Research on interventions for persons living with dementia and their care partners and caregivers has expanded greatly in the past three decades, with a better understanding of guiding principles and core components of dementia care, services, and supports (see Chapter 2). However, the findings of low-strength evidence in the Agency for Healthcare Research and Quality (AHRQ) systematic review highlight the need for a more rigorous and robust evidence base to inform decision making in this complex system. The evidence is lacking as a result of methodological challenges (see Chapter 5), inattention to all the key factors that go into implementation decision making (see Chapter 4), and the need for more ways to study this topic in the face of complexity (see Chapters 3 and 5). There are also key gaps in the evidence base for certain types of interventions, including at the community, policy, and societal levels (see Chapter 5).
In this chapter, the committee lays out a blueprint for future research on dementia care interventions by outlining the methodological improvements needed across the research enterprise to strengthen the evidence base at multiple levels, including prioritizing inclusive research and incorporating throughout the study process the priorities of persons living with dementia and their care partners and caregivers. Recognizing the complexity of dementia care interventions and the systems in which they operate, this chapter highlights the importance of partnerships to delivering care and implementing interventions, as well as the integration of multiple methodological approaches to provide a richer evidence base that accounts for this complexity. In addition, key factors for assessing the real-world effectiveness of dementia care interventions are presented at the end of the
chapter, along with strategies for improving the assessment of individual-level interventions and expanding the focus on community-, policy-, and societal-level interventions.
The AHRQ systematic review identified two types of interventions—collaborative care models and a multicomponent intervention for informal caregivers (REACH [Resources for Enhancing Alzheimer’s Caregiver Health] II and its adaptations)—as supported by low-strength evidence of effectiveness. The evidence for all other interventions examined in the AHRQ systematic review was insufficient to support conclusions regarding effectiveness. Although the inherent complexity of the area of study (e.g., the multifaceted and progressive nature of dementia and heterogeneity across populations and settings) poses its own challenges to the generation of a robust evidence base for dementia care interventions, the AHRQ review also noted challenges related to methodological limitations, as discussed in Chapter 5 and described in more detail below (Butler et al., 2020). These limitations include small sample sizes and limited-duration studies; heterogeneity of outcome measures and interventions that precludes aggregation of study results; the lack of measures related to intended and unintended benefits and harms important to persons living with dementia, care partners, and caregivers; overreliance on randomized studies with insufficient integration of other relevant evidence; insufficient reporting in the field in terms of fidelity, detailed methodology (e.g., nature and dose of intervention components), and null findings or negative results; and a lack of focus on community-, policy- and societal-level interventions. Despite encompassing a large number of studies, the AHRQ review also cast a relatively narrow net in excluding certain larger pilot studies and observational analyses of programs and policies that may yield benefit. The AHRQ review was limited in its ability to draw conclusions for specific interventions because of the high level of uncertainty of the evidence (Butler et al., 2020). This section of the chapter presents the committee’s recommendations for improving the evidence base across the field, with a focus on the addressable methodological limitations highlighted in Chapter 5.
Ensure a Balanced Portfolio of Short- and Longer-Term Studies with Sufficient Sample Sizes
Much of the dementia care research conducted to date has consisted of studies with small sample sizes and limited duration, making it challenging to detect significant effects. This limitation of the evidence base was highlighted in the AHRQ systematic review, in which the vast majority of
studies had small samples (defined in the systematic review as fewer than 10 participants per study arm) or were pilots (generally considered as a small-scale test of the feasibility of delivering the intervention, although as discussed in Chapter 5, this category as implemented in the AHRQ systematic review encompassed a heterogeneous set of studies). These studies were included in the review’s evidence map describing the overall landscape of evidence but were excluded from the analytic set, and the systematic review found insufficient evidence to support conclusions about effectiveness (Butler et al., 2020). Studies with small sample sizes have an important role in moving the field forward by enabling early testing and refinement of newly created interventions, as well as early feasibility and pilot testing, as described in Stage I of the National Institutes of Health (NIH) Stage Model for Behavioral Intervention Development (discussed in Chapter 4) (NIA, 2018). However, it is important that interventions ultimately advance through later stages of the NIH model to include efficacy testing, effectiveness research in diverse populations and settings, and dissemination and implementation. Given the heterogeneity of persons living with dementia, care partners, and caregivers, studies with larger sample sizes are needed to provide adequate power for detecting meaningful and statistically significant effects across subgroups (HHS, 2018).
In addition to sample size, the duration of a study is important to understanding the intervention effects on outcomes for persons living with dementia and their care partners and caregivers, and whether the intervention has been implemented successfully with sustainability and long-term fidelity. Recognizing the long trajectories of dementia and how the disease course unfolds over a decade or more, it may be appropriate to extend the observation period for some studies based on the study population (e.g., age and stage of disease among study participants) and expected outcomes. Longitudinal studies in community populations that are representative of persons living with dementia, care partners, and caregivers—as opposed to convenience samples often drawn from specialized clinics—may shed light on the progressive nature of the disease and how care, support, and service needs evolve. It will be important to understand whether an intervention implemented over the long term can adapt effectively to the changing needs of persons living with dementia, care partners, and caregivers, including the settings in which care is delivered (e.g., individual home, residential care facility, nursing home). It is also possible that certain interventions are effective at one phase of the disease and not others. Community-, policy-, and societal-level interventions are well aligned with study designs that extend over longer observation periods and larger populations in real-world settings (discussed further later in this chapter). In some cases, it may be desirable for persons living with dementia, care partners, and caregivers to adhere to an intervention indefinitely (e.g., interventions, such as physical activity, aimed
at establishing healthy lifestyles). In such instances, it would be important to assess long-term outcomes and any mediating effects that might contribute to the observed effect (Emsley et al., 2009; Richiardi et al., 2013).
Longer-duration studies may be feasible with alternative funding approaches that would enable investigators to design trials that could be renewed after the typical initial National Institute on Aging (NIA) 3- to 5-year funding term. Such studies would also be facilitated by NIA’s use of existing mechanisms and criteria that guide study sections in assessing initial proposals and renewal applications, including consideration of ways to overcome some of the research and methodological barriers specific to dementia care research that make the traditional pathways inhospitable to this type of research, or that are too focused on clinical trials. NIA might also consider providing a road map for researchers offering funding support for each stage of the NIH Stage Model, tailored to the unique challenges inherent in dementia care research. Particularly for longer-duration studies, methods need to be in place for addressing issues related to attrition and how best to retain study participants. While the inclusion criteria of the AHRQ systematic review did not impose a minimum duration or follow-up period, attrition bias was assessed differently if the duration of a study was less or more than 12 weeks (Butler et al., 2020). The review authors note that they allowed attrition to reach relatively high levels before assigning high risk of bias, an approach that is particularly important given the high likelihood of attrition due to death in longer-term studies in an elderly population. In addition, methods are needed for addressing issues related to blinding and preventing drift in how an intervention is implemented across the study and control groups.
To generate the robust evidence needed to move the field forward most efficiently, it will be important to have a balanced portfolio for dementia care research comprising short- and longer-term studies, all with sufficient sample sizes, including studies along the NIH Stage Model. As in other therapeutic areas (e.g., cancer, cardiovascular disease), development of this portfolio may include a strategic system for harvesting data from different types of studies (e.g., pilot, longitudinal, observational and quasi-experimental studies, and randomized controlled trials [RCTs]).
Use a Harmonized Core of Outcomes and a Taxonomy of Interventions to Enable Pooling of Study Findings
The variability of outcomes and measures used in dementia care research makes it challenging to pool results across studies for statistical analysis. Throughout the AHRQ systematic review, outcomes were often synthesized qualitatively for this reason (Butler et al., 2020). The problem is compounded by the difficulty of measuring social aspects of dementia (e.g.,
identity, autonomy, privacy, safety) and by the lack of sufficient measures that take into account how outcomes for persons living with dementia, care partners, and caregivers change in concert with the trajectory of illness, the helping network, and the environmental/service delivery context. Chapter 3 highlights this point, noting that outcomes related to quality of life and meaning are difficult to capture for persons living with dementia because of the cognitive impairments they experience, as well as the inherent goals of dementia care, services, and supports aimed at helping a person live well in the world. The AHRQ systematic review found that quality of life was rarely the outcome of primary interest in a study, and often not measured at all, despite its being a central goal of dementia care interventions (Butler et al., 2020). This gap may be due in part to the potential challenges of measuring quality of life as the disease progresses in persons living with dementia; however, a number of measures designed to assess this outcome in persons with late-stage Alzheimer’s disease and other dementias are available (e.g., Weiner et al., 2000).
As noted in Chapter 5, the AHRQ systematic review also noted a lack of clarity in how individual interventions are described within specific intervention categories, with outcomes often being measured and reported inconsistently, making it challenging to assess the evidence base (Butler et al., 2020). Many research publications examined in this and other related systematic reviews use inconsistent terminology to describe a particular intervention, and such inconsistency can often lead to discrepancies in classification (Gaugler et al., 2017). For example, the AHRQ systematic review notes that the terms “cognitive rehabilitation” and “cognitive training” were used interchangeably to describe intervention components within a single article, although the two terms have different meanings and interpretations (Butler et al., 2020).
To address these limitations, the committee agrees with the recommendation of the AHRQ systematic review regarding the need for better and consistent measures and measurements of psychosocial outcomes in persons living with dementia (Butler et al., 2020). In particular, having a general measure of well-being that could be adopted broadly is critical, particularly given the lack of consensus in the field as to what outcomes are most important to measure and matter to stakeholders (discussed further below). A harmonized core of meaningful outcomes is also important to help research converge with practice and policy; for example, greater attention is needed to identifying the types of measures that align with the Centers for Medicare & Medicaid Services’ (CMS’s) coverage decisions. Relatedly, measures are needed to examine the economic impacts of an intervention on persons living with dementia, care partners and caregivers, and other stakeholders, including cost-effectiveness and long-term viability (e.g., Gitlin et al., 2010; Livingston et al., 2020b). Interventions that do not improve outcomes and
reduce health care costs or provide enough benefit to justify additional expenditures are unlikely to be adopted beyond the research setting. Health care utilization and costs have implications for outcomes of importance to persons living with dementia, care partners, and caregivers, as well as the practicality of intervention dissemination (discussed further below).
In addition, a taxonomy of dementia care interventions needs to be developed to eliminate insufficient reporting, improve inferences of efficacy/effectiveness, and better understand how outcomes may differ by setting. As described in Chapter 3, dementia care interventions themselves are often complex, involving myriad interconnected components that interact with each other and with the context and system in which they are implemented. Using the organizational framework presented in that chapter as a guide, consideration of classifying interventions at all levels (i.e., individual/family, community, policy, and societal) is warranted.
Focus on Outcomes of Greatest Importance to Persons Living with Dementia and Their Care Partners and Caregivers
In reviewing the quality of the existing evidence on dementia care interventions, one finds that most studies to date have not taken stakeholder perspectives, and the diversity of those perspectives, into account from the outset. Chapter 2 highlights the field’s movement toward person-centered care; however, this term does not characterize the type of care most persons living with dementia currently receive. The broader task at hand is not only to assess the current evidence in order to decide which interventions to implement, but also to redefine the intervention development process to be more responsive to the needs of stakeholders—for example, designing interventions to address aspects of a person’s well-being, including personhood, financial strain, or social isolation; imparting to caregivers the skills to better manage complex medical conditions and medications; or reevaluating payment models to provide coverage for evidence-based interventions.1 In light of the COVID-19 pandemic, these evolving needs might also include the effects of social isolation on persons living with dementia and family care dyads, and the management of increasing medical complexity at home. In addition, a strengths-based approach (discussed in Chapter 2) is needed to examine interventions and outcome measures that focus on the strengths of persons living with dementia and their care partners and caregivers.
Outcomes that are important to persons living with dementia and their care partners and caregivers are also likely to vary depending on the age at onset and stage of disease and the needs at that time; for example, persons
1 Presented by Laura Gitlin of Drexel University at the Care Interventions for Individuals with Dementia and Their Caregivers workshop on April 15, 2020.
in the early stage of the disease who are independent and those in later stages will have different needs. The same is true for direct care workers as well (Jennings et al., 2017) (see Chapter 5). These variations over time have implications for the design of studies with longer durations, as there may be a need to modify the outcomes examined throughout the course of the study. Moreover, it will be important to elicit and assess goals that persons living with dementia and their care partners and caregivers may have for dementia care and how these goals may change along the continuum from early- to late-stage disease (Jennings et al., 2017, 2018). Identifying these evolving person-centered goals may help determine the outcomes that matter most to stakeholders at various time points and allow researchers to better evaluate the successes and shortfalls of an intervention. For example, Jennings and colleagues (2017, 2018) report that one goal noted among persons living with dementia and caregivers in terms of accessing services and supports was being able to feel that financial resources are not a barrier to care. Another study identified days spent at home (i.e., time spent out of hospitals, postacute rehabilitation, and nursing homes) as an outcome of importance to older adults, including persons living with dementia and their caregivers (Sayer, 2016). In light of the likelihood that the needs of persons living with dementia and their care partners and caregivers will evolve over time, research is needed to develop meaningful timescales for each outcome measure with an understanding of the expected pace of change in that measure, the ability for that behavior to be sustained, and the value placed on it by participants in the intervention. Needed as well are more effective ways of involving those who will implement emerging interventions (e.g., providers, persons living with dementia, care partners, and caregivers) in the process of intervention development.
Within the set of harmonized outcomes and measures for dementia care research, a person’s well-being is central. As noted in Chapters 1 and 2, this was a recurring theme highlighted among a group of persons living with dementia, care partners, and caregivers who served as advisers to the committee. Accordingly, it will be important for intervention studies to examine both the harms and benefits of an intervention, intended and unintended and for diverse stakeholders, including perspectives of future adopters. While caregiver burden was a common measure in the studies reviewed in the AHRQ systematic review, harms (e.g., elder abuse) were rarely assessed (Butler et al., 2020). Yet, Dong and colleagues (2014) report that, although 27.9–62.3 percent of older adults with dementia experience some form of psychological abuse and an estimated 3.5–23.1 percent experience physical abuse, elder abuse is often underreported in this population for a number of reasons (e.g., fear of retaliation and loss of support). Caregiver stress and burden is a common risk factor for elder abuse (Dong, 2017; Lee and Kolomer, 2005; Yan and Kwok, 2011). In a prior study of caregivers for persons living
with dementia, an expert panel found that 47 percent of persons living with dementia had experienced abuse, which corresponded to caregivers’ self-reported elder abuse (Wiglesworth et al., 2010). Caregiver anxiety, depressive symptoms, perceived burden, emotional status, and role limitations due to emotional problems were among the most common predictors of this abuse. “The combination of caregivers’ physical assault and psychological aggression provided the best sensitivity and specificity for elder mistreatment as defined by the expert panel” (Wiglesworth et al., 2010).
Similarly, the AHRQ systematic review revealed a lack of outcomes related to harms to care partners and caregivers. For example, it is estimated that severe aggression by persons living with dementia toward care partners and caregivers takes place at a rate greater than 20 percent and may be the strongest predictor of nursing home placement (Wharton and Ford, 2014). The same experiences occur among direct care workers, including those in long-term dementia care settings, who report higher rates of emotional and physical abuse by residents compared with nurses in hospitals (Boström et al., 2012).
The added stress and burdens caused by the COVID-19 pandemic have exacerbated some of the challenges faced by both persons living with dementia and their care partners and caregivers, resulting in increased risk for new abusive situations and potential increased severity of existing abusive relationships (Makaroun et al., 2020). These include elder abuse and caregiver and self-neglect, for which tools exist to predict at-risk individuals (Dong and Simon, 2014; Wang et al., 2020). Recognizing that outcomes related to intended and unintended harms are difficult to evaluate using quantitative metrics, further investment in qualitative and mixed-method research designs may be needed to capture these outcomes.
Include Qualitative Methods in Studies That Have Quantitative Outcomes
Given the complexity of dementia and the progression of the disease and care needs over time, qualitative and mixed-method research designs are needed to better understand the experience of persons living with dementia, care partners, and caregivers with a particular intervention and the context in which the intervention was implemented. While measurement of quantitative outcomes enables pooling of results across studies, qualitative methods offer greater insight into the acceptability and feasibility of an intervention and the relevance of context to its adoption among stakeholders. For example, qualitative methods might capture the experiences and perceptions of direct care workers caring for persons living with dementia, as well as the experiences and perceptions of that care among the persons living with dementia (Houghton et al., 2016; Reilly and Houghton, 2019). Synthesized qualitative findings can be developed
from a body of qualitative studies and graded using a process analogous to Grading of Recommendations Assessment, Development and Evaluation (GRADE). GRADE-CERQual (Confidence in the Evidence from Reviews of Qualitative research) is an approach that can be used to “assess how much confidence to place in findings from a qualitative evidence synthesis” using four criteria: methodological limitations, coherence, adequacy of data, and relevance (Lewin et al., 2015, 2018). Such evidence may complement the evaluation of the level of certainty of effectiveness for an intervention and help inform research, policy development, and practice.
Mixed-method research strategies have the potential to accelerate the translation of research findings into practice. As noted in Chapter 4, the hybrid effectiveness–implementation designs proposed by Curran and colleagues (2012) could help address issues related to internal and external validity and give researchers the latitude to study a range of outcomes (e.g., efficacy and implementation). Measuring these outcomes in both quantitative and qualitative ways would give researchers a better understanding of the implementation process, stakeholder perspectives, and areas for improvement (see Proctor et al., 2011). In addition, qualitative assessment of the factors outlined in the Normalization Process Theory (NPT) framework (described in Chapter 4)—coherence, cognitive participation, collective action, and reflexive monitoring—might help researchers understand how best to modify interventions as needed for different contexts and subpopulations (Murray et al., 2010).
Prior research in minority populations suggests that social and cultural context are important factors to consider (as discussed in greater detail below), but they are not commonly measured in traditional quantitative research (Brewster et al., 2019). Dementia is often perceived differently in various cultures (Dilworth-Anderson and Gibson, 2002), and substantial barriers exist in screening, diagnosing, and treating certain populations (Chin et al., 2011). Knowledge from the behavioral, social, and anthropological sciences may provide synergistic contributions to help address some of the above-noted limitations of dementia care research.
The committee acknowledges the ongoing efforts of NIH and NIA to institute new funding mechanisms that prioritize implementation science and the NIH Stage Model and align with the need for qualitative and mixed-method approaches.2 To continue to advance work in this domain, it may be necessary to establish expert working groups to develop standards and guidance on when best to use qualitative and mixed-method designs for dementia care research.
2 One such example is the NIA IMPACT (IMbedded Pragmatic Alzheimer’s disease and related dementias Clinical Trials) Collaboratory. For more information, see https://impactcollaboratory.org (accessed December 11, 2020).
Use Observational Studies as a Complement to Randomized Trials
While the AHRQ systematic review analytic set included only randomized trials, high-quality observational studies can also generate useful evidence on the effectiveness of interventions (see Table 6-1). Observational studies can provide insight into the magnitude of a problem and/or risk factors that serve as a point of intervention, and offer a cost-effective way to monitor progress toward remediation of identified problems over time. The data thus obtained can serve as benchmarks for interventionists or adopters of interventions in evaluating the populations they serve.
TABLE 6-1 Overview of Randomized and Nonrandomized/Observational Study Designs at the Individual and Community Levels, with Selected Examples
|Randomized Experiments||Nonrandomized Experiments and Observational Studies|
|Individual Level||Individually Randomized Trials
Randomized trials randomly assign individuals to one or more interventions, including a control, which may be standard care or some other comparator. The occurrence of an outcome is then compared between those assigned to each intervention.
The key advantage of randomized trials is that the groups receiving each intervention are, on average, comparable at the time of assignment to the interventions. However, some randomized trials have strict eligibility criteria or implement interventions in highly controlled conditions and thus do not enable evaluation of the interventions’ real-world effects. Pragmatic randomized trials address these issues by comparing realistic interventions in broader populations.
Observational follow-up studies compare the outcomes of individuals who happen to receive the interventions of interest without the investigators’ participation. Quasi-experiments (discussed further below under the community level) compare the outcomes of individuals whose interventions are assigned by the investigators but in a nonrandomized way. Because the assignment of interventions is not randomized, these designs may suffer from bias due to noncomparability between the groups receiving each intervention.
Follow-up studies may be particularly useful for generalizing results for purposes of clinical practice, because they, like pragmatic randomized trials, typically include individuals that more closely represent the target population, and they occur under real-world conditions. Statistical adjustment and sensitivity analyses are generally required to handle prognostic factors that are imbalanced across intervention groups.
|Randomized Experiments||Nonrandomized Experiments and Observational Studies|
|Community Level||Cluster Randomized Trials
Cluster randomized trials randomly assign groups, such as a local community or the patients seen at a specific clinic, to one or more interventions, including a control. The control may be standard care or some other comparator.
Quasi-experimental designs use a nonrandom method to assign participants to groups. Common types of these designs include nonequivalent group, posttest; nonequivalent group, pre–posttest; and interrupted time series.
|Cluster randomized trials may be particularly useful for evaluating public health, health policy, or health system interventions, since decisions to implement these interventions generally are made for a group of people (e.g., a community) rather than individuals. Cluster randomized trials are also used for interventions that are likely to be learned by participants who will have frequent contact with each other.||Quasi-experimental designs are used in the evaluation of interventions, including real-world effectiveness, when random assignment is not possible, ethical, or practical. While the findings may be more generalizable than those of randomized controlled trials, these designs are limited in their ability to determine the causal relationship between the intervention and outcome measure.|
|Natural Experiments of Policy Interventions
“Natural experiment” is a commonly used misnomer for an observational study in which groups of individuals receive a new intervention, often because of changes in health policies.
|Like cluster randomized trials, natural experiments are useful to study population-level interventions (e.g., changes in payment, or examination of differences in service outcomes by models). Like all observational studies, however, natural experiments are subject to bias due to noncomparability between the intervention groups.|
Numerous features of the research context for studying persons living with dementia and their care partners and caregivers—including substantial heterogeneity in outcomes that necessitates large sample sizes to achieve adequate power, the infeasibility of randomization in some settings and for some interventions, and the need for lengthy follow-up—make randomized trials difficult or infeasible to conduct. Longitudinal observational studies provide one means of generating complementary evidence for interventions that addresses these challenges (Concato et al., 2000). Such studies leverage existing or construct new cohorts to understand more precisely the risk/protective factors and potential causal mechanisms associated with the outcomes of interest. Given the rapid growth in U.S. minority populations, the construction of new representative, population-based state, regional, and national cohorts that remedy the limited diversity of existing cohorts discussed above is essential to improving the quality of the evidence base (an issue discussed in detail later in this chapter).
Finally, while the use of longitudinal observational studies has the potential to expand the evidence base, it is important that observational analyses be designed to emulate pragmatic randomized trials as closely as possible, when appropriate (Hernán and Robins, 2016). In some cases, observational studies may be used to study types of interventions (e.g., broad policy changes to reimbursement and regulations) for which pragmatic trials are not relevant (Craig et al., 2012). Further discussion of the use of these methods to assess the real-world effectiveness of an intervention is provided later in the chapter.
Commit to Comprehensive Study Reporting
The AHRQ systematic review highlights a variety of shortcomings in reporting of study results that impeded analysis. These included the need to improve and better understand fidelity in implementation, lack of reporting about the effects of the context in which an intervention was implemented, and lack of reporting on null findings or negative results and methodological approaches that did not work (Butler et al., 2020). Comprehensive study reporting would provide a richer evidence base on which to make decisions about implementation and dissemination.
Fidelity is the extent to which an intervention is administered as intended, in terms of both content and dose (Vernooij-Dassen and Moniz-Cook, 2014). An understanding of fidelity is critical to knowing the essential elements of an intervention and how it works. Problems with fidelity, regardless of the type of research, can have significant implications for the interpretation of findings, may result in a high risk of bias and the inability to replicate the study with the same level of effectiveness, and have a negative impact on translation and implementation. The inability
of a study to achieve fidelity across multiple sites may be a useful indication of whether an intervention is feasible or potentially shed light on contextual effects.
One challenge reported in the AHRQ systematic review is that interventions often have been conducted in selected populations, and their effectiveness when implemented more broadly and in other groups is unclear (Butler et al., 2020). This might be the case, for example, when an intervention tailored to the needs of a first-time care partner or caregiver is implemented with others who have prior experience caring for someone with dementia. Fidelity to an intervention may also vary depending on the number of care partners or caregivers a person living with dementia may have—none, one, or more than one (e.g., multiple family members taking shifts to help provide support and care to the person living with dementia).
The AHRQ systematic review found that many studies did not include detailed information on methodology outlining the delivery of the intervention, impeding an assessment of fidelity. In addition, fidelity to the intervention differed between formal and informal caregivers in the literature; measures to ensure that the intervention was delivered as designed were less likely to be used for informal caregivers.
Given the importance of implementation fidelity to understanding the effectiveness of an intervention, these are notable gaps. Furthermore, consensus is lacking in the field on the components of fidelity assessment approaches and how to measure them (Butler et al., 2020). Chapter 4 describes one conceptual framework for assessment of implementation fidelity, proposed by Carroll and colleagues (2007), which includes three areas of evaluation: (1) adherence; (2) moderators that might influence fidelity; and (3) identification of essential components of the intervention that have the most impact, which can be methodologically complex as it implies “breaking” the randomized assignment in the analysis.
The lack of standardized methods for improving implementation fidelity and for assessing how much real-world adaptation can be tolerated before fidelity is no longer realistic represents an important research need. In 2004, the Treatment Fidelity Workgroup of the National Institutes of Health Behavior Change Consortium made several recommendations for researchers to incorporate practices of treatment fidelity in their studies more consistently. The proposed strategies were related to the study design (e.g., ensure the same treatment dose across conditions); provider training (e.g., ensure provider skill acquisition); monitoring and improving the delivery of treatment (e.g., control for provider differences); receipt of treatment (e.g., ensure participant comprehension); and enactment of treatment skills (e.g., ensure participant use of cognitive skills) (Bellg et al., 2004). The adoption of similar approaches for dementia care research is needed to improve fidelity in the field.
Equally important is determining ways to better understand fidelity when its improvement may not be feasible. To this end, when interventions are implemented in real-world settings, researchers could provide a detailed report of specific adaptations made and analyze whether those differences affected the study outcomes. Also critical is examining and reporting on contextual effects, such as the care delivery system in which an intervention is delivered, and indicating whether the intervention was sustained after the study was completed. As discussed in Chapter 3, even when a community setting or system is not targeted by an intervention, the community, policy, and societal contexts in which the intervention is implemented may influence its effectiveness. Similarly, heterogeneity in study populations can contribute to variation in the observed effects of interventions. This context sensitivity contributes to the complexity of dementia care interventions and the challenges of evaluating their effectiveness, and understanding the effects of context is therefore critically important to implementation decisions. Moreover, variability in outcomes that is unattributed may lead to false conclusions about the ineffectiveness of interventions. Including and reporting on subgroup analyses/interaction testing in studies can help identify contextual effects and ultimately inform the better design of interventions and their targeting to those most likely to benefit.
To understand the contribution of contextual effects to observed variation in an intervention’s effectiveness across studies, sufficient detail is needed on the intervention’s implementation (how it was implemented and under which conditions). The TIDieR framework for intervention reporting (Hoffmann et al., 2014) is an extension of other reporting frameworks (e.g., CONSORT, SPIRIT) that goes beyond the description of an intervention to include contextual factors related to implementation, such as the intervention provider and setting. For large-scale research, including contextual variables in the analyses of specific interventions (e.g., Area Health Resources Files3) may help in better understanding outcomes. Realist review methods are increasingly being used to understand how complex public health, policy, and health services interventions work, for whom, and in which contexts (Pawson et al., 2005), and may be useful in elucidating these relationships for dementia care interventions.
One key contributor to advancing research is learning from others in the field about what has and has not worked. To this end, researchers have to commit to reporting on null findings, negative results, and methodological approaches not found to be successful. Doing so may illuminate for other researchers practices and interventions whose further implementation is not warranted (Largent and Karlawish, 2020), and, more important, is
3 For more information on Area Health Resources Files, see https://data.hrsa.gov/topics/health-workforce/ahrf (accessed October 13, 2020).
the researcher’s scientific and ethical obligation to the study participants. Yet, selective reporting of different outcomes that had a significant effect continues to bias the literature.
Critical to advancing the research enterprise, then, is addressing known barriers to reporting through the delivery of incentives and enforcement of requirements to report. While increasingly more funders are requiring researchers to report the results of their studies (NIH, 2017b), greater insights could be gained if researchers published their full protocols (e.g., in ClinicalTrials.gov) and made their data publicly available for others to review and possibly replicate. NIA, for example, is increasingly requiring data from its supported trials to be posted in repositories or consortiums for broad data sharing, when possible.4 When data were missing or not found to be significant, researchers could provide additional insights as to why that was the case (e.g., readiness of the researchers, lack of community buy-in or trust). Methodology papers detailing the delivery characteristics, fidelity, and setting (e.g., nursing homes, residential care facilities, adult day centers, individual homes) of an intervention might help address some of those questions.
According to the AHRQ systematic review,
dementia care research has been slow to incorporate key elements of rigorous intervention design. Until relatively recently, many dementia care intervention studies were not held to preregistration of trials, data safety and monitoring boards, or other standards more common in other areas of clinical science including reporting standards (e.g. the Consolidated Standards of Reporting Trials [CONSORT] statement). (Butler et al., 2020, p. 109)
Although NIH requires preregistration of all the clinical trials it funds (NIH, 2017a), this practice is variable across studies, and it is unclear how closely registration for trials is followed. In addition, the registration of observational studies remains an evolving area. Much of the early work in the field occurred before trial registration was mandated or the revised 2010 CONSORT standards existed, all of which contributes to the challenges in this field’s evidence base.
A critical aspect of strengthening the evidence base for care interventions for persons living with dementia, care partners, and caregivers is ensuring that the research is representative and generalizable across popu-
4 For more information on NIA Guidance for Sharing Data and Other Resources, see https://www.hhs.gov/guidance/document/nia-specific-funding-policies (accessed December 9, 2020).
lations. To this end, studies must have broadly inclusive research teams equipped with knowledge and experience working with underrepresented populations and in different settings; develop and adhere to recruitment strategies targeted at increasing the representation of racially, ethnically, culturally, linguistically, sexually, and socioeconomically diverse participants; and use study designs that support inclusivity. Strengthening the evidence base to advance the ultimate goal of improving well-being for all will require greater investments in increasing diversity across the entire research enterprise (e.g., researchers, study participants, and stakeholders), along with accountability measures to assess progress.
Conduct Studies Using Broadly Inclusive Research Teams
A lived experience cannot be taught. Diverse, multidisciplinary research teams are needed to ensure that team members collectively have insights into and sensitivity to different perspectives and cultures. Moreover, resources are needed to address knowledge and readiness gaps between researchers and stakeholders in the community. Researchers often do not know how to apply effective interventions in different populations and have not addressed fundamental questions before attempting to do so (Skinner et al., 2018). For example, how did the researchers conceptualize the problem in a social context, do they understand the social determinants of health, do they have knowledge of intersectionality and the optimal recruitment strategies for each targeted population, do they measure the intended and unintended consequences and harms of the intervention (Dong et al., 2014), and do they have the appropriate tools and training to apply the study methodology in diverse populations?
Before launching a study, then, researchers need to understand the people living in the community, as well as their history, culture, and resources. This point applies also to language and the use of terminology that resonates with the targeted population (e.g., “care partner” versus “caregiver”) (see Chapter 1), as well as an understanding of how different subpopulations may prioritize different needs. For example, in the Kame Project—a study designed to examine the rates of and risk factors for dementia and its subtypes among a Japanese American population in Seattle, Washington—the researchers took several steps to tailor the study to meet the needs of the community. These steps included hiring staff from the community, engaging a community advisory board, and adapting all of the study tools and instruments to ensure linguistic and cultural understanding (Graves et al., 1996). Understanding readiness within the community of interest is essential as well, as is determining what participants hope to gain from the study and how the researchers can engage them to be part of the research team to gain a better understanding of their cultural
values, belief systems, and what is important to them (e.g., serving on advisory boards). Once these questions have been answered, the researchers can determine what theory will drive the study, and then the specific methodology (Dilworth-Anderson et al., 2020).
Include Racially, Ethnically, Culturally, Linguistically, Sexually, and Socioeconomically Diverse Participants, and Assess Disparities in Access and Outcomes
As the racial and ethnic diversity of the U.S. population continues to rise (Frey, 2020; U.S. Census Bureau, 2020), it is projected that nearly half of Americans aged 65 or older will not be non-Hispanic whites by 2060 (U.S. Census Bureau, 2012). This demographic shift has clear implications for research on aging. For example, it is well known that racial and ethnic minorities (e.g., African Americans/Blacks, Latinos/Hispanics, American Indians/Alaska Natives) have a higher prevalence and incidence of Alzheimer’s disease and other dementias compared with older non-Hispanic whites, as well as different caregiver patterns and use of formal care (Dilworth-Anderson et al., 2008; Mayeda et al., 2016; Pinquart and Sörensen, 2011; Steenland et al., 2016). Yet, these populations are disproportionately excluded from dementia care research, as are low-income persons living with dementia and their care partners and caregivers (Dilworth-Anderson et al., 2020; Quiñones et al., 2020).
The AHRQ systematic review highlights this limitation, observing that few studies examined racial or ethnic differences (including subpopulations) and that culturally sensitive or culturally adapted interventions were rare (Butler et al., 2020). As noted in Chapter 5, the REACH II intervention and its adaptations were among the few interventions conducted in racially and ethnically diverse populations, delivered in multiple languages, and implemented in low-income communities or with low-income participants (Belle et al., 2006; Burgio et al., 2009; Cheung et al., 2014; Cho et al., 2019; Czaja et al., 2013, 2018; Lykens et al., 2014). Furthermore, the AHRQ review identified few interventions designed for low-resource areas or rural or tribal communities, other than pilot and small-sample studies, and no studies looked at LGBTQ populations or persons with Down syndrome with dementia (Butler et al., 2020). In short, there remains a critical gap in the development and implementation of dementia care interventions tailored to and driven by the needs of persons living with dementia, care partners, and caregivers from underrepresented groups and low-resource areas, as well as persons living with dementia who do not have care partners or caregivers.
To advance the field, it will be important to understand the cultural aspect of dementia and caregiving in diverse populations and the resultant
implications for methodological approaches and implementation (Apesoa-Varano et al., 2015; Chin et al., 2011). During the committee’s public workshop, J. Neil Henderson from the University of Minnesota noted that a better understanding of culture is needed in terms of designing and implementing care interventions.5 For example, some ethnocultural and indigenous populations whose numbers are small may be excluded from studies because of demands for large sample sizes. RCTs may also be unwelcomed in such communities because of the nature of the research design with respect to dividing the population into intervention and control groups. In such cases, delayed-start study designs used in other therapeutic areas, in which all study participants receive the intervention but at different time points, may be more appropriate (Crews et al., 2019; D’Agostino, 2009; Liu-Seifert et al., 2015; Tobe et al., 2014). Henderson noted that the term “culture” is often used incorrectly as a proxy when referring to minority populations, and he emphasized the importance of thinking about culture broadly as a process of adapting to life situations using precepts, beliefs, and values.6 He added that all people conduct caregiving in both macrocultural (e.g., the United States being a highly individualistic and independent society) and microcultural (e.g., cross-generational caregiving) systems. He stressed the critical importance of having people on research teams with deep knowledge of those cultural systems (including intragroup variance) who are involved in the study from the outset.
To achieve true progress in dementia care research, NIH will need to assume greater administrative accountability for ensuring increased representation of racial and ethnic minorities in research studies. The NIH Revitalization Act of 19937 amends the Public Health Service Act to incorporate a mandate for the inclusion of minorities in all NIH clinical research; however, progress on carrying out this mandate remains slow (Oh et al., 2015). The law was created in part as a result of the U.S. Public Health Service Syphilis Study at Tuskegee, in which African Americans were recruited for a study aimed at ensuring harm (Rencher and Wolf, 2013). Lack of adherence to this law has led to low percentages of underrepresented minorities in dementia clinical trials, representing a missed scientific opportunity to fully understand the effectiveness and safety of interventions in these populations, a gap that may exacerbate existing health disparities (Gilmore-Bykovskyi et al., 2018; Oh et al., 2015; Quiñones et al., 2020). Accordingly, follow-through by NIH and other
5 Presented by J. Neil Henderson of the University of Minnesota at the Care Interventions for Individuals with Dementia and Their Caregivers workshop on April 15, 2020.
6 Presented by J. Neil Henderson of the University of Minnesota at the Care Interventions for Individuals with Dementia and Their Caregivers workshop on April 15, 2020.
7 National Institutes of Health Revitalization Act of 1993, Public Law 103-43, 103 Cong. (June 10, 1993).
To close this gap, further training (i.e., retooling) is needed on learning and applying inclusive approaches in recruitment and on the impact of structural/system-level factors on recruitment approaches and successes (Dilworth-Anderson and Cohen, 2010; Hamel et al., 2016; Williams and Corbie-Smith, 2006). Such retooling can enable researchers to learn how to adapt, revise, and create new approaches to the recruitment and retention of diverse populations. It prepares researchers to use inclusive approaches to recruitment and retention of diverse populations in their research at both the conceptual and methodological levels, including thinking about and defining a problem and recruiting participants to help understand the problem. Such an approach goes beyond diversity to emphasize shared interest in and representation (inclusion) by researchers and participants in the research process. There are ongoing efforts to address this challenge at NIH and NIA, including the National Strategy for Recruitment and Participation in Alzheimer’s and Related Dementia Clinical Research,8 launched in 2018 (NIA, n.d.), and recruitment strategies of the Health and Retirement Study (Ofstedal and Weir, 2011). Nonetheless, progress remains slow.
Recognizing that a greater proportion of older adults reside in rural and remote settings than in urban environments (Smith and Trevelyan, 2019), greater attention to rural and low-resource areas is also needed (Prince et al., 2015). As noted above, few interventions considered in the AHRQ systematic review were designed for such areas, which therefore represent a critical gap in the implementation of care interventions to meet the needs of persons living with dementia, care partners, and caregivers. Closing this gap is a priority for NIA, as demonstrated by its funding opportunities and recent initiative to establish an Interdisciplinary Network on Rural Population Health and Aging,9 and continued efforts in this area are critical. In addition, U.S. Department of Veterans Affairs (VA) populations and safety net clinics may be another key target group setting for dementia care research.
Use Study Designs That Support Inclusivity
To address the diversity gaps detailed above, researchers need to use study designs that allow for inclusivity and increase the generalizability of research findings. Given the limitations of RCTs, the incorporation of other
8 For more information, see https://www.nia.nih.gov/research/recruitment-strategy (accessed October 20, 2020).
study designs, such as those discussed earlier in this chapter (e.g., quasi-experimental or longitudinal, well-designed observational studies; adaptive trial designs), is necessary to better understand whether an intervention will be effective in different subpopulations and settings. Intervention studies need to be designed with the goal of dissemination in mind, with consideration given to the potential for application in real-world care delivery and residential settings and in different cultural contexts (Damschroder et al., 2009; Green et al., 2009). As noted previously, including people from those different contexts and within the community in the study design can ensure the acceptability and feasibility of an intervention. The AHRQ systematic review notes that non-U.S.-based research may offer insights on future intervention adaptations for persons living with dementia, care partners, and caregivers with immigrant or related racial/ethnic heritages (Butler et al., 2020).
The Lancet Commission has reported that many risk factors for dementia cluster around inequalities that disproportionately affect minority and vulnerable populations, and emphasizes the need to tackle these inequalities through “not only health promotion but also societal action to improve the circumstances in which people live their lives” (Livingston et al., 2020a). For example, it has been shown that institutional racism is a factor in the social determinants of health that put minority populations at greater risk for disease, and also makes it more difficult to support and implement interventions in health systems (Brondolo et al., 2009; Chin et al., 2011; Phelan and Link, 2015). To address and help mitigate such inequalities through the continuum from prevention to late-stage care, a focus on inclusive research for dementia care interventions is needed as a complement to focusing on preventable risk factors.
In assessing the evidence base for dementia care interventions in the context of determining readiness for broad dissemination and implementation, the AHRQ systematic review was guided by the NIH Stage Model for Behavioral Interventions (Butler et al., 2020). This model delineates the full continuum of intervention research, ranging from basic science research and new intervention design to dissemination and implementation research (Onken et al., 2014). Interventions further along in that continuum (at Stage III or higher) have been studied under more real-world conditions and are more likely to be ready for broad dissemination. For the majority of interventions reviewed, however, the AHRQ review found few instances of progression along the NIH Stage Model beyond the basic explanatory stage (Stage III) (Butler et al., 2020), indicating little evidence of real-world effectiveness. As noted earlier in this chapter, these pilot
studies and studies with small sample sizes in the AHRQ systematic review are valuable for assessing feasibility or proof of concept. To be ready for broad dissemination and implementation, however, research must advance interventions further along the Stage Model and generate evidence about their application to individuals, communities, and systems. This section examines approaches to addressing this need.
Improve the Assessment of Individual-Level Interventions by Leveraging Complementary Study Methodologies
The framework for dementia care interventions presented in Chapter 3 depicts the various levels at which interventions may be implemented—at the individual, community, and policy levels. As discussed earlier in this chapter, much of the focus on interventions in the field has been on those targeting individual persons living with dementia, care partners, and caregivers. The effectiveness of these individual-level interventions can be assessed using randomized trials and observational studies that emulate randomized trials.
Randomized trials, and especially pragmatic trials, are helpful to establish effectiveness in the real world (see above). In some cases, pragmatic trials may be efficiently embedded in health care and other support delivery systems to leverage electronic health records (EHRs) for recruitment (to shorten enrollment times and improve recruitment of specific racial/ethnic groups or vulnerable populations), for data collection, or even as part of an intervention (e.g., alerts). However, the limitations of these trials need to be recognized. Although pragmatic trials are tested in real-world settings, they are still subject to contextual challenges that can impact the implementation and sustainability of the intervention (e.g., high caregiver turnover). Because randomized trials are logistically complex and expensive, they often have relatively small sample sizes and follow-up durations. As a result, they may yield imprecise effect estimates over short time horizons (Sim, 2019) or use proxy measures in lieu of clinically relevant outcomes (Krauss, 2018). In addition, failure to adhere to the trial protocol may obscure the effectiveness of otherwise promising interventions (if only adherence could be increased). That is, the usual intention-to-treat effect estimates need to be complemented with per-protocol effect estimates that adjust for deviations from the trial protocol (Hernán and Robins, 2017).
Observational studies may overcome some of the limitations of randomized trials, but they are vulnerable to several biases. First, the lack of randomized assignment may lead to noncomparable groups (Lu, 2009). The design of observational studies needs to incorporate the measurement of prognostic factors that may be imbalanced between intervention groups and that will have to be adjusted for in the statistical analyses. Second, a naïve
analysis of observational data may lead to serious biases, such as selection bias and immortal time bias. By designing observational analyses with the explicit goal of emulating a (hypothetical) pragmatic trial—the target trial—these biases can be reduced (Hernán et al., 2016). At the very least, sound observational analyses may help design better randomized trials.
Expand the Focus on Community/Policy-Level Interventions Using a Broad Set of Research Methodologies
Because much of the focus of dementia care research has been on interventions applied to individuals, the AHRQ systematic review includes a paucity of evidence for interventions applied to communities or to the entire system. Estimating the effectiveness of the latter interventions is difficult, especially when they are implemented in parallel with individual-level interventions, as it can be difficult to link many distal processes of care to desired outcomes (NQF, 2014). Improving the evidence for these kinds of interventions will necessitate methodologies other than those used to study individual interventions.
Cluster randomized trials (discussed earlier in this chapter) can be used to quantify the effectiveness of interventions implemented in communities (“clusters”). Observational studies that do not require individual-level data can also be used to evaluate these types of interventions or to help design cluster randomized trials. For example, ecological studies can be used to generate hypotheses that can later be studied in cluster randomized trials, and so-called quasi-experimental studies can estimate effects of a specific intervention by comparing the changes in outcomes over time between the intervention and control groups (e.g., using difference-in-difference analyses when the assumptions of the method hold).
Extreme instances of community-level interventions are those that are applied to an entire system in the form of new policies (e.g., payment change; state regulations; training grocery store clerks and bank tellers in the community to understand, recognize, and assist customers living with dementia to improve quality of life for them and their care partners and caregivers). The effect of system-wide interventions generally cannot be studied via randomized trials or observational studies that emulate randomized trials, and often cannot be studied using ecological or quasi-experimental approaches. Estimating the effects of system-wide interventions may therefore require the construction of policy models, and to this end, researchers can apply the modeling expertise developed in other health fields to expand their focus to system-wide interventions.
Address Key Factors Needed to Assess Real-World Effectiveness
As discussed in Chapter 4, decisions to implement interventions depend not only on evidence of efficacy from controlled research studies but also on factors that influence the feasibility of implementation (e.g., workforce needs, cost, alignment with current workflow), which will influence the intervention’s effectiveness in real-world settings. These considerations can be incorporated into decisions on whether interventions are ready to move into pragmatic trials (Baier et al., 2019) or should be recommended for dissemination and implementation (Moberg et al., 2018). As discussed in detail in Chapter 5, in considering whether either of the intervention types with low-strength evidence of effectiveness identified in the AHRQ systematic review were ready for broad dissemination and implementation, the committee evaluated the evidence using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) Evidence to Decision framework and noted significant gaps related to the quality and heterogeneity of the evidence. Such gaps will need to be addressed in future research to better inform the identification of interventions ready for broad dissemination and to meet the information needs of stakeholders (e.g., systems providing services and supports, payers, policy makers) responsible for making decisions about implementation and coverage of dementia care interventions.
NIA has already started to move in this direction, as exemplified by the development of the NIH Stage Model, as noted earlier in this chapter. In some cases, focused investigation of particular factors that are critical for assessing real-world effectiveness (e.g., financial constraints) could be conducted through studies that fit within NIH’s current R01 framework, incremental steps that would extend the R21 (exploratory/developmental grants) and R34 (grant planning program) approaches, or funding of replication studies.10 NIA could expand this approach, for example, by providing a road map for researchers offering funding support for each stage of the NIH Stage Model tailored to the unique challenges inherent in dementia care research and the many opportunities for future research identified in this chapter.
Approaches to addressing such gaps may include embedded pragmatic trials (which often involve stakeholders responsible for the adoption of interventions) and observational analyses carried out in practice settings that implement the interventions of interest (or by a third-party organization). In particular, a network of community-benefit organizations could help provide practical pragmatic, real-world information about the effectiveness, consistency, and impact of different organizations and service
10 For more information, see https://grants.nih.gov/grants/funding/funding_program.htm (accessed December 11, 2020).
providers, similar to a practiced-based research network.11 Observational studies can also be used to examine the relevance of policy and payment changes by geographies and organizational characteristics. The NIA IMPACT Collaboratory is building capacity to conduct embedded pragmatic clinical trials (ePCTs) to accelerate the translation of evidence-based interventions into practice as well (Mitchell et al., 2020).
As discussed previously, engaging the full range of stakeholders who will be using and implementing an intervention (e.g., persons living with dementia, care partners, and caregivers; community-based organizations; long-term services and supports providers; managed care organizations; health systems) in the design and evaluation of interventions can help ensure that implementation considerations (e.g., feasibility) are taken into account from the beginning and that interventions are appropriately tailored for the populations and contexts in which they will be implemented. This continues to be a priority for NIA, as demonstrated through the triennial National Research Summit on Care, Services, and Supports for Persons with Dementia and Their Caregivers, which engages a broad range of stakeholders, and the requests for applications emanating directly from those meetings.12
As discussed earlier, it is important to have evidence-based interventions that meet the needs of persons living with dementia, care partners, and caregivers at various time points in terms of the stage of disease and length of caregiving time.13 Community-based participatory research and community-partnered participatory research are existing models that support this kind of partnering. These models, which emphasize that the community of study and academic researchers are equal partners in the design, implementation, and dissemination of interventions (IOM, 2000; Viswanathan et al., 2004; Wells and Jones, 2009), have been applied in research with persons living with dementia, care partners, and caregivers, including a rural and remote dementia care program (e.g., Morgan et al., 2014). Morgan and colleagues note that stakeholder partnerships enhanced the quality, relevance, application, and sustainability of the research, which led to the adoption of a telehealth-delivered frontotemporal dementia support group model for a province-wide program in Canada and the transfer of the Rural and Remote Memory Clinic research project to a sustained program funded by the province’s ministry of health (Morgan et al., 2014).
Finally, while cost-effectiveness was beyond of the scope of the AHRQ systematic review and this committee did not conduct such an analysis, sev-
11 Presented by Patrick Courneya of HealthPartners at the Care Interventions for Individuals with Dementia and Their Caregivers workshop on April 15, 2020.
12 For more information, see https://aspe.hhs.gov/national-research-summit-care-services-and-supports-persons-dementia-and-their-caregivers (accessed December 11, 2020).
13 Presented by Kathleen Kelly of Family Caregiver Alliance at the Care Interventions for Individuals with Dementia and Their Caregivers workshop on April 15, 2020.
eral dementia care interventions have demonstrated benefits with respect to various types of utilization and overall Medicare costs, including research supported through Center for Medicare & Medicaid Innovation (CMMI) Health Care Innovation Awards (NORC at the University of Chicago, 2016). For example, the University of California, Los Angeles, Alzheimer’s and Dementia Care program observed significant reductions in hospitalizations for ambulatory care–sensitive conditions and 30-day readmissions, a 25 percent lower rate of nursing home placement, and a lower average cost of care among study participants (NORC at the University of Chicago, 2016).
While much progress has been made toward expanding and improving dementia care research, progress to date has been insufficient to meet the needs of the nation’s aging society with its increased numbers of persons needing services that advance their well-being. The evidence for care interventions for persons living with dementia, care partners, and caregivers remains complex and is lacking as the result of a number of methodological and implementation challenges. Over time, criteria for assessing the rigor and validity of research are becoming more standardized and rigorous, but this progress is not yet fully reflected in the overall body of literature assessed in the AHRQ systematic review. This progress should be encouraged, along with additional methodological improvements needed across the research enterprise to strengthen the rigor and representativeness of the evidence base for dementia care interventions at multiple levels, as well as the evidence base on the effect of interventions under real-world conditions.
The committee also emphasizes the significant impact the COVID-19 pandemic has had on quality of life for persons living with dementia and their care partners and caregivers, as well as the implications for research (e.g., challenges to study recruitment), implementation (e.g., decreased face-to-face interactions), and dissemination. This issue will require attention in the months and years to come.
The methodological improvements outlined in this chapter cannot all be achieved in a single study, but rather apply collectively across research in the field. In this context, it is essential for researchers, NIA, and other interested organizations to consider the specific actions they each can take to contribute to advancing dementia care research. Ensuring that well-being and personhood, as well as inclusive research, remain central is the responsibility of all.
To address the long-standing and urgent imperative to better support persons living with dementia and their care partners and caregivers in living as well as possible, there continues to be a critical need to build a more robust and useful evidence base. Studying dementia care interventions is challenging and complex, and the body of evidence is complicated to interpret. Two types of interventions are supported by sufficient evidence to warrant implementation in real-world settings, along with evaluation to continue to expand the evidence base. These interventions are practical instantiations of many of the core components of care, supports, and services, discussed in Chapter 2, that are needed to promote the well-being of persons living with dementia and their care partners and caregivers. Given current major deficits in the care, services, and supports that are available now, providing these interventions to those who could benefit would be a step forward. Yet, this is not a final answer. It is important that research continue to develop and evaluate other potentially promising interventions, many of which have shown some signal of benefit. The committee’s recommendations provide a path forward for building a more robust and useful evidence base by employing cutting-edge methods that are rigorous, most informative for this domain, inclusive, and equitable, and can yield critical information for real-world implementation. These exciting approaches can be implemented throughout the dementia care field, including by early-career researchers and others who want to harness new approaches to make a difference in addressing this critical societal need and better supporting
persons living with dementia and their care partners and caregivers in living as well as possible.
Apesoa-Varano, E. C., Y. Tang-Feldman, S. C. Reinhard, R. Choula, and H. M. Young. 2015. Multi-cultural caregiving and caregiver interventions: A look back and a call for future action. Generations 39(4):39–48.
Baier, R. R., E. Jutkowitz, S. L. Mitchell, E. McCreedy, and V. Mor. 2019. Readiness assessment for pragmatic trials (RAPT): A model to assess the readiness of an intervention for testing in a pragmatic trial. BMC Medical Research Methodology 19(1):156.
Belle, S. H., L. Burgio, R. Burns, D. Coon, S. J. Czaja, D. Gallagher-Thompson, L. N. Gitlin, J. Klinger, K. M. Koepke, C. C. Lee, J. Martindale-Adams, L. Nichols, R. Schulz, S. Stahl, A. Stevens, L. Winter, and S. Zhang. 2006. Enhancing the quality of life of dementia caregivers from different ethnic or racial groups: A randomized, controlled trial. Annals of Internal Medicine 145(10):727–738.
Bellg, A. J., B. Borrelli, B. Resnick, J. Hecht, D. S. Minicucci, M. Ory, G. Ogedegbe, D. Orwig, D. Ernst, and S. Czajkowski. 2004. Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH behavior change consortium. Health Psychology 23(5):443–451.
Berger, M. L., N. Dreyer, F. Anderson, A. Towse, A. Sedrakyan, and S. L. Normand. 2012. Prospective observational studies to assess comparative effectiveness: The ISPOR good research practices task force report. Value in Health 15(2):217–230.
Boström, A. M., J. E. Squires, A. Mitchell, A. E. Sales, and C. A. Estabrooks. 2012. Workplace aggression experienced by frontline staff in dementia care. Journal of Clinical Nursing 21(9–10):1453–1465.
Brewster, P., L. Barnes, M. Haan, J. K. Johnson, J. J. Manly, A. M. Nápoles, R. A. Whitmer, L. Carvajal-Carmona, D. Early, S. Farias, E. R. Mayeda, R. Melrose, O. L. Meyer, A. Zeki Al Hazzouri, L. Hinton, and D. Mungas. 2019. Progress and future challenges in aging and diversity research in the United States. Alzheimer’s & Dementia 15(7):995–1003.
Brondolo, E., L. C. Gallo, and H. F. Myers. 2009. Race, racism and health: Disparities, mechanisms, and interventions. Journal of Behavioral Medicine 32(1):1.
Burgio, L. D., I. B. Collins, B. Schmid, T. Wharton, D. McCallum, and J. Decoster. 2009. Translating the REACH caregiver intervention for use by area agency on aging personnel: The REACH OUT program. Gerontologist 49(1):103–116.
Butler, M., J. E. Gaugler, K. M. C. Talley, H. I. Abdi, P. J. Desai, S. Duval, M. L. Forte, V. A. Nelson, W. Ng, J. M. Ouellette, E. Ratner, J. Saha, T. Shippee, B. L. Wagner, T. J. Wilt, and L. Yeshi. 2020. Care interventions for people living with dementia and their caregivers. Comparative Effectiveness Review No. 231. Rockville, MD: Agency for Healthcare Research and Quality. doi: 10.23970/AHRQEPCCER231.
Carroll, C., M. Patterson, S. Wood, A. Booth, J. Rick, and S. Balain. 2007. A conceptual framework for implementation fidelity. Implementation Science 2(1):40.
Cheung, K. S., B. H. Lau, P. W. Wong, A. Y. Leung, V. W. Lou, G. M. Chan, and R. Schulz. 2014. Multicomponent intervention on enhancing dementia caregiver well-being and reducing behavioral problems among Hong Kong Chinese: A translational study based on REACH II. International Journal of Geriatric Psychiatry 30(5):460–469.
Chin, A. L., S. Negash, and R. Hamilton. 2011. Diversity and disparity in dementia: The impact of ethnoracial differences in Alzheimer disease. Alzheimer Disease and Associated Disorders 25(3):187–195.
Cho, J., S. Luk-Jones, D. R. Smith, and A. B. Stevens. 2019. Evaluation of REACH-TX: A community-based approach to the REACH II intervention. Innovation in Aging 3(3):igz022.
Concato, J., N. Shah, and R. I. Horwitz. 2000. Randomized, controlled trials, observational studies, and the hierarchy of research designs. New England Journal of Medicine 342(25):1887–1892.
Craig, P., C. Cooper, D. Gunnell, S. Haw, K. Lawson, S. Macintyre, D. Ogilvie, M. Petticrew, B. Reeves, M. Sutton, and S. Thompson. 2012. Using natural experiments to evaluate population health interventions: New medical research council guidance. Journal of Epidemiology and Community Health 66(12):1182.
Crews, D. C., A. M. Delaney, J. L. Walker Taylor, T. K. M. Cudjoe, M. Nkimbeng, L. Roberts, J. Savage, A. Evelyn-Gustave, J. Roth, D. Han, L. L. Boyér, R. J. Thorpe, D. L. Roth, L. N. Gitlin, and S. L. Szanton. 2019. Pilot intervention addressing social support and functioning of low socioeconomic status older adults with ESRD: The seniors optimizing community integration to advance better living with ESRD (SOCIABLE) study. Kidney Medicine 1(1):13–20.
Curran, G. M., M. Bauer, B. Mittman, J. M. Pyne, and C. Stetler. 2012. Effectiveness-implementation hybrid designs: Combining elements of clinical effectiveness and implementation research to enhance public health impact. Medical Care 50(3):217–226.
Czaja, S. J., D. Loewenstein, R. Schulz, S. N. Nair, and D. Perdomo. 2013. A videophone psychosocial intervention for dementia caregivers. The American Journal of Geriatric Psychiatry 21(11):1071–1081.
Czaja, S. J., C. C. Lee, D. Perdomo, D. Loewenstein, M. Bravo, J. H. Moxley, and R. Schulz. 2018. Community REACH: An implementation of an evidence-based caregiver program. Gerontologist 58(2):e130–e137.
D’Agostino, R. B., Sr. 2009. The delayed-start study design. New England Journal of Medicine 361(13):1304–1306.
Damschroder, L. J., D. C. Aron, R. E. Keith, S. R. Kirsh, J. A. Alexander, and J. C. Lowery. 2009. Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Science 4(1):50.
Dilworth-Anderson, P., and M. D. Cohen. 2010. Beyond diversity to inclusion: Recruitment and retention of diverse groups in Alzheimer research. Alzheimer Disease and Associated Disorders 24(Suppl):S14–S18.
Dilworth-Anderson, P., and B. E. Gibson. 2002. The cultural influence of values, norms, meanings, and perceptions in understanding dementia in ethnic minorities. Alzheimer’s Disease and Associated Disorders 16:S56–S63.
Dilworth-Anderson, P., S. Thaker, and J. M. D. Burke. 2005. Recruitment strategies for studying dementia in later life among diverse cultural groups. Alzheimer Disease and Associated Disorders 19(4).
Dilworth-Anderson, P., H. C. Hendrie, J. J. Manly, A. S. Khachaturian, and S. Fazio. 2008. Diagnosis and assessment of Alzheimer’s disease in diverse populations. Alzheimer’s & Dementia 4(4):305–309.
Dilworth-Anderson, P., H. Moon, and M. P. Aranda. 2020. Dementia caregiving research: Expanding and reframing the lens of diversity, inclusivity, and intersectionality. Gerontologist 60(5):797–805.
Dong, X. 2017. Elder abuse: Research, practice and policy, 1st ed. Springer International Publishing.
Dong, X., and M. A. Simon. 2014. Vulnerability risk index profile for elder abuse in a community-dwelling. Journal of the American Geriatrics Society 62(1):10–15.
Dong, X., R. Chen, and M. A. Simon. 2014. Elder abuse and dementia: A review of the research and health policy. Health Affairs 33(4):642–649.
Emsley, R., G. Dunn, and I. R. White. 2009. Mediation and moderation of treatment effects in randomised controlled trials of complex interventions. Statistical Methods in Medical Research 19(3):237–270.
Frey, W. H. 2020. The nation is diversifying even faster than predicted, according to new census data. https://www.brookings.edu/research/new-census-data-shows-the-nation-is-diversifying-even-faster-than-predicted (accessed August 12, 2020).
Gaugler, J. E., E. Jutkowitz, T. P. Shippee, and M. Brasure. 2017. Consistency of dementia caregiver intervention classification: An evidence-based synthesis. International Psychogeriatrics 29(1):19–30.
Gilmore-Bykovskyi, A., R. Johnson, L. Walljasper, L. Block, and N. Werner. 2018. Under-reporting of gender and race/ethnicity differences in NIH-funded dementia caregiver support interventions. American Journal of Alzheimer’s Disease and Other Dementias 33(3):145–152.
Gilmore-Bykovskyi, A. L., Y. Jin, C. Gleason, S. Flowers-Benton, L. M. Block, P. Dilworth-Anderson, L. L. Barnes, M. N. Shah, and M. Zuelsdorff. 2019. Recruitment and retention of underrepresented populations in Alzheimer’s disease research: A systematic review. Alzheimer’s & Dementia: Translational Research & Clinical Interventions 5:751–770.
Gitlin, L. N., N. Hodgson, E. Jutkowitz, and L. Pizzi. 2010. The cost-effectiveness of a nonpharmacologic intervention for individuals with dementia and family caregivers: The tailored activity program. American Journal of Geriatric Psychiatry 18(6):510–519.
Graves, A. B., E. B. Larson, S. D. Edland, J. D. Bowen, W. C. McCormick, S. M. McCurry, M. M. Rice, A. Wenzlow, and J. M. Uomoto. 1996. Prevalence of dementia and its subtypes in the Japanese American population of King County, Washington State: The Kame Project. American Journal of Epidemiology 144(8):760–771.
Green, L. W., J. M. Ottoson, C. García, and R. A. Hiatt. 2009. Diffusion theory and knowledge dissemination, utilization, and integration in public health. Annual Review of Public Health 30:151–174.
Gribbons, B., and J. Herman. 1996. True and quasi-experimental designs. Practical Assessment, Research, and Evaluation 5(1):14.
Hamel, L. M., L. A. Penner, T. L. Albrecht, E. Heath, C. K. Gwede, and S. Eggly. 2016. Barriers to clinical trial enrollment in racial and ethnic minority patients with cancer. Cancer Control 23(4):327–337.
Hernán, M. A., and J. M. Robins. 2016. Using big data to emulate a target trial when a randomized trial is not available. American Journal of Epidemiology 183(8):758–764.
Hernán, M. A., and J. M. Robins. 2017. Per-protocol analyses of pragmatic trials. New England Journal of Medicine 377(14):1391–1398.
Hernán, M. A., B. C. Sauer, S. Hernández-Díaz, R. Platt, and I. Shrier. 2016. Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses. Journal of Clinical Epidemiology 79:70–75.
HHS (U.S. Department of Health and Human Services). 2018. Report to the National Advisory Council on Alzheimer’s Research, Care, and Services. National Research Summit on Care, Services, and Supports for Persons with Dementia and Their Caregivers. Bethesda, MD: U.S. Department of Health and Human Services.
Hoffmann, T. C., P. P. Glasziou, I. Boutron, R. Milne, R. Perera, D. Moher, D. G. Altman, V. Barbour, H. Macdonald, M. Johnston, S. E. Lamb, M. Dixon-Woods, P. McCulloch, J. C. Wyatt, A.-W. Chan, and S. Michie. 2014. Better reporting of interventions: Template for intervention description and replication (TIDIER) checklist and guide. BMJ 348:g1687.
Houghton, C., K. Murphy, D. Brooker, and D. Casey. 2016. Healthcare staffs’ experiences and perceptions of caring for people with dementia in the acute setting: Qualitative evidence synthesis. International Journal of Nursing Studies 61:104–116.
IOM (Institute of Medicine). 2000. Promoting health: Intervention strategies from social and behavioral research. Washington, DC: National Academy Press.
Jennings, L. A., A. Palimaru, M. G. Corona, X. E. Cagigas, K. D. Ramirez, T. Zhao, R. D. Hays, N. S. Wenger, and D. B. Reuben. 2017. Patient and caregiver goals for dementia care. Quality of Life Research 26(3):685–693.
Jennings, L. A., K. D. Ramirez, R. D. Hays, N. S. Wenger, and D. B. Reuben. 2018. Personalized goal attainment in dementia care: Measuring what persons with dementia and their caregivers want. Journal of the American Geriatrics Society 66(11):2120–2127.
Krauss, A. 2018. Why all randomised controlled trials produce biased results. Annals of Medicine 50(4):312–322.
Largent, E. A., and J. Karlawish. 2020. Rescuing research participants after Alzheimer trials stop early: Sending out an SOS. JAMA Neurology 77(4):413–414.
Lee, M., and S. Kolomer. 2005. Caregiver burden, dementia, and elder abuse in South Korea. Journal of Elder Abuse & Neglect 17(1):61–74.
Lewin, S., C. Glenton, H. Munthe-Kaas, B. Carlsen, C. J. Colvin, M. Gülmezoglu, J. Noyes, A. Booth, R. Garside, and A. Rashidian. 2015. Using qualitative evidence in decision making for health and social interventions: An approach to assess confidence in findings from qualitative evidence syntheses (GRADE-CERQual). PLoS medicine 12(10):e1001895.
Lewin, S., A. Booth, C. Glenton, H. Munthe-Kaas, A. Rashidian, M. Wainwright, M. A. Bohren, Ö. Tunçalp, C. J. Colvin, R. Garside, B. Carlsen, E. V. Langlois, and J. Noyes. 2018. Applying GRADE-CERQual to qualitative evidence synthesis findings: Introduction to the series. Implementation Science 13(1):2.
Liu-Seifert, H., S. W. Andersen, I. Lipkovich, K. C. Holdridge, and E. Siemers. 2015. A novel approach to delayed-start analyses for demonstrating disease-modifying effects in Alzheimer’s disease. PLoS One 10(3):e0119632.
Livingston, G., J. Huntley, A. Sommerlad, D. Ames, C. Ballard, S. Banerjee, C. Brayne, A. Burns, J. Cohen-Mansfield, C. Cooper, S. G. Costafreda, A. Dias, N. Fox, L. N. Gitlin, R. Howard, H. C. Kales, M. Kivimäki, E. B. Larson, A. Ogunniyi, V. Orgeta, K. Ritchie, K. Rockwood, E. L. Sampson, Q. Samus, L. S. Schneider, G. Selbæk, L. Teri, and N. Mukadam. 2020a. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The Lancet 396(10248):413–446.
Livingston, G., M. Manela, A. O’Keeffe, P. Rapaport, C. Cooper, M. Knapp, D. King, R. Romeo, Z. Walker, J. Hoe, C. Mummery, and J. Barber. 2020b. Clinical effectiveness of the start (strategies for relatives) psychological intervention for family carers and the effects on the cost of care for people with dementia: 6-year follow-up of a randomised controlled trial. British Journal of Psychiatry 216(1):35–42.
Lu, C. Y. 2009. Observational studies: A review of study designs, challenges and strategies to reduce confounding. International Journal of Clinical Practice 63(5):691–697.
Lykens, K., N. Moayad, S. Biswas, C. Reyes-Ortiz, and K. P. Singh. 2014. Impact of a community based implementation of REACH II program for caregivers of Alzheimer’s patients. PLoS One 9(2):e89290.
Makaroun, L. K., R. L. Bachrach, and A.-M. Rosland. 2020. Elder abuse in the time of COVID-19-increased risks for older adults and their caregivers. The American Journal of Geriatric Psychiatry 28(8):876–880.
Mayeda, E. R., M. M. Glymour, C. P. Quesenberry, and R. A. Whitmer. 2016. Inequalities in dementia incidence between six racial and ethnic groups over 14 years. Alzheimer’s & Dementia 12(3):216–224.
Mitchell, S. L., V. Mor, J. Harrison, and E. P. McCarthy. 2020. Embedded pragmatic trials in dementia care: Realizing the vision of the NIA IMPACT Collaboratory. Journal of the American Geriatrics Society 68(Suppl 2):S1–S7. https://doi.org/10.1111/jgs.16621.
Moberg, J., and M. Kramer. 2015. A brief history of the cluster randomised trial design. Journal of the Royal Society of Medicine 108(5):192–198.
Moberg, J., A. D. Oxman, S. Rosenbaum, H. J. Schünemann, G. Guyatt, S. Flottorp, C. Glenton, S. Lewin, A. Morelli, G. Rada, P. Alonso-Coello, E. Akl, M. Gulmezoglu, R. A. Mustafa, J. Singh, E. von Elm, J. Vogel, and J. Watine. 2018. The GRADE Evidence to Decision (EtD) framework for health system and public health decisions. Health Research Policy and Systems 16(1):45.
Morgan, D., M. Crossley, N. Stewart, A. Kirk, D. Forbes, C. D’Arcy, V. Dal Bello-Haas, L. McBain, M. O’Connell, J. Bracken, J. Kosteniuk, and A. Cammer. 2014. Evolution of a community-based participatory approach in a rural and remote dementia care research program. Progress in Community Health Partnerships 8(3):337–345.
Murray, E., S. Treweek, C. Pope, A. MacFarlane, L. Ballini, C. Dowrick, T. Finch, A. Kennedy, F. Mair, C. O’Donnell, B. N. Ong, T. Rapley, A. Rogers, and C. May. 2010. Normalisation process theory: A framework for developing, evaluating and implementing complex interventions. BMC Medicine 8(1):63.
NIA (National Institute on Aging). 2018. Stage model for behavioral intervention development. https://www.nia.nih.gov/research/dbsr/nih-stage-model-behavioral-intervention-development (accessed August 18, 2020).
NIA. n.d. National strategy for recruitment and participation in Alzheimer’s and related dementias clinical research. https://www.nia.nih.gov/research/recruitment-strategy (accessed November 25, 2020).
NIH (National Institutes of Health). 2017a. NIH policy on the dissemination of NIH-funded clinical trial information. https://grants.nih.gov/policy/clinical-trials/reporting/understanding/nih-policy.htm (accessed August 12, 2020).
NIH. 2017b. Requirements for registering & reporting NIH-funded clinical trials in clinicaltrials.Gov. https://grants.nih.gov/policy/clinical-trials/reporting/index.htm (accessed December 11, 2020).
NORC at the University of Chicago. 2016. Third annual report: HCIA disease-specific evaluation. Bethesda, MD: NORC at the University of Chicago.
NQF (National Quality Forum). 2014. Priority setting for healthcare performance measurement: Addressing performance measure gaps for dementia, including alzheimer’s disease. http://www.qualityforum.org/priority_setting_for_healthcare_performance_measurement_alzheimers_disease.aspx (accessed September 4, 2020).
Ofstedal, M. B., and D. R. Weir. 2011. Recruitment and retention of minority participants in the health and retirement study. The Gerontologist 51(Suppl 1):S8–S20. https://doi.org/10.1093/geront/gnq100.
Oh, S. S., J. Galanter, N. Thakur, M. Pino-Yanes, N. E. Barcelo, M. J. White, D. M. de Bruin, R. M. Greenblatt, K. Bibbins-Domingo, A. H. B. Wu, L. N. Borrell, C. Gunter, N. R. Powe, and E. G. Burchard. 2015. Diversity in clinical and biomedical research: A promise yet to be fulfilled. PLoS Medicine 12(12):e1001918.
Onken, L. S., K. M. Carroll, V. Shoham, B. N. Cuthbert, and M. Riddle. 2014. Reenvisioning clinical science: Unifying the discipline to improve the public health. Clinical Psychological Science 2(1):22–34.
Pawson, R., T. Greenhalgh, G. Harvey, and K. Walshe. 2005. Realist review—A new method of systematic review designed for complex policy interventions. Journal of Health Services Research & Policy 10(Suppl 1):21–34.
Phelan, J. C., and B. G. Link. 2015. Is racism a fundamental cause of inequalities in health? Annual Review of Sociology 41(1):311–330.
Pinquart, M., and S. Sörensen. 2011. Spouses, adult children, and children-in-law as caregivers of older adults: A meta-analytic comparison. Psychology and Aging 26(1):1–14.
Prince, M. J., A. Wimo, M. Guerchet, G. Ali, Y.-T. Wu, M. Prina, and Alzheimer’s Disease International. 2015. World Alzheimer report 2015: The global impact of dementia: An analysis of prevalence, incidence, cost and trends. London, UK: Alzheimer’s Disease International.
Proctor, E., H. Silmere, R. Raghavan, P. Hovmand, G. Aarons, A. Bunger, R. Griffey, and M. Hensley. 2011. Outcomes for implementation research: Conceptual distinctions, measurement challenges, and research agenda. Administration and Policy in Mental Health 38(2):65–76.
Quiñones, A. R., S. L. Mitchell, J. D. Jackson, M. P. Aranda, P. Dilworth-Anderson, E. P. McCarthy, and L. Hinton. 2020. Achieving health equity in embedded pragmatic trials for people living with dementia and their family caregivers. Journal of the American Geriatrics Society 68(S2):S8–S13.
Reilly, J. C., and C. Houghton. 2019. The experiences and perceptions of care in acute settings for patients living with dementia: A qualitative evidence synthesis. International Journal of Nursing Studies 96:82–90.
Rencher, W. C., and L. E. Wolf. 2013. Redressing past wrongs: Changing the common rule to increase minority voices in research. American Journal of Public Health 103(12):2136–2140.
Richiardi, L., R. Bellocco, and D. Zugna. 2013. Mediation analysis in epidemiology: Methods, interpretation and bias. International Journal of Epidemiology 42(5):1511–1519.
Sayer, C. 2016. “Time spent at home”—A patient-defined outcome. NEJM Catalyst 2(2).
Sim, J. 2019. Should treatment effects be estimated in pilot and feasibility studies? Pilot and Feasibility Studies 5(1):107.
Skinner, J. S., N. A. Williams, A. Richmond, J. Brown, A. H. Strelnick, K. Calhoun, E. H. De Loney, S. Allen, A. Pirie, and C. H. Wilkins. 2018. Community experiences and perceptions of clinical and translational research and researchers. Progress in Community Health Partnerships 12(3):263–271.
Smith, A. S., and E. Trevelyan. 2019. The older population in rural America: 2012–2016. Washington, DC: U.S. Census Bureau.
Steenland, K., F. C. Goldstein, A. Levey, and W. Wharton. 2016. A meta-analysis of Alzheimer’s disease incidence and prevalence comparing African-Americans and Caucasians. Journal of Alzheimer’s Disease 50(1):71–76.
Tobe, S. W., M. Moy Lum-Kwong, S. Von Sychowski, K. Kandukur, A. Kiss, and V. Flintoft. 2014. Hypertension management initiative prospective cohort study: Comparison between immediate and delayed intervention groups. Journal of Human Hypertension 28(1):44–50.
U.S. Census Bureau. 2012. U.S. Census Bureau projections show a slower growing, older, more diverse nation a half century from now. https://www.census.gov/newsroom/releases/archives/population/cb12-243.html (accessed August 12, 2020).
U.S. Census Bureau. 2020. National population by characteristics: 2010–2019. https://www.census.gov/data/tables/time-series/demo/popest/2010s-national-detail.html (accessed August 10, 2020).
Vernooij-Dassen, M., and E. Moniz-Cook. 2014. Raising the standard of applied dementia care research: Addressing the implementation error. Aging & Mental Health 18(7):809–814.
Viswanathan, M., A. Ammerman, E. Eng, G. Garlehner, K. N. Lohr, D. Griffith, S. Rhodes, C. Samuel-Hodge, S. Maty, and L. Lux. 2004. Community based participatory research: Assessing the evidence: Summary. In AHRQ evidence report summaries. Agency for Healthcare Research and Quality (US).
Wang, B., D. R. Hoover, T. Beck, and X. Dong. 2020. A vulnerability risk index of self-neglect in a community-dwelling older population. Journal of the American Geriatrics Society 68(4):809–816.
Weiner, M. F., K. Martin-Cook, D. A. Svetlik, K. Saine, B. Foster, and C. S. Fontaine. 2000. The quality of life in late-stage dementia (QUALID) scale. Journal of the American Medical Directors Association 1(3):114–116.
Wells, K., and L. Jones. 2009. “Research” in community-partnered, participatory research. Journal of the American Medical Association 302(3):320–321.
Wharton, T. C., and B. K. Ford. 2014. What is known about dementia care recipient violence and aggression against caregivers? Journal of Gerontological Social Work 57(5):460–477.
Wiegersma, P. A., A. Hofman, and G. A. Zielhuis. 2001. Evaluation of community-wide interventions: The ecologic case-referent study design. European Journal of Epidemiology 17(6):551–557.
Wiglesworth, A., L. Mosqueda, R. Mulnard, S. Liao, L. Gibbs, and W. Fitzgerald. 2010. Screening for abuse and neglect of people with dementia. Journal of the American Geriatrics Society 58(3):493–500.
Williams, I. C., and G. Corbie-Smith. 2006. Investigator beliefs and reported success in recruiting minority participants. Contemporary Clinical Trials 27(6):580–586.
Yan, E., and T. Kwok. 2011. Abuse of older Chinese with dementia by family caregivers: An inquiry into the role of caregiver burden. International Journal of Geriatric Psychiatry 26(5):527–535.