This report is available in PDF only (Draft Report [2.7 MB]; Draft Appendixes [4.9 MB]). People using assistive technology may not be able to fully access information in these files. For additional assistance, please contact us.
Objective. To understand the evidence base for care interventions for people with dementia (PWD) and their caregivers, and to assess the potential for broad dissemination and implementation of that evidence.
Data sources. We searched Ovid Medline, Ovid Embase, Ovid PsycINFO, CINAHL, and the Cochrane Central Register of Controlled Trials (CENTRAL) to identify randomized controlled trials, nonrandomized controlled trials, and quasi-experimental designs published and indexed in bibliographic databases through October 2019.
Review methods. We searched for nondrug interventions targeting PWD, their informal or formal caregivers, or health systems. Two investigators screened abstracts and full-text articles of identified references for eligibility. Eligible studies included randomized controlled trials and quasi-experimental observational studies enrolling people with Alzheimer's disease or related dementias or their informal or formal caregivers. We extracted basic study information from all eligible studies. We assessed risk of bias, and summarized results for studies not judged to be NIH Stage Model 0 to 2 (pilot or small sample size studies) or to have high risk of bias. We grouped interventions into categories empirically based on intervention target.
Results. We identified 8409 unique references, of which 595 unique studies with an additional 255 companion articles were eligible. We classified interventions into 37 major categories. With few exceptions, we did not combine data quantitatively due to variability of interventions, comparison groups, outcomes measured, and study timing. Low-strength evidence shows that an intensive multicomponent intervention for informal caregiver support, with education, group discussion, in-home and phone support, and caregiver feedback (REACH II), may improve informal caregiver depression at 6 months. Low-strength evidence also shows that collaborative care models may improve quality of life for PWD and health system-level markers, including improvements in guideline-based quality indicators and reducing emergency room visits. The literature does not allow for further determination of whether the very small to small average effects applied to all enrolled PWD or if larger effects were concentrated in an unidentified subgroup. For all other interventions and outcomes, we found the evidence insufficient to draw conclusions. Insufficient evidence does not mean that the intervention is determined to be of no value to PWD or their caregivers. Rather, it means that due to the uncertainty of the evidence, we could not draw meaningful conclusions at this time.
Conclusions. Very little evidence supports widespread dissemination of any general care approaches for PWD. This review demonstrates the need for larger, longer-term, and more rigorous studies of interventions.
- Intensive multicomponent intervention with education, group discussion, in-home and phone support sessions, and caregiver feedback for informal caregiver support (REACH II), may improve informal caregiver depression at 6 months. (low-strength evidence)
- Collaborative care models may improve PWD quality of life. (low-strength evidence) The literature does not allow for further determination of whether the very small to small average effects applied to all enrolled PWD or if larger effects were concentrated in an unidentified subgroup.
- Collaborative care models may improve system-level markers, including guideline-based quality indicators and reduction in emergency department visits. (low-strength evidence)
- For all other outcomes and interventions, we found the evidence was insufficient because the uncertainty of the evidence was too high to draw conclusions.
Background and Purpose
The aging of the U.S. population and the concurrent rise in the number of adults living with dementia underscore the urgent need for a systematic review of the available evidence for care interventions for people with dementia (PWD) and their formal and informal caregivers. The National Institute on Aging (NIA) commissioned such a review from the Evidence-based Practice Center Program at the Agency for Healthcare Research and Quality (AHRQ). Dementia affects about 5 million U.S. adults 65 years and older (with disparities by race and ethnicity), and that number may grow to almost 14 million by 2060.1,2
The goal is to understand the evidence base for effective care interventions, and to assess the potential for broad dissemination and implementation of that evidence. Subsequently, a NASEM committee of experts, established at the request of the NIA, will use the evidence findings delivered in this report to help develop its own independent recommendations regarding which care interventions are supported by sufficient evidence to be widely disseminated and implemented, as well as to identify research gaps.
The methods for this systematic review follow the Agency for Healthcare Research & Quality (AHRQ) Methods Guide for Effectiveness and Comparative Effectiveness Reviews.9 See the review protocol and the full report of the review for additional details. We searched Ovid Medline, Ovid Embase, Ovid PsycINFO, CINAHL, and the Cochrane Central Register of Controlled Trials (CENTRAL) to identify randomized controlled trials, nonrandomized controlled trials, and quasi-experimental designs published and indexed in bibliographic databases through October 2019.
We identified 8409 unique references, 850 of which were eligible for our review, comprising 595 unique studies with an additional 255 companion publications. We sorted eligible studies into 37 major intervention categories. Approximately 60 percent of the literature emerged from research conducted outside of the United States.
While the literature was highly diverse, we found little information to determine whether interventions are equally appropriate for or have been successfully adapted to other race/ethnic cultures within the United States, with only a handful of studies providing sufficient inclusion of African-American or Hispanic/Latino populations. Interventions for rural communities were found to be even more rare in the literature. Additionally, the many countries in which care interventions were evaluated were almost exclusively high-resource, with very few low-resource countries represented. Finally, few studies gathered the granular detail necessary for a deeper understanding of the applicability of the interventions. Many important groups were rarely studied, including people with Down syndrome, who are living longer and who overall experience higher rates of dementia than the general population, and individuals with complex presentations of dementia.
We found low-strength evidence that one multicomponent intervention for informal caregiver support, REACH II (comprised of education, group discussion, in-home and phone support sessions, and caregiver feedback), may improve caregiver depression.3-5 Collaborative care models may improve quality of life for PWD6-9 and health system-level markers, including improvements in guideline-based quality indicators7,8 and reducing emergency room visits.6 For all other interventions and outcomes, we found the uncertainty of the evidence was too high to draw conclusions. The vast majority of studies had small sample size, were pilots that had not undergone traditional efficacy testing, and/or had high risk of bias, and provided insufficient evidence to draw conclusions.
As stated, the goal of the review was to understand the evidence base for effective care interventions for PWD and their caregivers in order to assess the potential for broad dissemination and implementation of that evidence. All decisions about the review methodology followed from this goal, which has implications for our findings. Therefore, some specific approaches for a particular intervention, or even whole classes of interventions, may not have been captured. Similarly, we may have missed some community services and support approaches such as tool kits, referral services and links, or awareness-raising outreach.
Because we excluded studies with fewer than 10 participants per study arm, we may not have identified some interventions with only very preliminary research. Our approach to risk of bias assessment was generous, relative to how risk of bias is assessed in more targeted systematic review topics. This is in part due to the unusually varied studies included in this review as well as the complexity of the condition and the care approaches.
Furthermore, included studies had methodological problems such as low subject retention, widely varying measures of success, and relatively small size trials that may have lacked the power to detect benefits. These problems stem from well-recognized challenges in researching these populations; therefore, we erred on the side of assessing a body of research as insufficient rather than ineffective. Additionally, given the progressive nature of dementia and the anticipated increase in care needs over time, it may be difficult to identify improvement in caregiver burden. The goal may in fact be to slow the rate of burden—and studies may be underpowered to detect such a small effect.
Finally, given the breadth of the topic, our systematic review is naturally reductionist in nature. That is, small but true differences may exist between many of the categories we summarized into our outcomes.
Implications and Conclusions
Ultimately, we uncovered very little evidence to support interventions and programs for active, widespread dissemination because evidence was insufficient to draw conclusions about the effects of the vast majority of interventions studied.
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 reporting standards (e.g. the CONSORT statement), pre-registration of trials, data safety and monitoring boards, or other standards more common in other areas of clinical science. As a direct result, despite a few positive findings, our global conclusions, largely similar to past reviews, are that the amount of high-quality evidence is insufficient to draw firm conclusions about interventions.
In order for federal funders and stakeholders to expedite the translational pipeline of idea development to implementation, as they aim to do, critical improvements are needed in dementia care research. Only with such improvements will we be able to draw clearer, less ambiguous conclusions related to efficacy.
- Matthews KA, Xu W, Gaglioti AH, et al. Racial and ethnic estimates of Alzheimer's disease and related dementias in the United States (2015-2060) in adults aged >65 years. Alzheimers Dement. 2019 Jan;15(1):17-24. doi: 10.1016/j.jalz.2018.06.3063. PMID: 30243772.
- Hebert LE, Weuve J, Scherr PA, et al. Alzheimer disease in the United States (2010-2050) estimated using the 2010 census. Neurology. 2013 May 7;80(19):1778-83. doi: 10.1212/WNL.0b013e31828726f5. PMID: 23390181.
- Belle SH, Burgio L, Burns R, et al. Enhancing the quality of life of dementia caregivers from different ethnic or racial groups: a randomized, controlled trial.[Summary for patients in Ann Intern Med. 2006 Nov 21;145(10):I39; PMID: 17116914]. Annals of Internal Medicine. 2006 Nov 21;145(10):727-38. PMID: 17116917.
- Berwig M, Heinrich S, Spahlholz J, et al. Individualized support for informal caregivers of people with dementia - effectiveness of the German adaptation of REACH II. BMC Geriatrics. 2017 Dec 12;17(1):286. doi: https://dx.doi.org/10.1186/s12877-017-0678-y. PMID: 29233097.
- Luchsinger JA, Burgio L, Mittelman M, et al. Comparative Effectiveness of 2 Interventions for Hispanic Caregivers of Persons with Dementia. Journal of the American Geriatrics Society. 2018 Sep;66(9):1708-15. doi: https://dx.doi.org/10.1111/jgs.15450. PMID: 30084133.
- Possin KL, Merrilees JJ, Dulaney S, et al. Effect of Collaborative Dementia Care via Telephone and Internet on Quality of Life, Caregiver Well-being, and Health Care Use: The Care Ecosystem Randomized Clinical Trial. JAMA Internal Medicine. 2019 Sep 30;30:30. doi: https://dx.doi.org/10.1001/jamainternmed.2019.4101. PMID: 31566651.
- Chodosh J, Colaiaco BA, Connor KI, et al. Dementia Care Management in an Underserved Community: The Comparative Effectiveness of Two Different Approaches. Journal of Aging & Health. 2015 Aug;27(5):864-93. doi: https://dx.doi.org/10.1177/0898264315569454. PMID: 25656074.
- Vickrey BG, Mittman BS, Connor KI, et al. The effect of a disease management intervention on quality and outcomes of dementia care: a randomized, controlled trial.[Summary for patients in Ann Intern Med. 2006 Nov 21;145(10):I31; PMID: 17116913]. Annals of Internal Medicine. 2006 Nov 21;145(10):713-26. PMID: 17116916.
- Thyrian JR, Hertel J, Wucherer D, et al. Effectiveness and Safety of Dementia Care Management in Primary Care: A Randomized Clinical Trial. JAMA Psychiatry. 2017 Oct 01;74(10):996-1004. doi: https://dx.doi.org/10.1001/jamapsychiatry.2017.2124. PMID: 28746708.