Engaging older adults in their transitional care: what more needs to be done?

Building upon a 2006 seminal report by Coulter and Ellins that described findings from 129 systematic reviews on the broad topic of ‘patient-focused interventions‘ [8], we conducted a state-of-the-art search for additional systematic reviews of research studies that focused on the concept of patient engagement [9]. The additional reviews, published from January 2006 through to February 2013, were not limited to intervention studies. Four electronic databases – the Cochrane Library, Medline, CINAHL and PsycINFO – were searched for combinations of the following terms: ‘patient engagement‘, ‘patient-centered care‘, ‘patient activation‘ and ‘patient preferences‘. From an initial search conducted by O‘Connor and replicated by Hirschman (both coauthors), 463 potential papers were identified (Figure 1). These coauthors independently reviewed the abstracts of all identified papers, applying two inclusion criteria previously established by all authors: the paper must be a report of a systematic review of primary research studies or secondary analyses of primary research studies on a topic related to patient engagement, and the paper must report studies of chronically ill adults (18 years of age and older) and include older adults. Of the 463 papers retrieved, 424 did not meet criterion one (systematic research review). An additional 24 did not meet criterion two (23 did not focus on adults and one did not focus on chronic illness). Reviewers disagreed on the status of two abstracts, independently read full article texts and achieved agreement on classification of papers following discussion; a total of 15 systematic reviews of adults were retained for this analysis.

In Table 1, we provide an overview of the primary foci (i.e., health literacy, clinical decision-making, self-care, patient safety, access, the care experience and service development), research designs, study populations and major findings for 129 systematic reviews included in Coulter and Ellins‘ work [8]. We also offer similar information for the 15 additional systematic reviews retained from the updated search and then indicate if knowledge generated from newer reviews adds to, reinforces or conflicts with major findings summarized by Coulter and Ellins [8].

Key concepts embedded in our definition of patient engagement involving older adults coping with multiple health transitions provide the framework for the analysis of retained papers. First, we examine the needs or preferences most relevant to this patient group and then explore what is known about promoting informed patients, shared decision-making between patients and clinicians, and mutual accountability for actions with clinicians.

Available evidence regarding the challenges experienced by older adults, especially during major health transitions, reinforces the need for patient engagement and suggests opportunities for CER to accelerate patients‘ involvement in care. Hospitalized older adults, for example, consistently express anxiety about the poor quality of communication with large numbers of clinicians, who are often strangers, exacerbated by at least some patients‘ perceptions that no one is in charge of their care [10,11]. At the same time, many elderly patients will not question or challenge clinicians‘ authority; some fear such responses will negatively affect their care [12]. Equally troubling are findings that patients have overall low expectations of positive interactions with clinicians [13]. However, a systematic review of older adults‘ and their family caregivers‘ experiences in acute care hospitals revealed that these common patient reactions to stressful health episodes can be mitigated when older adults believe they are being treated as unique human beings, have positive relationships with staff and are included in decision-making [12].

The importance of a care delivery approach that engenders older adults‘ trust and communicates respect is not just relevant to care experiences [14,15]; patients who perceive that care is focused on their unique needs and preferences are more likely to adhere to therapies [16]. In addition, engaged patients can help to reduce medical errors and prevent costly and avoidable outcomes such as infections [8,17,18]. Optimizing engagement requires that clinicians understand the outcomes patients describe as most relevant to them. For the majority of chronically ill older adults, improving or maintaining function, the activities of daily living that give these patients a sense of purpose and perceived higher quality of life, matters the most [19]. As they look to the future, older adults‘ greatest fear is loss of independence. However, findings from multiple studies suggest that anticipating the changing health needs of chronically ill older adults and incorporating future preferences into plans of care, often referred to as advance care planning, is not the norm [20].

Identifying preferences of older adults coping with advanced serious illness as well as those of their family caregivers can be exceptionally challenging near the end of life. A consensus is growing that that multiple conversations about patients‘ preferences at the end of life should be guided by trusted clinicians and take place under the most relaxed circumstances possible. Some argue that the focus of such communication should not be limited to decisions about the use of specific interventions, such as cardiopulmonary resuscitation (referred to as advance directives), but rather emphasize a broader discussion about the overall trajectory of illness and the process of dying. Content analyses of almost 4000 papers reinforce this view and contribute to our understanding of patients‘ goals of care toward the end of life [21]. Older adults‘ desires include: “being cured,” “living longer,” “improving or maintaining function, quality of life and independence,” “being comfortable,” “achieving life goals” and/or “providing support for family/caregiver.” These goals offer a framework for structuring challenging conversations about how treatment alternatives may or may not align with patients‘ values. In addition, a review of 38 studies revealed that the majority of older adults want family caregivers to act as surrogates during difficult healthcare transitions so that treatments are consistent with their preferences [22].

Assuring people are well informed, often referred to as health literacy, is complex and influenced by many factors including knowledge, language and cultural preferences [23]. Health literacy challenges are magnified among some older adults not only due to cognitive or physical deficits, but also because our healthcare system stresses processes and outcomes related to a given disease, when most of these patients are coping with multiple chronic conditions. However, informed patients are a prerequisite for effective engagement in care management. In 2004, an Institute of Medicine report revealed that patients with low literacy were less likely to adhere to treatments, experienced more medical errors, had poorer health status‘ and higher use of costly services such as hospital admissions [23].

Fortunately, selected interventions have been proven to mitigate low literacy and improve patient knowledge, behaviors and care experiences, but these have primarily focused on adults with a specific chronic condition (e.g., diabetes). Effective health literacy interventions typically include personalized or tailored information available in a variety of forms (e.g., printed and web-based) coupled with clinicians‘ direct advice and guidance [8,14,16,24–32]. For example, structured telephone support and monitoring contributed to reduced hospitalizations and mortality rates among heart failure patients [30]. Information provided as part of preoperative and postdischarge processes resulted in shorter lengths of stay and fewer follow-up visits for cancer patients [33]. In addition to coaching and educational materials, the use of patient-reported outcome measures has been identified as a valuable strategy to increase patients‘ health literacy [25]. Other methods of providing information such as educational programs without verbal reinforcement from clinicians are generally ineffective [8,16,28,34]. The jury is still out regarding which health literacy strategies work best for older adults with multiple chronic conditions.

Recent data suggest patients seek more than information; they want to collaborate with clinicians in care management decisions, a process known as shared decision-making [8]. Aids (ranging from printed sheets to interactive websites), tested in multiple randomized controlled trials, have been shown to increase patient participation in decisions related to specific conditions, improve their knowledge, self-confidence and adherence, and achieve more appropriate use of health services [29]. Cochrane has assembled a sample inventory of such aids [103]. The positive effects of decision aids are increased when combined with face-to-face counseling or targeted coaching [35]. In Coulter and Ellins‘ review, younger adult patients with multiple complex problems seem to require more direct and longer-term involvement with clinicians [8,24,104]. More recently, in 11 of 17 studies of socially disadvantaged adults (including older adults) with Type I or II diabetes, significant improvements in health outcomes were achieved when interventions were one-to-one, of higher intensity and longer duration [36]. Despite proven benefits, there is limited evidence that clinicians employ such shared decision-making strategies with chronically ill older adults [37]. Results from a few randomized controlled trials that have tested strategies to promote clinicians‘ adoption of shared decision-making have been inconclusive [38].

Accountability for self-management is shared between patients and clinicians. Informed patients, whose clinicians have actively engaged in decision-making, are accountable for adhering to mutually agreed upon plans of care. However, self-management is usually very challenging for older adults coping with complex chronic illnesses and social issues and they often require more than traditional clinical support. Patients with selected chronic conditions have demonstrated both enhanced care management capacity and improved adherence from peer support groups and culturally tailored interventions involving lay or community health workers [36]. A wide range of support tools (e.g., personal health records) have also contributed positively to behavioral and health outcomes for patients with selected health problems. However, the effectiveness of these strategies for especially vulnerable patients, including chronically ill older adults, is not known.

In summary, findings from Coulter and Ellins‘ work, augmented with findings from 15 additional systematic reviews, suggest multiple opportunities and strategies to enhance care management by emphasizing engagement of chronically ill older adults [8]. We have a robust body of knowledge about what matters most to these complex patients. In addition, the review offers evidence of improved patient experiences and outcomes among better informed and more involved patients, but the strategies that work best to advance health literacy, shared decision-making and accountability for self-management among chronically ill older adults are not as clear. We need to build upon what is known about proven approaches in each of these overlapping dimensions of patient engagement, a central focus of effective care management, by assessing their effects on complex, chronically ill patient populations. The following case offers an example of the application of CER designed to enhance care management of chronically ill older adults experiencing major transitions in health, with particular attention to patient and family caregiver engagement.

Effective care management of older adults coping with multiple chronic conditions is frequently complicated by several other health (e.g., functional deficits) and social (e.g., poverty) risk factors [2,11,38,39]. While ‘standard care‘ is increasingly recognized as less optimal than emerging models, key questions remain about which of the evolving approaches are better aligned with the goals of chronically ill older adults and improve their health outcomes while reducing total healthcare costs. The PCMH and the transitional care model (TCM) have each demonstrated improvements in the outcomes of older adults at different points on the chronic illness trajectory. However, reported studies of the PCMH [40–42] and, until recently, the TCM [43–45], have been limited to comparisons with standard care. Neither model has been observed to have an immediate and longer-term positive impact on patients‘ health status or costs, nor has either care management approach optimally engaged vulnerable patients to address needs throughout both acute and more stable transitions in health.

Despite promising findings, including improved health outcomes and reduced resource utilization for patients with chronic conditions such as diabetes [40,42], the PCMH approach has not been rigorously tested among older adults coping with multiple chronic conditions. In addition, system challenges may both delay and diminish the impact of this longitudinal care model. Frequent episodes of acute illness are the norm among chronically ill older adults and often result in hospitalizations [1,46,47]. Due to poor communication with clinicians in acute and postacute settings, primary care clinicians typically have little input into plans of care for their patients during critical health transitions to and from hospitals and postacute settings. By contrast, the TCM, a short-term care management approach, has consistently demonstrated improved health and quality-of-life outcomes among acutely ill older adults coping with multiple chronic conditions throughout and following episodes of acute illness [43–45]; however, over time, the model‘s effects diminish [48,49]. Optimal implementation of the TCM has also been challenged by system issues that limit interactions with vulnerable older adults after their acute illnesses have been resolved.

The PCMH‘s and TCM‘s performance patterns suggest an opportunity to enhance collaboration between primary and acute care clinicians (e.g., hospitalists) throughout patients‘ episodes of acute illness and to promote longer-term engagement with, and support for, a population coping with complex care needs and their family caregivers. The TCM protocol includes continuity of care provided by advanced practice nurses (APNs) across hospital, postacute skilled nursing facility (if referred) and community settings via 7-day per week access (via face-to-face visits, telephone and secure email) during acute illnesses and ongoing monitoring during more stable periods. Combining the PCMH plus TCM could enable APNs to partner with PCMH clinicians and more effectively and efficiently address the preferences and needs of this patient group and their family caregivers throughout major transitions in health. Similarly, this approach could better position patients to prevent future episodes of acute illness. The combined strategy may contribute to improved care experiences and enhanced shorter- and longer-term health and quality-of-life outcomes for high-risk older adults experiencing stressful transitions associated with multiple chronic conditions, while increasing achievement of their health goals and reducing healthcare costs. However, the relative benefits and costs of the proposed comparators (PCMH only vs PCMH plus TCM) among older adults coping with multiple chronic conditions and their family caregivers have yet to be demonstrated.

With the support of the Gordon and Betty Moore Foundation (CA, USA), the Rita and Alex Hillman Foundation (NY, USA) and the Jonas Center for Nursing Excellence (NY, USA), a multidisciplinary team based at the University of Pennsylvania (PA, USA) is collaborating with nine PCMHs in southeastern Pennsylvania (USA) to compare two evidence-based care management approaches, the PCMH only (n = 4 sites) to the PCMH combined with the TCM (n = 5 sites), to assess which model is more patient-centered and contributes to more effective and efficient shorter- and longer-term outcomes among recently hospitalized, community-based, chronically ill older adults. Guided by a multistakeholder advisory committee, the University of Pennsylvania team and clinical partners are comparing patient care experiences and selected health (e.g., functional and symptom status), quality of life, caregiver (e.g., burden) and resource utilization outcomes among recently hospitalized, community-dwelling, chronically ill older adults and their family caregivers who either receive care management only from a PCMH to a similar group of older adults who receive care from a PCMH combined with the TCM (PCMH plus TCM). This study has been approved by the University of Pennsylvania‘s Institutional Review Board as well as the Institutional Review Boards of participating sites.

Patients are enrolled in either the PCMH only or PCMH plus TCM group based on the site. Each enrolled older adult is 65 years of age or older, lives within 30 miles of a partnering PCMH, speaks English, has a minimum of four chronic conditions, and is considered ‘high risk‘ for poor outcomes based on a risk profile agreed upon by the University of Pennsylvania team and collaborating PCMH partners. Family caregivers of cognitively impaired patients are also enrolled. The intervention protocol for the combined PCMH plus TCM group is focused on identifying and implementing plans of care consistent with patients‘ and family caregivers‘ goals, promoting informed and shared decision-making, and fostering shared accountability for outcomes.

Preliminary findings on the PCMH plus TCM group are presented below; data collection of the comparison group (PCMH only) and longer term follow-up of both intervention groups are ongoing. To date, 72 patients and seven family caregivers have been referred to the PCMH plus TCM intervention group in this study; 56 patients and seven caregivers have enrolled (71% acceptance rate). Lack of interest in having services in their homes and/or in participating in research studies, coupled with lack of perceived need for services, are the most common reasons for refusals. Two enrolled patients did not return to their communities following hospital discharge. Thus, a total of 54 patients and seven caregivers have completed the PCMH plus TCM intervention. Enrolled patients are a mean age of 78 years; 70% are female, 19% are African–American and 41% live alone. Enrolled patients have an average of 12 active chronic conditions and are prescribed an average of 12 daily medications. A total of 20% (11 out of 54) of patients had been hospitalized once in the 6 months prior to study enrollment; 24% (13 out of 54) had been hospitalized two or more times in the prior 6 months and 30% (16 out of 54) had been hospitalized 30 days before enrollment. Preliminary findings from standardized surveys, coupled with individualized feedback from PCMH collaborators, suggest a high level of receptivity to the combined intervention by patients, family caregivers, primary care clinicians and APNs in the PCMH plus TCM intervention group, and improvements over time in health and quality of life within the combined group. Only one patient was hospitalized in the 30 days following enrollment.

APNs are maintaining detailed case studies of their experiences working with patients, family caregivers and PCMH clinicians in the PCMH plus TCM group. In addition to the patient‘s health history, case studies include a list of the goals established with each patient, an overview of the collaboratively (patient, family caregiver, primary care physician and APN) developed plan of care and summary of progress in achieving goals, including description of facilitators and barriers. Preliminary content analyses of 40 case studies completed to date suggest facilitators that support patient engagement (e.g., multiple contacts with patients to establish trust, patient goal setting, high level of patient motivation, and supportive family caregivers). Barriers for some patients (e.g., health literacy, cultural differences, behavioral health problems and severity of illness) suggest that optimal engagement of patients will be challenging. Qualitative analyses of all 54 case studies will be conducted to confirm or provide new insights into facilitators or barriers to effective care management.

The transformation from the current clinician- and system-centric care delivery model to one with an unwavering focus on patients and their family caregivers will require more than a universally accepted definition, a robust body of evidence and exemplar models [50–53]. Fundamental changes in the central roles and relationships between chronically ill older adults and clinicians in our healthcare system will be challenged by a complex set of factors. In their narrative review of the literature of such factors, Longtin and colleagues identified significant differences in the proportion of patients who do and do not want to participate in decision-making; the latter group prefers to rely on their clinicians [54]. These scholars concluded that many of the factors identified in this analysis (e.g., health literacy) as well as others (e.g., socioeconomic level) are substantial obstacles to patient engagement. A challenge for all stakeholders will be how to respond when patients choose not to become engaged. Among older adults, in particular, unique challenges may be posed by factors such cognitive status. For this and other groups of vulnerable older adults, engagement of family caregivers will be critically important.

While this analysis has focused on a re-envisioned role for patients, the engagement of the clinician in this new partnership is also of paramount importance. Evidence suggests some clinicians will resist such change for a number of reasons including: personal beliefs about the role of patients; perceived loss of control; time and other resources required to inform and share decisions with patients; and inadequate preparation for this new partnership role [54]. Contributing to the barriers for both patients and clinicians are a set of system issues, such as: lack of robust measures; limited communication of information about patients‘ needs and preferences; lack of adoption of shared decision-making aids and other tools to support clinicians‘ engagement with patients; and poor alignment of financial incentives [54].

Patient engagement is an emerging concept and, thus, our definition may have inadvertently missed relevant constructs. Since this language is relatively new, we relied on an evidence base that was initially broader in reach and then narrowed the search during the period when the terminology became more commonly used. Therefore, this state-of-the-art approach may have missed relevant research. In addition, our evidence was generated primarily from systematic reviews and, thus, relied on analyses of findings completed by other teams.

Patient engagement provides an important opportunity to improve care management of complex chronically ill older adults with the promise of substantial benefits to patients, family caregivers, clinicians and society. CER provides an immediate path to accelerate advances in knowledge essential to promote such engagement.

Patient engagement is important to maximize the health of all individuals and populations [55]. This paper focuses on the possibility that engaging the most complex group of chronically ill older adults and their family caregivers, especially during major health transitions, will improve their care management and the health system‘s overall performance. Informed by our analyses, the following recommendations relevant to CER are designed to advance the knowledge needed to accelerate engagement of this vulnerable patient group in the near future:

Papers of special note have been highlighted as: ▪ of interest ▪▪ of considerable interest