INTRODUCTION

Geriatrics is the arm of medicine dedicated to treating the aging patient. With its focus on holistic approaches to patient care, caregiver support, and delivery system redesign, geriatrics has advanced our understanding of optimal care during transitions. Older adults experience many transitions across healthcare settings. Models of care for geriatric populations can provide insight into how to provide optimal transitional care for those with complex needs, regardless of age. Such populations include those with developmental disabilities, traumatic brain injuries, end-stage renal disease, or malignancies.

Lessons learned from the specific transitions mentioned in this article can be broadened to other transitions. This article will primarily discuss examples of care transitions from the hospital to the community (e.g., home, skilled nursing facility, home health care) because much of the research in the field has been done on these transitions, and the Centers for Medicare and Medicaid Services (CMS) are working towards improving care during hospital to community transitions. This article provides a framework for incorporating geriatrics principles into care transition activities by discussing the following elements: (1) identifying factors that make transitions more complex, (2) engaging care “receivers” and tailoring home care to meet patient needs, (3) building “recovery plans” into transitional care, (4) predicting and avoiding preventable readmissions, and (5) adopting a palliative approach, when appropriate, that optimizes patient and family goals of care. The article concludes with a discussion of practical aspects of designing, implementing, and evaluating care transitions programs for those with complex care needs, as well as implications for public policy.

THE CHALLENGE OF ENSURING OPTIMAL CARE DURING TRANSITIONS

Older adults’ care transition patterns are heterogeneous and complicated.1 Almost one-fifth of seniors are re-hospitalized within 30 days of hospital discharge, costing Medicare $17.4 billion yearly.2 Nationwide, approximately 22 % of older adults experience a transition annually. Half of those transitions involve going to and from a hospital setting from either a skilled nursing facility or home, but the other half often involve complicated trajectories across different settings. Those with dementia experience an increased number of transitions through acute and post-acute care settings.3

Care transitions have become a target for reform efforts; yet, our current understanding of how to optimize care transitions remains incomplete. Most improvement efforts have focused on the hospital discharge process and the immediate post-discharge period. These efforts often include the creation of a “navigator,” “transition coach,” or other role to provide continuity of care. Numerous interventions to improve care transitions have been developed. Yet, to date, no single intervention has reliably reduced 30-day readmission rates.4 Interventions are often disease-specific,5 – 8 require substantial financial investments in training allied healthcare professionals,6 , 9 , 10 or focus primarily on hospital-based discharge planning with mixed results.4 , 11 – 13 Additionally, interventions are often multi-faceted, and it is difficult to identify which elements drive improvements in patient outcomes.14 Besides older adults, there are also other populations with complex needs that may benefit from an approach to improve their care transitions.15

IDENTIFYING FACTORS THAT MAKE TRANSITIONS MORE COMPLEX

Several factors characterize complex transitions from the hospital to the community. Figure 1 demonstrates a conceptual model adapted from previous work16 that incorporates factors affecting implementation of the post-discharge plan of care. These factors encompass domains at the healthcare system level, provider level, and patient level. For example, lack of care integration within a healthcare system can lead to suboptimal communication across settings. At the provider level, unclear roles during care transitions can lead to suboptimal execution of care processes. Finally, factors inherent to the patient and their post-discharge environment can affect implementation of the care plan. Examples of these are listed in Fig. 2. Many patients with complex needs have cognitive impairment, functional impairments, and limited health literacy.17 When discharged from a hospital, these patients encounter frequent medication changes, complex multi-step instructions, and numerous details about follow-up appointments.18 Following hospitalization, they encounter high rates of further functional decline, cognitive changes, adverse drug reactions, pressure ulcers, bowel and bladder dysfunction, malnutrition, and dehydration.19 – 21

Figure 1
figure 1

Conceptual model of factors affecting care transitions.

Full size image
Figure 2
figure 2

Common phenomena affecting older adults and subsequent factors to consider when delivering care during transitions to any patient with complex needs.

Full size image

Specifically relating to the patient and their post-discharge environment, solutions can be developed based on the presence of the phenomena in Fig. 2. These phenomena are frequently encountered in the care of older adults, and are often also present in others with complex illness. Though not meant to be comprehensive, Fig. 2 depicts some of the most common challenges faced by geriatricians caring for older adults. Hearing and vision impairments are common among vulnerable populations and should be considered when tailoring transitions interventions. For example, an 88-year-old man with hearing impairment may not be able to understand well-intentioned care providers making a follow-up phone call after hospital discharge. Recommendations: Post-discharge phone calls may need to be enhanced with written instruction guides or education of caregivers. During hospitalization, vulnerable populations frequently experience cognitive dysfunction that would potentially impair them from understanding discharge instructions. With cognition shown to improve following hospitalization,22 programs could reinforce instructions when patients are better able to understand (e.g., in a follow-up phone call or visit, perhaps incorporating telehealth tools), or they must incorporate caregivers.

The patient’s support network is often pivotal to a successful transition. Most patients will need assistance from support people following a hospitalization. Support people include family members, friends, paid caregivers, and the outpatient clinicians to whom patients will turn for answers. Recommendations: Family members and friends should be a routine part of patient education, and timing of discharge discussions should accommodate the ability of family members to be present. This may mean discharge planning may need to begin several days prior to discharge and take place during early evening hours to accommodate family members who work.

ENGAGING CARE “RECEIVERS” AND TAILORING HOME CARE TO MEET PATIENT NEEDS

Populations with complex needs transition through many settings, including rehabilitation, skilled nursing, long-term care, assisted living, and home. As they move through these settings, their health and safety are frequently at risk due to fragmented, episodic, and uncoordinated care.2 , 16 Much of the work to improve care during transitions focuses disproportionately on improving the process of hospital discharge. Recommendations: Organizations may want to also focus on understanding and meeting the needs of the “receivers”: outpatient care providers, post-acute care facilities, home healthcare agencies, family members and paid caregivers. This can be done with a needs assessment asking receivers to articulate their challenges during care transitions. Healthcare providers describe a lack of education on how to execute optimal transitions and a lack of feedback on their performance delivering care during transitions.23 Various tools exist to assist with identifying the right post-discharge setting based on patients’ needs.24 Providers can be given feedback on a subset of transitions they help execute.

Skilled home health care (SHHC, e.g., nursing services, physical therapy) can help patients to transition back home safely.25 Of the 14 million annual hospitalizations of older adults, approximately one in six patients are discharged with a referral to home health, and this figure is increasing.26 The main goal of SHHC services for patients leaving the hospital is to provide services on a non-chronic basis to facilitate discharge, by providing skilled nursing or therapy services and teaching the patient or their caregiver to care for themselves safely in the home. Often SHHC services are not considered until the end of a hospitalization, severely limiting the time for services to be made, given the unique nature of SHHC. Recommendations: Well before the transition takes place, consider whether the patient would benefit from SHHC and notify the patient and agency early.

While all care transitions are potential points for lapses in continuity of care, care transitions to SHHC are particularly risky. For reasons that are incompletely understood, older adults who require SHHC after hospital discharge are among those at highest risk of hospital readmission and adverse events.27 – 30 SHHC services are unique in that medical professionals are present only intermittently, and health-related responsibilities are often shared with informal caregivers (e.g., family or friends).31 SHHC agencies perform a comprehensive assessment of all admissions to SHHC, but they currently lack the means by which to effectively track care transitions,28 in part because SHHC providers are not immediately aware of when patients are re-hospitalized.32 SHHC agencies are required to submit patient-level and agency-level data on process and outcome measures to CMS, but these measures do not provide information on satisfaction with care during transitions and subsequent hospital utilization. Recommendations: SHHC agencies must be actively engaged as “receivers” in care transitions improvement efforts. Additionally, regulators can encourage reporting on patient-centered measures related to care transitions, such that SHHC agencies can better tailor services in the home. One evidence-based model addresses the needs of SHHC populations by having hospital social workers partner with SHHC agencies to coordinate post-discharge care.33 Further research is needed in the area of improving care transitions for recipients of SHHC services.

Other receivers that must be engaged during care transitions are informal caregivers. They act as receivers, yet they are often unprepared and overwhelmed. Paid or formal caregivers are hired to support patients in their home following a hospitalization. With the goal to keep the patients out of a long-term care facility, paid caregivers assist with reminding patients to take their medications, ambulating, and are also required to manage agitation and mental confusion when patients return home from hospitalizations.34 However, these caregivers are largely unskilled and many lack adequate training to handle these responsibilities.35 Recommendations: Include activities that activate and support caregivers during and after care transitions.

Finally, clinical care providers in ambulatory care and skilled nursing facility settings must be engaged during care transitions. Recommendations: Clinical providers’ roles during care transitions remain poorly defined, and use of recently developed conceptual frameworks36 may help incorporate clinician perspectives into care transition improvement efforts. Standard formats for reporting discharge summaries, medication lists, cognitive status, and other important information should be discussed among all stakeholders in a community across the patient’s continuum of care. CMS innovations projects are underway to design strategies to improve communication between hospitals and post-acute care facilities.

To summarize, transitional care interventions need to ensure that the receivers, including skilled nursing facilities, home health, formal and informal caregivers, are educated at discharge and capable of handling potential scenarios that could lead to readmission.

BUILDING “RECOVERY PLANS” INTO TRANSITIONAL CARE

To be resilient, high-performing organizations need to have systems (1) to minimize problems (e.g., readmission, poor communication across care sites, failures to implement care plan), and (2) to recover if failures occur. This includes having the ability to apply evidence in real time and draw from knowledge of healthcare delivery processes to improve care.37 Because someone with complex needs may fail at any time (e.g., clinically deteriorate, experience lapses in care), making emergency contingency plans becomes extremely important. Most care transitions programs focus on implementing a structured intervention. What is often absent is an emergency plan or “recovery scenario,”—what to do when the patient begins to experience failures in the implementation of the plan of care.

Failures can be influenced by factors at several levels: patient level (not reporting the development of warning signs of clinical deterioration, not filling a prescription); health system level (failing to engage the patient and caregiver in discharge planning efforts; failing to notify outside care providers of changes to the plan of care); and regional level (lacking interchangeable or accessible electronic medical records). Table 1 lists several types of scenarios for which recovery plans could be designed. Though some care transitions efforts do have elements that recommend action plans if red flags are noted, these primarily act upon the patient level (e.g., clinical symptoms). Recommendations: To be comprehensive, recovery scenarios must be broadened to include the health system and region. Organizations may consider developing recovery scenarios that go beyond referring patients back to their primary care provider or others in the health system. At the regional level, for example, organizations may develop bi-directional communication across health systems that allows for messaging among providers and has automatic notification of care transitions. Table 1 provides examples of recovery plans at each level to possibly prevent unplanned healthcare utilization.

Table 1 Examples of Care Transition Scenarios and Possible Recovery Plans
Full size table

PREDICTING AND AVOIDING PREVENTABLE READMISSIONS

The ability to predict which patients will be readmitted remains poor.38 Likewise, the extent to which readmissions are preventable and can be reduced through hospital-based intervention programs is unclear. More attention needs to be directed at what causal risk factors exist for preventable hospitalizations, and how to utilize post-discharge networks and community resources to reduce readmission. While patients with previous readmissions seem an obvious target, it may be that these individuals are very sick with multiple comorbidities and complex care needs, and readmission is not preventable.39 The challenge remains on targeting interventions to those most likely to benefit.

A preventable hospitalization is a one that could have been avoided if modifiable system-level or patient-level factors had been identified and intervened upon. It is important to include factors beyond the patient level and include the other domains in Fig. 1. Among those with multiple comorbidities, targeting those with particular medical conditions (e.g., congestive heart failure) is unrealistic when patients have multiple chronic conditions, and social and functional needs. National data demonstrate that the odds of re-hospitalization are increased by living alone, having unmet functional need, lacking self-management skills, and having limited education.16 Recommendations: Social and functional factors can be routinely examined by a comprehensive evaluation (geriatric assessment), yet they are frequently overlooked in determining risk of readmission. Collecting such data elements in an accessible and accurate electronic medical record may be helpful in identifying preventable readmissions.

Additionally, system-level factors may be modifiable. Recommendations: In determining preventable readmissions, performing a root cause or failure mode effect analysis (a proactive risk assessment method used to ask groups to identify solutions to safety problems before they occur)40 can uncover factors leading to the readmission. Partnering with experts in human factors engineering can also be critical to the design of interventions to improve patient safety.41 – 44 Using the following criteria, one could ask if the hospitalization could have been avoided if: (1) care processes were changed, (2) education of the patient or proxy was appropriate, (3) social and functional issues were remedied, (4) timely and adequate health provider follow-up had occurred, or (5) recovery plans had been established and followed. Through these criteria, we emphasize that preventability is not limited to patient-level characteristics alone. Instead, applying concepts from geriatric medicine, a more encompassing view is needed to determine preventability of hospitalizations.

ADOPTING A PALLIATIVE APPROACH, WHEN APPROPRIATE, THAT OPTIMIZES PATIENT AND FAMILY GOALS OF CARE

Good palliative care can lead to good transitional care by matching care processes to desired goals and outcomes. A patient whose medications are streamlined and whose goals are clear is easier to transition across settings. Younger patients with complex illness also need help easing treatment burden and prioritizing goals of care. Patients and caregivers can utilize healthcare services more appropriately if their expectations are clear. Recommendations: Primary care and hospital providers should have conversations with all patients and surrogate decision makers regarding their goals and expectations for care. Conversations must go beyond determinations of “code status,” and in some cases may include offering concepts such as “do not hospitalize.” Establishing end-of-life care goals in long-term care settings could effectively reduce unnecessary transitions or readmissions to hospitals.45 The use of Provider Orders for Life-Sustaining Treatment (POLST) is an example of how forms can ensure goals of care are known throughout various care settings.

The five principles outlined in this paper thus far provide a framework for incorporating principles from geriatric medicine into care transitions improvement activities, regardless of age. In providing this framework, we emphasize the importance of moving beyond the patient level and incorporating system and regional factors into improvement efforts. Informed by a geriatrics perspective, we now discuss practical aspects of designing, implementing, and evaluating care transitions programs and implications for public policy.

THE NEED FOR CONTINUED HIGH-IMPACT, HIGH-QUALITY RESEARCH ON CARE TRANSITIONS

Progress in reducing readmission rates has been stagnant. Organizations partnering with researchers to improve care transitions may consider using innovative research and implementation methods, such as those adapted from the fields of qualitative research, quality improvement, implementation science, and human factors engineering. These methods allow for inclusion of a wide variety of stakeholders and provide an in-depth analysis of the complex phenomena involved in healthcare delivery. They have been incorporated successfully for use in other industries and in other aspects of healthcare delivery (e.g., surgical and intensive care).46 These methods may provide insights not previously noted as important in care transitions efforts.

The implementation of transitional care programs is more complex than introducing a new drug or diagnostic procedure. Before implementing transitional care programs, health systems should consider several aspects of measurement: needs assessment to inform program planning, process measurement, and outcome measurement. This should be done within a structure that facilitates continuous quality improvement.

PRACTICAL ASPECTS OF DESIGNING, IMPLEMENTING, AND EVALUATING CARE TRANSITIONS PROGRAMS FOR THOSE WITH COMPLEX CARE NEEDS

The first step is to understand local needs. A needs assessment for transitional care improvements should take into consideration patient needs and local factors such as community resources and practice patterns. Key steps of a needs assessment are: (1) identifying key stakeholders for each setting a patient experiences along the care continuum; (2) surveying of key stakeholders (patients and caregivers, post-acute care settings, regional health systems, and community networks) regarding gaps in the delivery of optimal care; and (3) reviewing local and regional data relevant to care transitions (e.g., utilization patterns, satisfaction).

The second step is identification of high-impact process and outcome measures. Process measurement and tracking provides critically important information about intervention fidelity and will facilitate dissemination. Without this information, it is difficult to know what components comprise an intervention, their relative importance, and their feasibility. Several groups, including the Agency for Healthcare Research and Quality, the National Transitions of Care Coalition and the Physician Consortium for Performance Improvement have identified process measures that link to successful outcomes.47 – 49 Process measures should reflect major intervention components, identify measurable outcomes at various stages of the intervention, and be easily integrated into workflows. Examples include timely transfer of information across settings, effective coordination, and timely delivery of care. Table 2 summarizes outcome metrics and includes those related to care quality, functional status, and quality of life. Adopting a robust approach to measuring processes and outcomes requires tracking patients across healthcare settings and payment systems. A community-wide data sharing agreement could facilitate collection of more comprehensive data.

Table 2 Examples of Outcome Metrics
Full size table

IMPLICATIONS FOR PUBLIC POLICY

System-level improvements to provide better transitional care will only occur if public policy guides systems to do so. Policy efforts can influence data collection, which in turn can inform the design and implementation of new care transition interventions. The approach outlined in this paper can provide policymakers with a wider lens through which to look at measures and outcomes aside from rehospitalizations.

Because hospital readmissions are common and costly, it is not surprising that most care transitions discussions have focused on the goal of reducing readmission rates.2 Moreover, the biggest national efforts to incentivize change—such as financial penalties for high readmission rates, and public reporting of readmission rates—function at the individual hospital level. The broader community context within which care transitions take place is an under-recognized and critical factor. Indeed, the single biggest factor associated with readmission rates in an analysis of national Medicare data was not patient comorbidity, discharge planning quality, or supply-level factors.50 Rather, communities with the highest admission rates had the highest readmission rates, suggesting that communities differ in the way health care is organized and in the way they respond to patients with urgent complaints. Hence, we recommend understanding the context of the admission rate before interpreting the readmission rate.

Community-level problems deserve community-level solutions. Emergence of accountable care organizations (ACOs), shared savings models, and bundled payments create opportunities to both improve measurement capability and integrate additional outcome measures. For example, ACOs could facilitate data sharing and continuity across administrative data sources. In addition, ACOs could reward outcomes such as timely care coordination and follow-up. Opportunities exist to introduce additional transitional care measures into publicly obtained and reported measures. The measures listed in Table 2 can serve as a guide.

CONCLUSION

Patients with complex needs can benefit from optimal transitional care, regardless of age. A 50-year-old man with uncontrolled diabetes and limited mobility has similar needs to an 80-year-old woman with cognitive impairment and limited transportation. Both would benefit from additional transitional care visits post-discharge to enhance self-management and caregiver support. Additionally, lessons learned from transitions from hospitals back to the community can be applied more broadly to other transitions. The field of geriatric medicine provides a framework for others to use when analyzing how to improve care transitions, especially for those with complex needs. Geriatricians’ holistic approach to patient management, caregiver support, and understanding of the continuum of care (and associated payment systems), facilitates the design, implementation, and evaluation of care delivery systems responsive to the needs of older adults and those with complex needs. Ultimately, the principles used to improve care transitions for the most complex older adults can lead the way to optimizing care for all populations.