13: Transitions and Continuity of Care

Older adults with chronic illnesses are frequently in contact with the health care system. They periodically require hospitalization for acute exacerbations of chronic illness, falls, infections, and other problems. For these patients, hospitalization often marks the beginning of a journey through a series of disconnected settings and providers. Because of this lack of connection, the journey is not smooth. Mishaps along the way are sometimes insignificant, sometimes even undetected, but others are life altering. The discussion here focuses on how this journey became so complex, the risks of the journey, and the best practices and innovations aimed at minimizing these risks.

The term care transition refers the transfer of a patient’s care from one team of health care providers to another. Transitions usually occur when a patient physically moves between sites of care. Care transitions can be grouped into 3 broad categories. The first category, perhaps the most studied, includes community dwellers discharging from the hospital. For example, an older adult may develop a chronic condition for which he receives care from his primary care physician. Exacerbation of the chronic condition might prompt hospitalization, where hospital-based physicians, nurses, and therapists care for the patient. The patient may then move to a skilled nursing facility (SNF) for rehabilitation and/or nursing care and in that setting encounter an entirely new care team. When rehabilitation goals are achieved, the patient may return home, resume care with his primary care physician and may also receive home care from a new team to complete any remaining tasks in recovering from the original exacerbation. In this example, the patient underwent 3 separate transitions in care: primary care provider to hospital provider, hospital to SNF, SNF provider back to primary care provider and to a new home health care team. The second category includes nursing home residents transitioning to and from the hospital. Although many of the problems seen in these transitions are the same as those encountered for community dwellers, additional challenges exist in transitions for these frail, functionally, and often cognitively, impaired patients. Finally, the third category includes patients who are facing the end of life. They often experience multiple transitions in providers in the moves between home, emergency room, and hospital with disease progression. For these patients, again, transitions carry the same risks as those for the community dwellers with chronic diseases, but also special problems come up for those near the end of life. Transitional care broadly refers to time-limited care processes aimed at ensuring safe and minimally disruptive transfers of care between different sites and providers.

The frequency of care transitions may be in part a consequence of changes in the structure and financing of the health care system over the past 30 years. In 1983, faced with ever-increasing costs, Medicare adopted a prospective payment scheme, whereby hospitals were no longer paid individual fees for individual services provided. Instead, hospitals received predetermined diagnosis-based fees for entire hospitalizations. Thus, hospitals had a financial incentive to increase efficiency and shorten length of stay. Indeed, length of stay decreased with the new legislation, but patients were not only discharged “quicker,” they were also discharged “sicker.” The “sicker” patients were more likely to be transferred to SNFs for rehabilitation and continued skilled nursing care prior to discharge home. Concurrently, fast-paced advances in hospital medicine coupled with the push to improve efficiency prompted physicians to restrict their practice to single sites (ie, clinic or hospital). Fewer primary care physicians continued to follow their patients while hospitalized. This shift in practice patterns meant that patients would routinely transfer care from one provider to another when moving from one setting to another.

Finances may also subtly encourage transfers of nursing home residents to the hospital. Medicare certified nursing facilities often provide temporary skilled rehabilitation and nursing services to their residents returning from the hospital. Reimbursement for skilled services is higher than the rate paid, often by Medicaid, for room, board, and custodial care. Medicare will only pay nursing homes for skilled services if patients have a preceding qualifying hospital stay. Thus, transfers to the hospital can be financially beneficial to nursing homes. Furthermore, nursing home residents using their Medicare Skilled Nursing Facility benefit can rarely simultaneously access their Medicare Hospice benefit, thus increasing the possibility of potentially avoidable end-of-life transitions.

For patients at the end of life, new symptoms and eventual functional decline prompt many care transitions. Because there are few comprehensive home care programs for people with life-limiting illnesses still seeking disease-directed therapies, patients with new symptoms often need to access care through the emergency room. Although functional decline is expected late in the course of life-limiting illness, often there is not time or staffing to develop plans for increased care when needed.

Systems factors have contributed to the frequency of care transitions, and many factors inherent to the system increase the risk of negative outcomes during transitions. Transitions provide opportunities for new providers to reevaluate unresolved problems and refine care plans, but transitions are also fraught with risk. Older patients and those with multiple chronic illnesses—those who are most likely to experience multiple transitions—are particularly vulnerable to the risk of adverse events.

Studies describing the adverse events associated with transitions have focused on the transition of community dwelling adults from hospital to home or other facility and the transfer of nursing home residents from hospital to nursing home and back. A growing body of research describes transitions at the end of life. A prospective observational study of hospitalized adults showed that one in five discharged patients experienced an adverse event associated with discharge. The adverse events were most frequently related to medications but also included nosocomial infections, falls, and complications of procedures. Half of the adverse medication events were felt to preventable or at least “ameliorable.” In a retrospective analysis of Medicare claims data, nearly one-fifth of beneficiaries who were hospitalized were readmitted within 30 days. Ninety percent of these rehospitalizations were considered “unplanned”; that is, not for follow-up treatments or procedures. In a study of Medicare claims in nursing home resident decedents, nearly one-fifth had at least one “burdensome transition” in the last 90 days of life. “Burdensome transitions” were defined as hospitalization in the last 3 days of life, or multiple hospitalizations, or residing in different nursing home facilities in those last 90 days. In another study, geographic areas with greater rates of care transitions among nursing home residents also had greater rates of feeding tube placement in patients with severe cognitive impairment, a group that is unlikely to benefit from this invasive procedure. Transfers of nursing home residents have also been associated with drug regimen changes and adverse medication effects. Smith et al found that 50% of Medicare decedents visited emergency departments in the last month of life, 77% of these were admitted to the hospital and 68% of those admitted died there, despite the fact that several studies document that a majority of people would prefer to die at home. These findings suggest that patients’ goals of care can be lost in transitions. Moreover, transitions seem to promote overuse of health care services, thereby contributing to ever-increasing health care spending. The goal of transitions research has been to identify the factors that contribute to these adverse events, particularly those that lead to readmission, and to design interventions to decrease the rate of readmissions and, ultimately, to better integrate health care across settings.

A successful care transition is one in which providers have timely, complete information about the hospitalization and patients have the same information, as well as easy access to answers and support when problems arise. Coleman groups barriers to successful care transitions into 3 levels: systems, provider, and patient. Our health care system is composed of many autonomous health care facilities and networks of facilities. Communication and collaboration across sites and between networks is challenging for many reasons—from simple lack of readily available contact information for providers at different sites to laws protecting confidentiality. Information systems are often not shared between different systems, thereby slowing down the transfer of key data. Although the Health Insurance Portability and Accountability Act (HIPPA) has a provision allowing transfer of information between providers for the purpose of continuing care, many health care workers and staff are not familiar with that provision. Furthermore different institutions have different drug formularies based on contractual relationships with pharmaceutical companies, leading to medication substitutions with each transition. Another systems barrier has been the lack of incentives to ensure the quality of care transitions. However, Medicare’s contracted Quality Improvement Organizations were tasked improving transitions in selected areas in with the 9th Scope of Work in 2008 and 10th Scope of Work in 2011; additionally, the Affordable Care Act proposes a pilot program using bundled payments for entire episodes of care across sites (ie, hospital and postacute SNF care), as an incentive to reduce readmissions.

Provider-level barriers stem from communication difficulties. The increasing prevalence of site-specific providers generates inpatient–outpatient physician discontinuity. Communication between inpatient and outpatient providers is most frequently accomplished via discharge summaries, but these often do not arrive in the receiving physician’s office in a timely manner, if at all. Discharge summaries have also been shown to omit key information, such as which test results are pending and which follow-up appointments have been scheduled. Other forms of direct communication between inpatient and outpatient providers, such as telephone calls or e-mails, are infrequent.

Patient-level barriers include limitations in health literacy and self-efficacy. Patients may not know the details of their health history, or even the names and dosages of their medications, leading to the possibility of inaccurate medication prescription in the hospital. Additionally, with shorter inpatient stays, patients are generally still recovering and perhaps facing new diagnoses at the time of discharge. Thus, they are likely to have new self-care responsibilities, including monitoring of symptoms and signs, taking new medications, and keeping follow-up appointments either on their own or with the help of family or friends. Providers frequently overestimate patients and families’ abilities (physically, socially, and cognitively) to manage their medical conditions. All of these problems can be traced back to the limitations in provider–patient communication. Discordance between physician explanations and patient understanding has been well documented. Communication can be even less effective when patients have low functional literacy or primarily speak a language other than English.

Excellent transitional care is associated with reduced rates of readmission to the hospital, cost savings and greater patient satisfaction. Readmissions can be avoided if inpatient and outpatient providers communicate effectively, medications are carefully reconciled at multiple key time points, and patients and families are educated about monitoring and care needs after discharge or transfer. The Joint Commission Guidelines for discharge summaries recommend that the following information be included: diagnoses, abnormal physical findings, important test results, discharge medications including reasons for changes, follow-up appointments, education provided to patient and family, and tasks to be completed (Table 13–1). For older adults, documentation of cognitive and functional status, skin condition including description of any pressure ulcers, nutritional status, goals of care, and surrogate decision makers are also important. Detailed medication reconciliation, with the assistance of a clinical pharmacist for patients with complex regimens, is essential to reducing adverse drug events. For patients with cognitive or functional disabilities or psychosocial challenges, a multidisciplinary team including social workers, nurse discharge planners, physical and occupational therapists is essential. Finally, using clear language and a trained interpreter if needed, discharging teams should counsel patients and families about medication changes, outpatient appointments, self-care and “red flags” signaling a call to the doctor or return to the hospital. For the patient transferring to an intermediate site of care, counseling should include a description of what to expect in the next site of care. If there have been major changes in goals of care and treatment limitations, a Physician Order for Life-Sustaining Treatment (POLST) form increases the likelihood that patients will have orders consistent with their wishes in the next setting. Finally, because written discharge summaries do not capture every detail, direct discussion between transferring and accepting providers can be helpful in complicated situations.

The transition from one setting to another is often a good time to review overall goals of care. This type of discussion could include the patient and family’s understanding of the hospitalization and what they are expecting in the next setting. Eliciting specific hopes of future therapies can help discharging providers set realistic goals with the patient and initiate discussion of alternate plans in case those goals are not met.

Coordination between different disciplines to accomplish all these tasks is key for effective transitional care. Naylor and colleagues designed an intervention utilizing advanced practice nurses for hospitalized elders at risk of poor postdischarge outcomes. The nurses provided personalized care coordination during the hospitalization, as well as follow-up visits and calls after discharge. They followed patients beginning at the time of hospital admission through 3 months after discharge. In a randomized trial, intervention patients were significantly less likely to be readmitted to the hospital within 24 weeks of the original discharge. Moreover, the intervention group incurred roughly half the health care costs of the control group in those 24 weeks.

Beyond care coordination, qualitative studies suggest that patient activation and self-management are also important to reducing adverse events associated with care transitions. The Care Transitions Intervention utilized a “transition coach,” an advance practice nurse who worked with hospitalized older adults during the admission and for 4 weeks after discharge. The aim of the intervention was to empower patients to be more involved in their own self-management. Thus, rather than act as another provider, the transition coaches helped patients and caregivers take more active roles in their care. A second component of the intervention was a personal health record, carried by patients between settings, containing key information including diagnoses, medications, allergies, and advance directives. The intervention was studied in 2 randomized, controlled trials, 1 with patients enrolled in Medicare managed care plans, and 1 with patients using traditional fee-for-service Medicare. In both studies, intervention patients had lower rates of hospital readmission at 30, 90, and 180 days.

Other studies have focused on improving transitions for nursing home residents. Many hospitalizations for this population are considered avoidable, given a high prevalence of preventable admission diagnoses and significant geographic variation in readmission rates. Thus, many interventions designed to improve transitions for nursing home residents have focused simply on preventing hospitalizations. Berkowitz and colleagues studied the use of an admission template and automatic palliative care consultation for patients living in a SNF who had 3 or more hospitalizations in the 6 months preceding admission to the facility. A second component of the study was a regular meeting of the interdisciplinary team to examine root causes for rehospitalization events. Disposition was studied in the year prior to the intervention and after 1 year of implementation. Rehospitalization rates declined, home discharges increased, discharges to long-term care decreased and death on the unit increased (all deaths were felt to be expected). The Interventions to Reduce Acute Care Transfers (INTERACT) is a set of tools designed to help nursing home staff detect, evaluate and communicate early changes in resident status. A quality improvement project implementing these tools in multiple facilities showed a significant decline in rehospitalization, as compared with facilities not using the tools.

Beyond making changes to care processes as described above, novel methods of information transfer and reimbursement for services have been proposed. These include improvement in accessibility of patient information across multiple sites of care. Options include universally accessible computer-based records and portable records that could be taken by patients themselves from site to site. Additionally, financial incentives for improving transitional care are being introduced, such as the development of Accountable Care Organizations that will receive bundled payments for entire episodes of care across several settings. Also part of the Affordable Care Act, the Community Based Care Transitions program will fund hospitals and community organizations to use proven interventions to integrate care for Medicare beneficiaries who are at high risk of hospital readmissions.

Older adults with complex medical conditions are at risk of adverse events as they traverse the various sites within the health care system. A number of interventions have been shown to improve post-discharge outcomes and reduce readmissions. Comprehensive changes at the patient, provider, institution and overall system levels are needed to improve transitional care as the population ages.

Interventions to Reduce Acute Care Transfers (Interact II). http://interact2.net

Society of Hospital Medicine. Project BOOST (Better Outcomes for Older adults through Safe Transitions). http://www.hospitalmedicine.org/ResourceRoomRedesign/RR_CareTransitions/CT_Home.cfm

The Care Transitions Project (Coleman, et al). http://www.caretransitions.org/

Transitional Care Model (Naylor, et al). http://www.transitionalcare.info