ADDRESSING SOCIAL ISOLATION

Thomas Cudjoe (Robert and Jane Meyerhoff endowed Professor and Assistant Professor of Medicine and Nursing, Johns Hopkins University) began his presentation by sharing personal anecdotes related to issues of social isolation and engagement among older adults. He emphasized that environment and culture are important factors in people's trajectories and also that there are mutual benefits for older adults and those who are supporting them. For example, he said, when he was a medical student he rented a room from an older woman: he gained shelter, and she gained financial support, a sense of safety, ongoing social connection, and new experiences. Carefully balancing the autonomy and safety of older people and those around them is a critical piece of the discussion on social isolation and aging in place. For instance, when providing home-based primary care for a homebound older couple “aging in place,” he witnessed many safety challenges (e.g., leaving the stove on during the winter for warmth) and hoarding issues, which also negatively affected the couple's neighbors.

Before exploring these issues in greater depth, Cudjoe provided a definition of social isolation: objectively having few social relationships, social roles, and group memberships, and having infrequent social interactions (Badcock et al., 2022; Holt-Lunstad, 2018). He emphasized that social isolation differs from loneliness, which is a subjective feeling of isolation. Social isolation includes low levels of social contact, may be chosen owing to a preference for solitude, and is not necessarily unpleasant; loneliness emerges from a mismatch between actual and desired relationships and may result in feelings of emotional distress and a low sense of control or choice. He explained that a framework for social connection considers the structure (e.g., marital status, social networks, social integration, living alone, social isolation), function (e.g., received support, perceived social support, perceived loneliness), and quality (e.g., marital quality, relationship strain, social inclusion or exclusion) of social connections and how these elements affect individuals who are living with cognitive impairment (Holt-Lunstad, 2018).

Cudjoe stressed that additional research is needed to better support older people who may be experiencing social isolation or loneliness. For example, a report recommended developing a more robust evidence base, as well as translating research into health care practice, improving awareness, strengthening education and training, and optimizing ties between the health system and community networks and resources (National Academies of Sciences, Engineering, and Medicine, 2020). More recently, the U.S. Surgeon General (2023) released an advisory on the healing effects of social connection and community; it also recommended the need to deepen knowledge so as to better address social isolation and loneliness.

Careful measurement and identification of individuals who experience social isolation and loneliness is a critical first step, Cudjoe said (for example, see Pomeroy, Mehrabi, et al., 2023). He underscored that social isolation is widespread among older adults: 25% of those aged 65 and older experience social isolation (Cudjoe et al., 2020). He also noted that experiences of social isolation may change over time (Cudjoe, 2023). Social isolation also matters at the cellular level, indicated by an elevation of inflammatory markers (Cudjoe et al., 2021).

Cudjoe pointed out that research over several decades has considered how upstream factors affect cognition (for example, see Berkman et al., 2000). More recently, scholars have begun to study the relationship between social connections and health conditions that could lead to hospitalizations and living in a nursing home (Pomeroy, Cudjoe et al., 2023). Many studies (Evans et al., 2019; National Academies, 2020) have found that social isolation increases risk for dementia in particular: for example, Huang et al. (2023) found that socially isolated older adults have a 27% higher chance of developing dementia than other older adults over time.

However, challenging the workshop session's framing, Cudjoe indicated that few studies have examined whether older people living with dementia are at greater risk than other older adults for social isolation. He argued that not enough data from people living with dementia are available to make that assertion, and more research is needed to understand the social connections of people living with dementia. He said it is even possible that dementia could increase social connection as a result of the increased use of clinical and community-based services and family and community supports. Of course, dementia could also decrease social connection because of memory challenges related to maintaining connection; the compounding risk of poor health and function; stigma, stress, fear, anxiety, and misunderstanding; and communication issues. He underscored that research to understand this relationship is complex because of the heterogeneity of individual and community resources, challenges in diagnosis and care of people living with dementia, and the evolution in individuals' cognition over time. He said that additional work to understand the bidirectional relationship between social isolation and health conditions would be valuable.

Reflecting on one of the session's guiding questions about how people living with dementia are vulnerable to the risks of isolation, Cudjoe noted that people living with dementia may live alone and experience safety and memory challenges; have trouble performing activities of daily living; experience difficulty taking medications and going to medical appointments; struggle to obtain adequate food, water, shelter, and clothing; or be victims of elder mistreatment. Responding to the question about factors that improve or worsen the consequences of isolation for people with dementia, he said that individual-level factors include each person's life course, demographic factors, personality, and preferences; and community-level factors include housing arrangements, transportation, networks of support, and social infrastructure (e.g., senior centers and cafes). Finally, considering the question about what interventions are available to support those without available or willing family and friends, he highlighted that intervention development could be strengthened with more measures of social isolation; identification of those in need before the intervention; consideration of upstream factors when determining who to target and when; use of a menu of solutions with a person–place focus to meet individual needs; and group-based and purpose-oriented strategies.

In concluding his presentation, Cudjoe presented four approaches to addressing social isolation among older people living with dementia: (1) use the Educate, Assess, Respond framework (Holt-Lunstad & Perissinotto, 2023); (2) create easy, adaptable, and sustainable solutions; (3) start proximally; and (4) adopt the sawubona¹ (“I see you”) mentality. He emphasized the need to balance personal, clinical, and population perspectives; expand awareness; improve measurement; invest in research and intervention development; center equity and justice at the core of efforts; and realize assets instead of focusing solely on the deficits of people with dementia.

TECHNOLOGY TO SUPPORT SOCIAL ENGAGEMENT

Raksha Mudar (Professor of Speech and Hearing Science, University of Illinois Urbana-Champaign) emphasized that although loneliness and social isolation—two key metrics of social health—are moderately correlated, they are not the same. Reiterating the point of the previous speaker, she stressed that people who are socially isolated might not be lonely, and people who are lonely might not be socially isolated. She noted that physical, cognitive, and emotional health are affected by both social isolation and loneliness, which may alter people's quality of life.

Mudar underscored that older adults are at an increased risk of both social isolation and loneliness and their negative consequences. She explained that data from general population–based studies show that social isolation and loneliness increase the risk of developing cognitive impairment. However, because the relationship between social health and cognitive health in populations known to be at risk of dementia is not well understood, key research questions remain: Are social isolation or loneliness related to accelerated cognitive decline in populations known to be at risk of dementia (i.e., those with mild cognitive impairment and older caregivers of people with dementia)? How do other modifiable risk factors (e.g., hearing or vision loss) mediate or moderate the relationship between social health and cognitive health in at-risk populations? She said that understanding these issues could lead to the development of more targeted approaches to delay the progression of cognitive impairment and allow people to be better able to age in place.

Mudar indicated that social engagement—that is, participation in social activities and maintenance of social connections—is a key target to address isolation and loneliness in older adults. Social engagement includes both structural elements (e.g., network size) and functional elements (e.g., purpose and belonging). To better inform assessments of social engagement and guide intervention design, she advocated for developing more integrated frameworks of social engagement. For example, Lydon et al. (2022) developed a multidimensional framework that represents social activity and the social network as two interrelated dimensions that include the structural and functional components. This framework also represents the role of contextual and health factors in social engagement: see Figure 6-1.

Figure 6-1

A framework of social engagement. SOURCE: Lydon et al. (2022, Figure 1). Reprinted with permission.

Mudar explored the use of technology-based interventions to offer social engagement opportunities for older adults in the comfort of their home environments. She noted that video technology platforms, in particular, enable relatively natural opportunities for social engagement. However, accommodating the unique technology requirements and socialization needs of older adults is critical in the success of such interventions. Nie et al. (2020) published design guidance for using video technology platforms to support social engagement interventions for older adults,² which is based on the team's redesign of the OneClick platform to optimize use for older adults with and without mild cognitive impairment.

Mudar elaborated on the research of her team. They conducted a randomized controlled trial of the efficacy of a social engagement intervention delivered using the optimized OneClick technology platform. The goal of the intervention was to provide opportunities for older adults to connect with others, attend to social needs, reflect with others casually, and engage with others on topics of shared interest. To achieve these outcomes, she explained that collaboration with content development experts was key to ensure that the content was curated carefully for the participants: The content needed to be engaging, lead to casual conversation, lead to positive reminiscing, be accessible, capture diversity, and cater to a wide audience. Sixty different topics were developed across five broad content areas: arts and culture; nature, health, and wellness; life experiences; science and technology; and recreation and sports. The structure of the events for social interaction also had to be carefully curated to create a shared framework for conversation for the participants. After instructions from the host of the event, participants engaged in conversation in breakout rooms, with discussion prompts available as needed.

As part of the trial, 99 participants were initially divided into an intervention group and a waitlist group that began after 8 weeks. Before beginning the intervention, participants received extensive, customized technology training, and live technology support was available throughout the intervention. Even for such problems as a dropped internet connection, participants could call technology support for assistance, which motivated them to continue participating. She underscored that hiring and training the right type of technology support staff, with the ability to walk people through instructions over the phone clearly and patiently, is critical.

Mudar explained that participants were asked to attend two events per week, based on the topics of greatest interest to them and their preferred time frames. Preliminary results of the trial revealed that participants with and without mild cognitive impairment were able to use the technology platform successfully to engage with one another, enjoyed the variety of conversation topics offered, found pretraining to be helpful, did not view occasional technology issues as a barrier to participation, and expressed interest in continuing to use the platform after the study. Home- and community-based organizations also expressed interest in using the platform to support their programs. Qualitative and quantitative analyses are ongoing to explore the efficacy of the intervention to support social engagement and quality of life. She explained that the scope of the study is being expanded to other populations, including care partners of people with dementia, 61 of whom are participating in a social engagement intervention, delivered via Zoom, designed to suit their needs for respite.

In closing, Mudar discussed the scalability issues of social engagement interventions delivered using technology. She emphasized that although one-on-one technology-based social engagement interventions target the individual needs of older adults, this approach is resource intensive. Group-based social engagement interventions using technology are more easily scalable and not as resource intensive, but they need to be designed carefully to meet individuals' varied needs, interests, and technology abilities. She pointed out that having flexible technology platforms is critical to support and scale any social engagement intervention. Developing training protocols to support the use of technology by users with varying proficiency is also essential. Beyond randomized controlled trials, she continued, pragmatic trials and adaptive intervention designs could be considered. Home- and community-based organizations could be used as test beds to optimize interventions before launching randomized controlled trials, but training protocols to support community-based implementation by home- and community-based organizations need to be developed. Finally, she emphasized the need to identify the minimum intervention dose that yields maximum benefits before increasing the reach of interventions.

SOCIAL ISOLATION AND SURROGATE DECISION MAKING

Andrew Cohen (Assistant Professor of Internal Medicine and Geriatrics, Yale School of Medicine, Yale University) defined decision-making capacity as the ability to understand the risks and benefits of a decision, reason through them, and communicate a decision about oneself. Health care professionals often speak to patients about situations in which their capacity to make decisions could be diminished as a result of a catastrophic illness, injury, or surgery, and a surrogate would be needed to make decisions on behalf of that person. This paradigm is also relevant for people with certain types of chronic disease. Many of these patients have full capacity to make and express decisions and participate in advanced care planning (preferences for which evolve over time as they adapt to their illness) until a serious illness arises close to the end of their lives, at which point a surrogate may be needed to make decisions.

In contrast, however, Cohen stressed that dementia creates a fundamentally different paradigm for decision-making capacity. The capacity to make decisions declines much more slowly over time as impairment increases, evolving from full capacity, to supported decision making, to a need for a surrogate to make decisions about housing and end-of-life decisions: see Figure 6-2.

Figure 6-2

Decision-making capacity trajectory for older adults with dementia. SOURCE: Workshop presentation, Andrew Cohen.

Cohen explained that this paradigm for decision-making capacity among older people with dementia implicitly assumes that the surrogate will be a family member. Many appointed health care proxies are family members, and most states have default surrogate statutes that allow family members to make decisions for an impaired person in the absence of an official health care proxy. However, using data from a nationally representative study of older adults without dementia, Cohen et al. (2020) showed that 8% of older Americans do not have a desired surrogate. Among that 8% of older Americans, 3% had no nuclear family or confidantes; 8% had no nuclear family but did have confidantes; and 89% had nuclear family and confidantes, but they were not desired as surrogates. Therefore, he emphasized that this paradigm does not fit for a significant number of older people.

Cohen and his team are currently conducting qualitative interviews with older adults who do not have anyone in their lives whom they would select as a health care surrogate. Some participants expressed interest in completing an inverse health care proxy form, if it was available, to list people who are not allowed to make decisions on their behalf. Questions remain about these “unrepresented adults” who may develop impaired decision-making capacity but have no suitable surrogate. He pointed out that this is especially challenging when incapacity is prolonged, such as for people living with dementia, and the types of decisions that need to be made are many and varied. In this case, the default mechanism is guardianship, in which a probate court appoints a paid professional to make decisions on behalf of the unrepresented adult. He commented that although this approach has not been studied in depth—in part because no centralized data exist about guardianship in most states—many people find it to be a problematic option.

To study people under guardianship, Cohen and his team used Department of Veterans Affairs data, which indicated that two-thirds of people under professional guardianship live in nursing homes; fewer than half of these people have no nuclear family; professional guardians receive little training in how to make decisions for people with dementia; and, contrary to popular belief, professional guardians' medical decisions are similar to those made by family surrogates for other patients. In another study of people with advanced dementia at the end of life, Cohen et al. (2021) confirmed that professional guardianship decisions are overall about the same as decisions made by family members.

Cohen also challenged implied assumptions about surrogates, which are not only that they will be family members, but also that they know how to make decisions on someone else's behalf. This assumption about decision-making capability is especially problematic when the surrogate is an appointed professional who has no previous relationship with the person, although family members making decisions for their loved ones also have an incredibly difficult task. Surrogate decision making for people with dementia in particular is extremely difficult, especially around questions about aging in place, because the person's capacity to express preferences may diminish long before such critical decisions need to be made.

Moving forward, Cohen said, the following questions require additional exploration to address these complex issues of decision making: How and when can and should people who do not have a desired surrogate be identified? What can one do when someone does not have a desired surrogate before that person's decision-making capacity becomes impaired? How can professional guardians and other surrogates be trained to make good decisions?

DISCUSSION

Reflecting on key themes from this session of the workshop, Vega observed that understanding the vital characteristics of a support system is critical. He reiterated that social isolation may occur even among older people with caregivers, especially among those who experience aggression and behavior issues as their dementia progresses. He said that social isolation is a complex problem, owing in part to cultural and linguistic issues that could enable or hinder engagement, as well as to family dynamics that change over time. Questions remain about which interventions are best suited to which individuals, how to assess these interventions systematically, and how to select the right people for interventions and tailor their training appropriately. Additional questions remain about legal decision making, especially as it relates to the social distance of and social power in families, which could also influence the success of interventions.

With consideration for declining civic participation in the United States and the general individualistic orientation of U.S. culture, Lis Nielsen (National Institute on Aging [NIA]) asked if particular characteristics of societies or communities make them more or less likely to embrace the task of reducing social isolation among individuals living with dementia. Without attempting to make any society-level generalizations, Cudjoe cautioned that the increased political polarization in the United States could influence whether and how people engage; more research is needed to examine the broader effect of the current polarization. He highlighted several specific promising efforts to reduce social isolation, including work by the Robert Wood Johnson Foundation and activity in the Blue Zones,³ which promote meaningful interaction.

Elena Fazio (Director of the Office of Alzheimer's Disease and Alzheimer's Disease-Related Dementias Strategic Coordination, Division of Behavioral and Social Research, NIA) invited the speakers to expand on their discussions of how people living with dementia are vulnerable to risks of isolation. Mudar replied that as these people try to understand what is happening to them, they may withdraw from their connections. For example, people beginning to experience mild cognitive impairment may avoid leaving the house and seeing other people to prevent uncomfortable situations (e.g., not remembering people's names). Cudjoe added that community-based organizations are doing good work to support the social connections of people living with dementia, but this work needs to be examined and disseminated more broadly so that more people can benefit. However, he cautioned against “overmedicalizing” the social connection aspect of aging. Cohen noted that people aging with dementia who do not have supportive family structures are particularly susceptible to social isolation and its negative consequences even though they may not be identified as such.

Fazio asked Cohen to elaborate on the notion of “unrepresented patients” across the range of dementia. Cohen responded that a challenge persists in the inconsistency of the terminology itself. For example, some use the term “elder orphans” to describe those who have the ability to make their own decisions but do not have a surrogate in the event that that ability is lost. However, these people are not “orphans”; they have friends who may not be the right people to serve as surrogates. He explained that people in the early stages of dementia might not be considered “unrepresented” because they are still able to make decisions; however, these people are still vulnerable to becoming unrepresented as their dementia progresses. Ideally, he continued, unrepresented adults would be identified upstream to prevent some of the issues that arise when cognitive impairment worsens. Because this level of identification is as difficult to achieve as is developing next steps to plan for this person's future, he said that further research is warranted.

Fazio expressed interest in learning more about issues specific to people with or without dementia who are aging in place alone. Cudjoe encouraged workshop participants to consult relevant research of Elena Portacolone at the University of California, San Francisco (for example, see Portacolone, 2020).

Fazio posed additional questions about technology interventions for those with cognitive impairment. She inquired about the cognitive ability of the participants in Mudar's technology-based social engagement intervention. Mudar explained that the participants had early stages of cognitive impairment. The only requirements to participate were that they had a computer or tablet that they could use with some assistance, a webcam, a quiet place, and a stable internet connection. Fazio suggested that additional research on tailoring interventions to those with more advanced stages of cognitive impairment or physical comorbidities could be helpful. Mudar agreed and highlighted another area for further exploration: information on how rural communities address issues of social isolation, and thus may face barriers in the implementation of interventions. For example, individuals in rural communities might not have access to broadband required to participate in a technology-based intervention for social engagement.

Emily Agree (Research Professor and Associate Director, Hopkins Population Center, Johns Hopkins University; planning committee chair) observed that as relatively younger cohorts age, people will have more knowledge about technology and more comfort with virtual settings. Therefore, she asked how to build sustainability into technology-based interventions for social engagement. For example, she wondered if people would be given the tools and the skills to continue communicating on their own, or if they would only be able to communicate through a specific platform. Cudjoe mentioned research on how people's use of technology affects their risk of social isolation. To sustain interventions—technology-based or otherwise—he advocated for the development of “capable” interventions that meet individuals where they are and meet their unique needs; purpose-driven, goal-oriented interventions to increase social connection are critically important.

REFERENCES

• Aranda MP, Baier R, Hinton L, Peak KD, Jackson JD, Dilworth-Anderson P, Gitlin LN, Jutkowitz E, Quiñones AR. Journal of the American Geriatrics Society. 2023. Preparing for pragmatic trials in dementia care: Health equity considerations for nonpharmacological interventions. https://doi​.org/10.1111/jgs.18568 . [PMC free article: PMC10841288] [PubMed: 37656062]
• Badcock JC, Holt-Lunstad J, Bombaci P, Garcia E, Lim MH. Position statement: Addressing social isolation and loneliness and the power of human connection. Global Initiative on Loneliness and Connection; 2022.
• Berkman LF, Glass T, Brissette I, Seeman TE. From social integration to health: Durkheim in the new millennium. Social Science and Medicine. 2000;51(6):843–857. [PubMed: 10972429]
• Cohen AB, Costello DM, O'Leary JR, Fried TR. Older adults without desired surrogates in a nationally representative sample. Journal of the American Geriatrics Society. 2020;69:114–121. [PMC free article: PMC7854949] [PubMed: 32898285]
• Cohen AB, Han L, O'Leary JR, Fried TR. Guardianship and end-of-life care for veterans with dementia in nursing homes. Journal of the American Geriatrics Society. 2021;69(2):342–348. [PMC free article: PMC7902349] [PubMed: 33170957]
• Cudjoe TKM. Correlates of loneliness and social isolation in old age: Poverty. In: Hajek A, Riedel-Heller SG, König H-H, editors. Loneliness and social isolation in old age. Routledge; 2023.
• Cudjoe TKM, Roth DL, Szanton SL, Wolff JL, Boyd CM, Thorpe RJ Jr. The epidemiology of social isolation: National Health & Aging Trends Study. Journal of Gerontology Series B: Psychological Sciences and Social Science. 2020;75(1):107–113. [PMC free article: PMC7179802] [PubMed: 29590462]
• Cudjoe TKM, Selvakumar S, Chung S, Latkin CA, Roth DL, Thorpe RJ Jr, Boyd CM. Getting under the skin: Social isolation and biological markers in the National Health and Aging Trends Study. Journal of the American Geriatrics Society. 2021;70(2):408–414. https://doi​.org/10.1111/jgs.17518 . [PMC free article: PMC8821117] [PubMed: 34698366]
• Dodge HH, Zhu J, Mattek N, Bowman M, Ybarra O, Wild K, Loewenstein DA, Kaye JA. Web-enabled conversational interactions as a means to improve cognitive functions: Results of a 6-week randomized controlled trial. Alzheimer's & Dementia. 2015;1(1):1–12. https://doi​.org/10.1016/j​.trci.2015.01.001 . [PMC free article: PMC4507295] [PubMed: 26203461]
• Evans CJ, Ison L, Ellis-Smith C, Nicholson C, Costa A, Oluyase AO, Namisango E, Bone AE, Brighton LJ, Yi D, Combes S, Bajwah S, Gao W, Harding R, Ong P, Higginson IJ, Maddocks M. Service delivery models to maximize quality of life for older people at the end of life: A rapid review. Milbank Quarterly. 2019;97(1):113–175. https://doi​.org/10.1111/1468-0009.12373 . [PMC free article: PMC6422603] [PubMed: 30883956]
• Holt-Lunstad J. Why social relationships are important for physical health: A systems approach to understanding and modifying risk and protection. Annual Review of Psychology. 2018;69:437–458. [PubMed: 29035688]
• Holt-Lunstad J, Perissinotto C. Social isolation and loneliness as medical issues. New England Journal of Medicine. 2023;388:193–195. [PubMed: 36648080]
• Huang AR, Roth DL, Cidav T, Chung S-E, Amjad H, Thorpe RJ Jr, Boyd CM, Cudjoe TKM. Social isolation and 9-year dementia risk in community dwelling Medicare beneficiaries in the United States. Journal of the American Geriatrics Society. 2023;71(3):765–773. [PMC free article: PMC10023331] [PubMed: 36628523]
• Lydon EA, Nguyen LT, Nie Q, Rogers WA, Mudar RA. An integrative framework to guide social engagement interventions and technology design for persons with mild cognitive impairment. Frontiers in Public Health. 2022;9 https://doi​.org/10.3389/fpubh​.2021.750340 . [PMC free article: PMC8795670] [PubMed: 35096730]
• National Academies of Sciences, Engineering, and Medicine. Social isolation and loneliness in older adults. The National Academies Press; 2020. [PubMed: 32510896]
• Nie Q, Nguyen LT, Myers D, Gibson A, Kerssens C, Mudar RA, Rogers WA. Design guidance for video chat system to support social engagement for older adults with and without mild cognitive impairment. Gerontechnology. 2020;20(1):1–15. https://doi​.org/10.4017/gt​.2020.20.1.398.08 . [PMC free article: PMC10615365] [PubMed: 37904899]
• Pomeroy ML, Cudjoe TKM, Cuellar AE, Ihara ES, Ornstein KA, Bollens-Lund E, Kotwal AA, Gimm GW. Association of social isolation with hospitalization and nursing home entry among community-dwelling older adults. JAMA Internal Medicine. 2023;183(9):955–962. [PMC free article: PMC10366946] [PubMed: 37486647]
• Pomeroy ML, Mehrabi F, Jenkins E, O'Sullivan R, Lubben J, Cudjoe TKM. Reflections on measures of social isolation among older adults. Nature Aging. 2023;3(12):1463–1464. [PubMed: 37640906]
• Portacolone E. Older Americans living alone: The influence of resources and intergenerational integration on inequality. Journal of Contemporary Ethnography. 2020;44(3):280–305.
• U.S. Surgeon General. Our epidemic of loneliness and isolation: The U.S. Surgeon General's Advisory on the healing effects of social connection and community. 2023. https://www​.hhs.gov/sites​/default/files/surgeon-general-social-connection-advisory.pdf . [PubMed: 37792968]
• Yu K, Wild K, Potempa K, Hampstead BM, Lichtenberg PA, Struble LM, Pruitt P, Alfaro EL, Lindsley J, MacDonald M, Kaye JA, Silbert LC, Dodge HH. The Internet-Based Conversational Engagement Clinical Trial (I-CONECT) in socially isolated adults 75+ years old: Randomized controlled trial protocol and COVID-19 related study modifications. Frontiers in Digital Health. 2021;3:714813. https://doi​.org/10.3389/fdgth​.2021.714813 . [PMC free article: PMC8521795] [PubMed: 34713183]