Science into Practice: Effective Solutions for Social Isolation and Loneliness

Abstract:

Research confirms the adverse health effects of social isolation and loneliness, with increasing focus on finding effective solutions. This article describes themes that can inform program and intervention design: multiple underlying causal factors may require a tailored approach; the time course of relationships and disease development versus length of interventions; the protective effects of relationships based on naturally occurring or existing close relationships; not all relationships are positive, and negativity in relationships is associated with greater risk. If ignored, these themes could lead to reduced effectiveness or harm. The article also outlines key components to evaluating interventions.

Key Words:

social isolation, loneliness, solutions, program design, interventions, relationships

Humans are inherently social, and social relationships influence all aspects of life, including development, happiness, and health. Beyond the obvious emotional and psychological consequences, social isolation and loneliness independently predict premature mortality (Holt-Lunstad et al., 2015), significant morbidity (Valtorta et al., 2016), and increased healthcare costs (Flowers, Houser, and Noel-Miller, 2017). Alarmingly, rates of social disconnection have been increasing, with approximately one third of older adults being socially isolated, lonely, or both (Flowers, Houser, and Noel-Miller, 2017). Preliminary data (SocialPro, 2020) from the time of widespread COVID-19 restrictions showed that younger adults were resilient, with little change in reported loneliness, but that older adults were more vulnerable, and showed increases (Luchetti, 2020). The adverse consequences of social disconnection will only increase unless effective solutions can be identified and promoted.

Despite robust evidence of the deleterious health effects of social isolation and loneliness, evidence of effective solutions is more limited (National Academies of Sciences, 2020). Several large-scale reviews of intervention literature have pointed to mixed and poor-quality evidence (Fakoya, McCorry, and Donnelly, 2020; Gardiner, Geldenhuys, and Gott, 2018; Poscia et al., 2018). Given this disconnection, an essential question is what evidence can we extract from epidemiology and relationship science, which established the importance of social connection as a robust predictor of health, to assist in designing more effective solutions?

Factors Relevant to Designing Effective Social Interventions

Defining the issue Current research uses a variety of definitions and measures to evaluate social connections, with a focus on the indicators of social deficits, namely social isolation and loneliness. While social isolation and loneliness are related terms, they are distinct concepts. Recognizing the distinction may have important implications for how to effectively intervene, thus clear definitions are needed. Social isolation is often defined as objectively being alone or having few or infrequent social contacts; whereas loneliness is a subjective feeling, often defined as the discrepancy between one’s actual and desired level of connection (Peplau and Perlman, 1982).

Although social isolation and loneliness are distinct, intervention efforts often target both. This may be appropriate in some cases but reducing social isolation may not necessarily result in changes in loneliness and vice versa. It is important for interventions to clearly identify which condition is being targeted. In this article specific terms will be used except when the concept applies to both social isolation and loneliness (SIL), in which case the acronym SIL will be used.

Outcomes The mixed evidence in the social intervention literature may be due in part to the wide variety of outcomes assessed across studies. For example, some interventions evaluate decreases in loneliness, others assess increases in social behaviors (e.g., social participation), while others consider issues such as feasibility or acceptability. To determine the effectiveness of an intervention, care should be taken as to the outcome(s) the intervention was designed to reduce. Moreover, measuring uptake or satisfaction is insufficient (e.g., people widely use and/or enjoy practices that are not necessarily health promoting), yet in some cases these are the only reported indicators of success. Interventions should clearly identify whether they are designed to reduce loneliness, social isolation, or both, and use validated outcome measures (National Academies of Sciences, 2020). Interventions also should include measurement of the intended health outcome (e.g., emotional/psychological, physical, or cognitive) and the potential mechanisms of change.

Guiding Theoretical Framework The fields of social psychology and health psychology are characterized by strong theoretical foundations, yet too often social interventions provide commonsense solutions without a guiding theoretical framework for reducing SIL. Moreover, social interventions need to consider multiple factors, because social connections promote positive health outcomes for a variety of reasons (see Holt-Lunstad, 2018, for review). Being socially connected provides safety and security, buffers stress responses, aids emotion regulation, improves health behaviors, and improves access to resources. Conversely, SIL represents the absence or inadequacy of these social resources.

Are interventions attempting to recreate the key missing factors that are protective or address the reasons why they are missing? For example, a review of interventions on loneliness broke them down by those attempting to improve social skills, social support, social contact, or maladaptive cognition (Masi et al., 2011). Those focused on social skills and maladaptive cognition concentrated more on potential barriers to social resources, whereas social support and social contact interventions concentrated more on access to resources. Both types of interventions often lack a guiding theoretical framework. Social contact could potentially tap into any of the protective effects listed earlier, but tests of effectiveness would differ substantially and could be implemented more efficiently if focused on certain aspects more than others. Regardless, a guiding theoretical framework can improve how an intervention is designed, implemented, and evaluated.

Targeting Causal Mechanisms To design effective interventions, it is important to recognize that SIL has no single cause. Interventions that fail to identify and address the underlying causes of an individual’s SIL may result in no change or worsening. Multiple risk factors for SIL have been identified in the literature. These include death of loved ones, worsening of health and chronic illness, new sensory impairments, retirement, and changes in income (National Academies of Sciences, 2020), with each of those conditions necessarily requiring distinct amelioration strategies. Given the diversity of risk factors, many scholars have argued that a one-size-fits-all approach is unlikely to succeed. The consensus committee report recommended that tailored approaches be implemented and evaluated (National Academies of Sciences, 2020).

Also, interventions to reduce SIL need to be designed with the mechanisms of health change in mind. Interventions are often developed to reduce SIL because of the well-established negative influences on mental, physical, and cognitive health outcomes. SIL has also been linked to biological (e.g., inflammation, cardiovascular reactivity) and behavioral (e.g., impaired sleep, physical inactivity) changes that lead to these outcomes; however, very few interventions determine whether the intervention successfully modifies any of the mechanisms. Several biological mechanisms (e.g., cardiovascular, neuroendocrine, immune) have been implicated as pathways explaining the associations between SIL and health outcomes (Uchino, 2006). These biological pathways often interact with each other, and with behavioral and cognitive pathways. For example, loneliness has been linked to chronic inflammation (Nersesian et al., 2018), which is a common pathway implicated in mental (e.g., depression), physical (e.g., cardiovascular disease), and cognitive (e.g., Alzheimer’s disease) health outcomes—yet few interventions test whether any measurable change in inflammation may occur as a result of the intervention. Thus, to reduce the health burden caused by loneliness it is critical to interrupt or change the course of these disease pathways.

Time Course Another major consideration is the time course of relationships and disease development in the context of interventions. If intervening to reduce risk, one must account for the fact that meaningful social relationships develop over time. Delimited positive interactions may provide temporary boosts to emotional and biological markers, but they may not necessarily yield long-term benefits to mental or physical health unless the relationships are sufficiently trustworthy and enduring. Further, many physical health conditions decline over long periods of time prior to the development of fatal conditions. Scholars have strongly recommended a long-term perspective on the interactions between social and physical well-being (Uchino, 2009). However, when evaluating the effectiveness of interventions, the vast majority only assess short-term outcomes or do not follow participants long enough to detect long-term changes. Metrics that undergo rapid fluctuation (e.g., emotional affect, blood glucose levels) also tend to predict mental or physical health to a much lesser extent than do markers that change more slowly (e.g., depression, hA1c). 

The lack of long-term, follow-up data among SIL interventions is most likely attributable to limited resources; however, the consequence is inadequate information regarding long-term outcomes and sustainability. Finding the most cost-effective approaches is imperative to promoting long-term follow-up (National Academies of Sciences, 2020). Until SIL are elevated to the status of other public health concerns such as obesity and diabetes (which are clearly understood to require long-term interventions and follow-up), it is unlikely that we will find adequate solutions for SIL (Holt-Lunstad, 2018).

Creating and/or Improving Social Connections

Another potential challenge to existing interventions is that the protective effects of relationships are based on naturally occurring or existing close relationships, yet most intervention programs involve strangers (e.g., hired staff, other patients with the same medical condition). To clarify the different types of social interventions, the UK Campaign to End Loneliness classified interventions into three categories: those that use existing relationships, those that change thinking about relationships, and those that create new relationships. The epidemiological literature linking the protective effects of strong social connections with reduced risk for premature mortality is almost entirely based on the first category of naturally occurring relationships. Nevertheless, most interventions to foster social connectivity focus on the second and third categories, using personnel previously unknown to the individual receiving the intervention. This situation must be rectified.

Decades of research have consistently confirmed that familial relationships are the most enduring, intimate, and consequential of any social relationship (Goodfriend, 2020). They may also be among the most important relationships that can mitigate SIL. For example, those who transition from being lonely to unlonely were characterized by greater family support and lower family strain (Hawkley and  Kocherginsky, 2018), and improvements in their relationships (Victor and Bowling, 2012).

The effectiveness of familial relationships in reducing loneliness is influenced by the level of trust and responsiveness in the relationship. Relationships that are responsive convey understanding, validation, and care. “They are warm, sensitive to their partners’ feelings, and want to make their partners feel comfortable, valued, listened to, and understood” (Canevello and Crocker, 2010).

That responsiveness builds security and trust and provides a foundation for how individuals perceive and respond to circumstances that arise, including health challenges. When individuals feel secure about their social connections, they tend to be more resilient, less reactive and able to proactively cope. The resilience factors most beneficial to long-term physical health arise from accumulated evidence that one is supported by others (Uchino, 2006). If individuals’ physical condition deteriorates, they can have reasonable confidence that their close associates will provide for their needs. That kind of trust may not accrue from short-term relationships specific to a brief intervention.

Optimally, social interventions should align with the realities of individuals’ pre-existing social lives. Those who are at greatest risk may be those who do not have reliable social networks to draw upon, necessitating support groups of peers and trained (hired or volunteer) support personnel. Nonetheless, it is clear from the mixed effectiveness of these interventions that in many cases these sources may not be able to adequately approximate the functions or qualities of naturally occurring close relationships. Thus, care is needed to identify and more closely approximate protective relationship characteristics (e.g., responsiveness, trustworthiness, support).

Quality of Relationships Matters 

Another potential reason for such mixed findings among social interventions may be a lack of attention to relationship quality. Quality goes beyond the level of depth or superficiality of a relationship to address the positive and negative qualities of relationships and interactions. Not all relationships are positive, and negativity in relationships is associated with greater health risks (Birmingham and Holt-Lunstad, 2018; Holt-Lunstad and Uchino, 2019). Many social interventions seek to improve the frequency, duration, and number of social contacts among those experiencing SIL. However, research has consistently indicated that the beneficial effects of social relationships is attributable to consistent interactions of high quality (Hawkley, 2018; Cohen-Mansfield et al., 2016).

Social interventions are likely to be the most effective when developing, repairing, or maintaining high-quality interactions characterized by mutual sharing, actions demonstrating mutual benefits rather than mere verbal exchange, and commitments to one another despite inconveniences and common barriers (Hogan, 2002). The quality of close relationships strongly predicts multiple aspects of physical and mental health (Holt-Lunstad and Uchino, 2019). Therefore, strengthening the quality of existing family relationships, such as reducing discord or distress by improving coping skills within family units, or by improving caretakers’ abilities to maintain positive interactions with physically ill family members, may be more effective in the long-term than fostering temporary interactions with strangers.

Factors Relevant to Evaluating Interventions

Measurement Among reviews of social interventions, a frequently noted limitation is the variability in measurement tools used across studies. SIL are assessed using myriad constructs and measurement tools, which complicates the ability to draw strong conclusions about what works best, for whom, and in which circumstances (Veazie et al., 2019). The NASEM report calls for the need to use well-validated measurement, and to avoid creating and using unvalidated measures or altering existing validated measures (National Academies of Sciences, 2020).

Appropriate Comparison A randomized controlled trial is generally thought to be the gold standard when it comes to psychological and behavioral interventions (Gatlin and Czaja, 2015). However, most existing social interventions do not use this methodology or include an appropriate comparison group. Many studies only examine participants pre- and post-intervention and test for change; but such an approach does not rule out whether participants may have become more or less lonely over time regardless of the intervention.

In a “friendly visitor intervention” it is possible that measurable improvements in loneliness are unrelated to the intervention itself. Participants may have improved because the intervention was conducted in the springtime and they were able to get outside the home more often, or because of the increased attention by intervention staff. Similarly, it is possible that participants in a “friendly visitor” intervention show health declines over the intervention period, but the participants were all highly vulnerable older adults with pre-existing chronic health conditions and the intervention was effective in slowing the rate of this decline. In both cases the effectiveness (or ineffectiveness) of the intervention is not readily apparent until a similar group of participants who do not receive the “friendly visitor” can be used as a comparison.

Scalability and Sustainability Social interventions can be conducted in a variety of ways, including peer support groups, one-on-one meetings with hired staff, family or couple sessions, and telephone conversations. Increasingly, however, because of demands for scalability, many interventions are being conducted online via video connections. Similarly, in response to COVID-19 restrictions online or virtual methods are being scaled and disseminated widely to replace in-person programs. Many logistical limitations associated with in-person meetings (e.g., homebound individuals) can be diminished through online technology, but in-person meetings may be more effective in establishing and maintaining relationships of higher perceived quality. Further, more than a quarter of older adults do not have access to the Internet or broadband (Pew Research Center, 2019). Despite inconsistent and weak evidence across twelve reviews and twenty-two independent studies using e-interventions for loneliness in older people (Chipps, Jarvis, and Ramlall, 2017), there is still widespread use and push for greater access and utilization among older adults. Research has yet to confirm the relative benefits and limitations of social interventions provided at a distance. Caution should be exercised when scaling an intervention before it is deemed effective.

Given that research has most often focused on formalized/structured social interventions, informal/unstructured social interventions will need to be evaluated in future research as these may be more sustainable (the degree to which implementation is maintained over time). Some examples of informal interventions in apartment/retirement communities or rehabilitation centers include designating space for informal gatherings or meal preparations, adding unstructured social exchanges to the activities calendar, mixing tenants across age/ability levels, and seeking community volunteers for visiting. Options for informal gatherings multiply when considering online programs, such as shared interest groups, pro-social interactive gaming, and promotion of contacts with extended family or friends who live far away (e.g., providing Zoom or Facetime accounts along with access to a bedside device).

Ethical Considerations

Although social interactions are usually benign and associated with minimal risk of harm, all human interactions involve possible risks and ethical considerations. A recent review found that 67 percent of intelligent assistive technologies (e.g., handheld devices, smart home sensors, robots, mobility aids) to be used among older adults with dementia were developed without any consideration of ethical implications (Ienca, 2018). Any social intervention, formal or informal, should be designed and monitored with attention to issues of safety, accessibility, privacy, and autonomy (National Academies of Sciences, 2020).

Interventions should be accessible to a range of physical and cognitive abilities, potential sensory impairments, economic backgrounds (if intervention involves costly equipment or monthly expenses), and literacy levels. Interventions, particularly those that enlist consumer-facing technologies that collect user (e.g., usage, audio, visual) data, must meet HIPPA standards. Privacy concerns also must be considered when data is shared with family members. Interventions should take care to maintain the autonomy of older adults, such that older adults’ values, goals, and preferences are preserved. These ethical considerations point to precautions and ongoing vigilance against possible manipulation, fraud, misconduct, and abuse.

Conclusion

Although it is clear that human beings thrive when supported in strong social networks, health sciences have historically overlooked the intersection of physical health with interpersonal well-being. Increasingly, programs and interventions are being developed in medical clinics and hospitals, rehabilitation centers, and retirement communities to address individuals’ social needs. However, these interventions tend to be delimited in nature, conducted by staff, and lacking in theoretical grounding. It is essential to identify and target the needs of individuals. In that way, the field can switch from haphazard generic approaches (e.g., bringing strangers together with intentions that they form bonds), to approaches that are evidence-based and informed by social psychology and relationship science, and that account for life circumstances. People experience SIL for many reasons. Generic intervention approaches to reduce SIL will not be as helpful as those designed to be responsive to specific needs and to foster long-term connectivity. 

Julianne Holt-Lunstad, Ph.D., is a professor in the Psychology Department at Brigham Young University (BYU) in Provo, Utah. Ryan Layton and Bonnie Barton are students at BYU, and Timothy B. Smith, Ph.D., is a professor in Counseling Psychology and Special Education at BYU.

References

Birmingham, W. C., and Holt-Lunstad, J. 2018. "Social Aggravation: Understanding the Complex Role of Social Relationships on Stress and Health-relevant Physiology." International Journal of Psychophysiology 131: 13–23. doi:10.1016/j.ijpsycho.2018.03.023.

Canevello, A., and Crocker, J. 2010. "Creating Good Relationships: Responsiveness, Relationship Quality, and Interpersonal Goals." Journal of Personality and Social Psychology 99(1): 78–106. doi.org/10.1037/a0018186.

Chipps, J., Jarvis, M. A., and Ramlall, S. 2017. "The Effectiveness of E-Interventions on Reducing Social Isolation in Older Persons: A Systematic Review of Systematic Reviews." Journal of Telemedicine and Telecare 23(10): 817–27.

Cohen-Mansfield, J. H., et al. 2016. "Correlates and Predictors of Loneliness in Older-adults: A Review of Quantitative Results Informed by Qualitative Insights." International Psychogeriatrics 28(4): 557–76.

Fakoya, O. A., McCorry, N. K., and Donnelly, M. 2020. "Loneliness and Social Isolation Interventions for Older Adults: A Scoping Review of Reviews. BMC Public Health 20(1): 129. doi:10.1186/s12889-020-8251-6.

Flowers, L., Houser, A., and Noel-Miller, C. 2017. "Medicare Spends More on Socially Isolated Older Adults." AARP Public Policy Institute. Retrieved August 12, 2020.

Gardiner, C., Geldenhuys, G., and Gott, M. 2018. "Interventions to Reduce Social Isolation and Loneliness Among Older People: An Integrative Review." Health and Social Care in Community 26(2): 147–57. doi:10.1111/hsc.12367.

Gitlin, L. N., and Czaja, S. 2016. Behavioral Intervention Research: Designing, Evaluating, and Implementing. New York: Springer Publishing Company.

Goodfriend, W. 2020. Intimate Relationships. Thousand Oaks, CA: Sage.

Hawkley, L. C., and Kocherginsky, M. 2018. "Transitions in Loneliness Among Older Adults: A 5-Year Follow-Up in the National Social Life, Health, and Aging Project." Research on Aging 40(4): 365–87. doi:10.1177/0164027517698965.

Hogan, B. E., Linden, W., and Najarian, B. 2002. "Social Support Interventions: Do They Work?" Clinical Psychology Review 22(3): 383–442.

Holt-Lunstad, J. 2018. "Why Social Relationships Are Important for Physical Health: A Systems Approach to Understanding and Modifying Risk and Protection." Annual Review of Psychology 69: 437–58. doi:10.1146/annurev-psych-122216-011902.

Holt-Lunstad, J., et al. 2015. "Loneliness and Social Isolation as Risk Factors for Mortality: A Meta-analytic Review." Perspectives on Psychological Science 10(2): 227–37. doi:10.1177/1745691614568352.

Holt-Lunstad, J., and Uchino, B. N. 2019. "Social Ambivalence and Disease (SAD): A Theoretical Model Aimed at Understanding the Health Implications of Ambivalent Relationships." Perspectives on Psychological Science 14(6): 941–66. doi:10.1177/1745691619861392.

Ienca, M., T., et al. 2018. "Ethical Design of Intelligent Assistive Technologies for Dementia: A Descriptive Review." Science and Engineering Ethics 24(4): 1035–55.

Luchetti M., et al. 2020. "The Trajectory of Loneliness in Response to COVID-19." American Psychologist. doi: 10.1037/amp0000690.

Masi, C. M., et al. 2011. "A Meta-analysis of Interventions to Reduce Loneliness." Personality and Social Psychological Review 15(3): 219–66. doi:10.1177/1088868310377394.

The National Academies of Sciences, Engineering, and Medicine. 2020. Social Isolation and Loneliness in Older Adults: Opportunities for the Health Care System. Retrieved August 12, 2020.

Nersesian, P. V., et al. 2018. "Loneliness in Middle Age and Biomarkers of Systemic Inflammation: Findings from Midlife in the United States." Social Science & Medicine 209: 174–81. doi:10.1016/j.socscimed.2018.04.007.

Peplau, L. A., and Perlman, D. 1982. "Perspective on Loneliness." In L. A. Peplau and D. Perlman, Eds., Loneliness: A Sourcebook of Current Theory, Research and Therapy. New York: John Wiley and Sons.

Pew Research Center. 2019. "Internet/Broadband Fact Sheet." Retrieved September 2, 2020.

Poscia, A., et al. 2018. "Interventions Targeting Loneliness and Social Isolation Among the Older People: An Update Systematic Review." Experimental Gerontology 102: 133–44. doi:10.1016/j.exger.2017.11.017.

SocialPro. 2020. "Loneliness and Anxiety During Lockdown." Retrieved September 2, 2020.

Uchino, B. N. 2006. "Social Support and Health: A Review of Physiological Processes Potentially Underlying Links to Disease Outcomes." Journal of Behavioral Medicine 29(4): 377–87. doi:10.1007/s10865-006-9056-5.

Uchino, B. N. 2009. "What a Lifespan Approach Might Tell Us about Why Distinct Measures of Social Support have Differential Links to Physical Health." Journal of Social and Personal Relationships 26(1): 53–62. doi:10.1177/0265407509105521.

Valtorta, N. K., et al. 2016. "Loneliness and Social Isolation as Risk Factors for Coronary Heart Disease and Stroke: Systematic Review and Meta-analysis of Longitudinal Observational Studies." Heart 102(13): 1009–16. doi:10.1136/heartjnl-2015-308790.

Veazie, S., J. et al. 2019. Addressing Social Isolation to Improve the Health of Older Adults: A Rapid Review. Rockville, MD: Agency for Healthcare Research and Quality.

Victor, C. R., and Bowling, A. 2012. "A Longitudinal Analysis of Loneliness Among Older People in Great Britain." The Journal of Psychology 146(3): 313–31. doi:10.1080/00223980.2011.609572.