing readmission rates. Some other examples of factors that may contribute to a readmission, but are beyond the control of the hospital, include:
Congress is considering major reforms to the health care delivery system, including how to prevent unnecessary readmissions. Options currently on the table would link hospital performance on readmissions to Medicare payments.
Preventing unnecessary hospital readmissions is a complex, systemwide goal that involves hospitals, physicians and other providers who manage patients' care, as well as patients and their families. Hospitals play an important role in preventing unplanned admissions that are related to the initial admission, but other sectors of the health care system also must do their part.
The following principles reflect the hospital field's view of how such policies should be shaped to be fair and effective, including ensuring that ongoing, thorough evaluation is conducted of any new policies.
The goal of any effort to reduce hospital readmissions should be to improve patient care. The use of payment or regulatory policies to create incentives to reduce hospital readmissions should focus on improving all hospitals' performance and providing the best care possible. Such policies should not be used as further cost-cutting measures for payers. A critical component of improving patient care involves sharing of best practices and collaborative learning among hospitals. Programs to reduce readmissions should not discourage hospitals from participating in shared-learning activities with others in the field.
Any policy that offers incentives to hospitals to reduce readmissions should begin by focusing on the unplanned, related readmissions. Any proposal that does not segregate planned readmissions from unplanned readmissions may prevent hospitals and doctors from providing appropriate services for patients in need of re-hospitalization. Among unplanned readmissions, there are some factors that are within the control of the hospital, and holding hospitals accountable for these factors through the payment system may be appropriate. For example, hospitals should ensure that timely information about a patient's care is communicated to post-acute providers when the patient is discharged. However, other factors that are outside the control of the hospital, as noted above, may affect patients' conditions.
Wherever possible, hospitals should implement evidence-based practices to reduce readmissions, and public policies and programs should support this implementation. While the knowledge base of evidence-based interventions to prevent readmissions is still small, there are some identified actions that hospitals can take. Making post-discharge follow-up phone calls, thoroughly completing medication reconciliation and promptly providing patients' primary care physicians with discharge information may help prevent readmissions for some patients. Regulatory and payment policies should not inhibit the implementation of these practices.
To effectively reduce unplanned, related readmissions, public policies should support the application of evidence-based practices across the broad health care system. Programs supporting rehabilitation, home visits, end-of-life decision-making and assistance to patients in managing their own care are resource intensive. Many are not adequately or explicitly supported by payment from insurers. Public policies, including payment policies, should better support the delivery of timely and appropriate care services so that readmissions resulting from inadequate access to needed services post-hospitalization or those that are not in keeping with patients' wishes for end-of-life care can be avoided.
Policies should be developed collaboratively with providers and other stakeholders. Payers should be encouraged to work with hospitals, physicians and other providers to develop policies to reduce readmissions. Together, they should develop shared performance objectives, payment methods and measures to assess performance.
Information on readmissions should be transparent to providers now to enable understanding of what the data mean and, subsequently, to the public for accountability. Any measures used to assess hospital performance on readmissions should be developed through an open, transparent and consensus-based process, such as is used by the National Quality Forum and the Hospital Quality Alliance. Information on hospital performance should be transparent to stakeholders and the public. However, the current data sources available do not provide useful and reliable insights into which are planned versus unplanned readmissions, and which of the unplanned readmissions might have been avoided. Data displayed for the purposes of helping consumers and others understand hospital quality, such as on the Hospital Compare Web site, portray important differences in performance that reflect differences in quality.
To enable investigation into the myriad issues surrounding readmission data, data should be made available to providers, researchers and public agencies, but should not be displayed on consumer Web sites for a period of at least two years. When these data are made available to the public, special attention will be needed to ensure that information is provided to help the public understand the data and how hospital readmission rates are related to other attributes of the hospital's structure and the characteristics of its patients.
Policies to reduce hospital readmissions should recognize differences among hospitals, their communities and the patients they serve. When incentives to reduce readmissions are introduced, care should be taken to ensure that all hospitals have an opportunity to participate and succeed without bias or disadvantage, particularly hospitals that serve sicker or more vulnerable patient populations. The limited availability of important post-acute and ambulatory health care services in some communities could affect a hospital's performance on readmission measures, but for the sake of the patients, it should not have an adverse effect on the hospital's payment.
While there are actions hospitals can take to reduce readmissions, truly decreasing repeat hospitalizations will take combined efforts by hospitals, physicians, other post-acute and ambulatory care providers, patients and their families to better manage ongoing health care needs. To implement better care coordination across the continuum of care, hospital, physician and other providers' incentives need to be better aligned. This likely will require broader system-level changes. Until these or similar changes introduce better coordination of care among providers, some potentially preventable hospital readmissions will remain a reality.
Caring for patients is often the shared responsibility of several providers, and hospitals should be partners with others in their communities to address readmissions. Hospitals and payers should work across the spectrum of providers and their communities to better coordinate patient care.
Greater investment in research is needed to identify and develop a better understanding of readmissions, discover effective practices to reduce readmissions, determine effective ways to implement those practices, and monitor the impact of readmissions policy changes. Because the examination of hospital readmission data is relatively new and, as of now, depends on the use of administrative data that do not contain many clinically important facts, ongoing efforts are needed to understand what can be learned from the data and what else would be needed to foster appropriately informed policy decisions. More work is needed to understand how readmission rates are affected by the characteristics of patients' conditions and other factors related to hospital stays, such as mortality rate and length of stay.
Similarly, the research on effective strategies for reducing readmissions is insufficient. Because most studies have focused on readmissions for heart failure patients, very little is known about causes or potential strategies to reduce readmissions among other patients. More research is needed to determine how readmissions are affected by community and patient factors and what actions can be taken to prevent readmissions among patients with various health conditions.
Financial incentives and other regulatory policies can be a powerful inducement to change behaviors but significant unintended consequences can occur when those changes are made before there is a sufficient base of knowledge to know what the right care approaches are and before the infrastructure is in place to support them. Monitoring the impact of any policies adopted and their intended and unintended consequences should be a part of any plan for change.
We know that hospitals have some ability to reduce unplanned readmissions if they understand the causes. Some effective strategies can be implemented solely by making changes in hospital processes, although others will include changes in the processes of other providers, patients and family members, and perhaps in regulation, accreditation standards or payment. In addition to our advocacy work, the AHA will make resources available to hospitals to help them understand the reasons why patients are readmitted and implement the suggested strategies to reduce readmissions. These resources will be made available through the AHA Quality Center.
Here are ideas and interventions that hospitals can adopt:
For more information about readmissions, visit the AHA Quality Center at www.hpoe.org.