Do Words Matter? Detecting Social Isolation and Loneliness in Older Adults Using Natural Language Processing

Introduction: Social isolation and loneliness (SI/L) are growing problems with serious health implications for older adults, especially in light of the COVID-19 pandemic. We examined transcripts from semi-structured interviews with 97 older adults (mean age 83 years) to identify linguistic features of SI/L.Methods: Natural Language Processing (NLP) methods were used to identify relevant interview segments (responses to specific questions), extract the type and number of social contacts and linguistic features such as sentiment, parts-of-speech, and syntactic complexity. We examined: (1) associations of NLP-derived assessments of social relationships and linguistic features with validated self-report assessments of social support and loneliness; and (2) important linguistic features for detecting individuals with higher level of SI/L by using machine learning (ML) models.Results: NLP-derived assessments of social relationships were associated with self-reported assessments of social support and loneliness, though these associations were stronger in women than in men. Usage of first-person plural pronouns was negatively associated with loneliness in women and positively associated with emotional support in men. ML analysis using leave-one-out methodology showed good performance (F1 = 0.73, AUC = 0.75, specificity = 0.76, and sensitivity = 0.69) of the binary classification models in detecting individuals with higher level of SI/L. Comparable performance were also observed when classifying social and emotional support measures. Using ML models, we identified several linguistic features (including use of first-person plural pronouns, sentiment, sentence complexity, and sentence similarity) that most strongly predicted scores on scales for loneliness and social support.Discussion: Linguistic data can provide unique insights into SI/L among older adults beyond scale-based assessments, though there are consistent gender differences. Future research studies that incorporate diverse linguistic features as well as other behavioral data-streams may be better able to capture the complexity of social functioning in older adults and identification of target subpopulations for future interventions. Given the novelty, use of NLP should include prospective consideration of bias, fairness, accountability, and related ethical and social implications.

Download Full-text

The ‘loneliness pandemic’: implications for gerontological nursing

British Journal of Nursing ◽

10.12968/bjon.2021.30.11.652 ◽

2021 ◽

Vol 30 (11) ◽

pp. 652-655

Author(s):

Carlos Laranjeira

Keyword(s):

Mental Health ◽

Older Adults ◽

Older People ◽

Chronic Diseases ◽

Old Age ◽

Social Isolation ◽

Social Relationships ◽

Vulnerable Groups ◽

Negative Consequences ◽

Gerontological Nursing

The COVID-19 pandemic compelled states to limit free movement, in order to protect at-risk and more vulnerable groups, particularly older adults. Due to old age or debilitating chronic diseases, this group is also more vulnerable to loneliness (perceived discrepancy between actual and desired social relationships) and social isolation (feeling that one does not belong to society). This forced isolation has negative consequences for the health of older people, particularly their mental health. This is an especially challenging time for gerontological nursing, but it is also an opportunity for professionals to combat age stereotypes reinforced with COVID-19, to urge the measurement of loneliness and social isolation, and to rethink how to further adjust interventions in times of crisis, such as considering technology-mediated interventions in these uncertain times.

Download Full-text

How Relationship Satisfaction Influences Post-Pandemic COVID-19 Anxiety: A Four-Wave Study in China

10.31234/osf.io/vbwcg ◽

2021 ◽

Author(s):

Jingyi Ou ◽

yunhanqi ◽

Ke Zhang ◽

Yuexiao Du ◽

Yihang He ◽

...

Keyword(s):

Mental Health ◽

Social Support ◽

Psychological Distress ◽

Relationship Satisfaction ◽

Social Interactions ◽

Social Isolation ◽

Social Relationships ◽

Crucial Role ◽

Perceived Social Support ◽

The Social

The social isolation due to the COVID-19 pandemic exerts lasing impacts on people’s mental health. However, whether and how people’s pre-existing positive social relationships can serve as stable reserves to alleviate people psychological distress following the disaster remains unknown. To address the question, the current study examined whether pre-pandemic relationship satisfaction would predict post-pandemic COVID-19 anxiety through middle-pandemic perceived social support and/or gratitude using four-wave data in China (N = 222, 54.50% female, Mage = 31.53, SD = 8.17). Results showed that people’s COVID-19 anxiety decreased from the peak to the trough pandemic stage; perceived social support increased markedly from the pre-pandemic to the peak and remained stable afterwards, while relationship satisfaction remained unchanged throughout. Further, it was middle-pandemic perceived social support, but not gratitude, mediated the association between pre-pandemic relationship satisfaction and post-pandemic COVID-19 anxiety, indicating perceived social support played a more crucial role than gratitude in this process. Last, it is suggested to distinguish perceived social support from gratitude as two different components of social interactions.

Download Full-text

Social Support Patterns of Middle-Aged and Older Adults Within a Physical Activity App: Secondary Mixed Method Analysis

JMIR Aging ◽

10.2196/12496 ◽

2019 ◽

Vol 2 (2) ◽

pp. e12496 ◽

Cited By ~ 2

Author(s):

Zakkoyya H Lewis ◽

Maria C Swartz ◽

Eloisa Martinez ◽

Elizabeth J Lyons

Keyword(s):

Physical Activity ◽

Social Support ◽

Older Adults ◽

Social Interaction ◽

Emotional Support ◽

Mixed Method ◽

Middle Aged ◽

Promising Tool ◽

The Social ◽

Method Analysis

Background Physical activity (PA) is critical for maintaining independence and delaying mobility disability in aging adults. However, 27 to 44% of older adults in the United States are meeting the recommended PA level. Activity trackers are proving to be a promising tool to promote PA adherence through activity tracking and enhanced social interaction features. Although social support has been known to be an influential behavior change technique to promote PA, how middle-aged and older adults use the social interaction feature of mobile apps to provide virtual support to promote PA engagement remains mostly underexplored. Objective This study aimed to describe the social support patterns of middle-aged and older adults using a mobile app as part of a behavioral PA intervention. Methods Data from 35 participants (mean age 61.66 [SD 6] years) in a 12-week, home-based activity intervention were used for this secondary mixed method analysis. Participants were provided with a Jawbone Up24 activity monitor and an Apple iPad Mini installed with the UP app to facilitate self-monitoring and social interaction. All participants were given an anonymous account and encouraged to interact with other participants using the app. Social support features included comments and likes. Thematic coding was used to identify the type of social support provided within the UP app and characterize the levels of engagement from users. Participants were categorized as superusers or contributors, and passive participants were categorized as lurkers based on the literature. Results Over the 12-week intervention, participants provided a total of 3153 likes and 1759 comments. Most participants (n=25) were contributors, with 4 categorized as superusers and 6 categorized as lurkers. Comments were coded as emotional support, informational support, instrumental support, self-talk, and other, with emotional support being the most prevalent type. Conclusions Our cohort of middle-aged and older adults was willing to use the social network feature in an activity app to communicate with anonymous peers. Most of our participants were contributors. In addition, the social support provided through the activity app followed social support constructs. In sum, PA apps are a promising tool for delivering virtual social support to enhance PA engagement and have the potential to make a widespread impact on PA promotion. Trial Registration ClinicalTrials.gov NCT01869348; https://clinicaltrials.gov/ct2/show/NCT01869348

Download Full-text

Social Support Patterns of Middle-Aged and Older Adults Within a Physical Activity App: Secondary Mixed Method Analysis (Preprint)

10.2196/preprints.12496 ◽

2018 ◽

Author(s):

Zakkoyya H Lewis ◽

Maria C Swartz ◽

Eloisa Martinez ◽

Elizabeth J Lyons

Keyword(s):

Physical Activity ◽

Social Support ◽

Older Adults ◽

Social Interaction ◽

Emotional Support ◽

Mixed Method ◽

Middle Aged ◽

Promising Tool ◽

The Social ◽

Method Analysis

BACKGROUND Physical activity (PA) is critical for maintaining independence and delaying mobility disability in aging adults. However, 27 to 44% of older adults in the United States are meeting the recommended PA level. Activity trackers are proving to be a promising tool to promote PA adherence through activity tracking and enhanced social interaction features. Although social support has been known to be an influential behavior change technique to promote PA, how middle-aged and older adults use the social interaction feature of mobile apps to provide virtual support to promote PA engagement remains mostly underexplored. OBJECTIVE This study aimed to describe the social support patterns of middle-aged and older adults using a mobile app as part of a behavioral PA intervention. METHODS Data from 35 participants (mean age 61.66 [SD 6] years) in a 12-week, home-based activity intervention were used for this secondary mixed method analysis. Participants were provided with a Jawbone Up24 activity monitor and an Apple iPad Mini installed with the UP app to facilitate self-monitoring and social interaction. All participants were given an anonymous account and encouraged to interact with other participants using the app. Social support features included comments and likes. Thematic coding was used to identify the type of social support provided within the UP app and characterize the levels of engagement from users. Participants were categorized as superusers or contributors, and passive participants were categorized as lurkers based on the literature. RESULTS Over the 12-week intervention, participants provided a total of 3153 likes and 1759 comments. Most participants (n=25) were contributors, with 4 categorized as superusers and 6 categorized as lurkers. Comments were coded as emotional support, informational support, instrumental support, self-talk, and other, with emotional support being the most prevalent type. CONCLUSIONS Our cohort of middle-aged and older adults was willing to use the social network feature in an activity app to communicate with anonymous peers. Most of our participants were contributors. In addition, the social support provided through the activity app followed social support constructs. In sum, PA apps are a promising tool for delivering virtual social support to enhance PA engagement and have the potential to make a widespread impact on PA promotion. CLINICALTRIAL ClinicalTrials.gov NCT01869348; https://clinicaltrials.gov/ct2/show/NCT01869348

Download Full-text

Social network members who engage in activities with older adults: do they bring more social benefits than other members?

Ageing and Society ◽

10.1017/s0144686x17001490 ◽

2018 ◽

Vol 39 (5) ◽

pp. 1050-1069 ◽

Cited By ~ 1

Author(s):

SATO ASHIDA ◽

DANIEL K. SEWELL ◽

ELLEN J. SCHAFER ◽

AUDREY SCHROER ◽

JULIA FRIBERG

Keyword(s):

Social Support ◽

Older Adults ◽

Social Network ◽

Social Influence ◽

Social Relationships ◽

Social Engagement ◽

The United States ◽

Social Benefits ◽

Social Activities ◽

Generalised Linear Mixed Model

ABSTRACTActive participation in social activities is important for the wellbeing of older adults. This study explored benefits of active social engagement by evaluating whether relationships that comprise active involvement (e.g. co-engagement in activities) bring more social benefits (i.e. social support, companionship, positive social influence) than other relationships that do not involve co-engagement. A total of 133 adults ages 60 years and older living in a rural Midwestern city in the United States of America were interviewed once and provided information on 1,740 social network members. Among 1,506 social relationships in which interactions occurred at least once a month, 52 per cent involved engagement in social activities together and 35 per cent involved eating together regularly. Results of the generalised linear mixed model showed that relationships involving co-engagement were significantly more likely to also convey social support (i.e. emotional, instrumental, informational), companionship and social influence (encouragement for healthy behaviours) than relationships that do not involve co-engagement. Having more network members who provide companionship was associated with higher sense of environmental mastery, positive relations with others and satisfaction with social network. Interventions may focus on maintaining and developing such social relationships and ensuring the presence of social settings in which co-engagement can occur. Future research may explore whether increasing co-engagement leads to an enhanced sense of companionship and psychological wellbeing.

Download Full-text

SOCIAL ISOLATION AND FALLS RISK AMONG COMMUNITY DWELLING OLDER ADULTS: THE MEDIATING ROLE OF DEPRESSION

Innovation in Aging ◽

10.1093/geroni/igz038.1078 ◽

2019 ◽

Vol 3 (Supplement_1) ◽

pp. S293-S293

Author(s):

Jeffrey Burr ◽

Lien Quach

Keyword(s):

Social Support ◽

Older Adults ◽

Social Isolation ◽

Perceived Social Support ◽

Well Being ◽

Community Dwelling ◽

Outcome Variable ◽

Falls Risk ◽

Risk Of Falls ◽

Perceived Social Isolation

Abstract Relatively little is known about the relationship between social isolation and the risk of falls among older adults. Yet, a considerable amount of research demonstrates that lack of sufficient social relationships, broadly defined, represents a modifiable risk factor for many indicators of well-being in later life. This study examines the association between two types of social isolation and the risk of falls. The study also examines whether depression mediates the association between social isolation and risk of falls. Longitudinal data from the Health and Retirement Study (2006-2012) were collected from community-dwelling participants aged 65 and older (N=8,464). The outcome variable was number of falls self-reported over the observation period. Independent variables included perceived isolation (feeling lonely, perceptions of social support), social disconnectedness (e.g., having no friends or relatives living nearby, living alone), and number of depressive symptoms. Results from regression models indicated that social disconnectedness was associated with a 5% increase in the risk of falls (IRR=1.05, 95% CI=1.01-1.09). Perceived social support was associated with a 21% increase in the risk of falls; when examined together, perceived social support and loneliness were associated with a combined 37% increase in falls risk. Depression was associated with a 47% increase in falls. Depression mediated the association between perceived isolation and falls. Further, perceived isolation mediated the association between social disconnectedness and falls. Reducing perceived social isolation and social disconnectedness may be an avenue for designing interventions to reduce the risk of falls, especially for older adults with depression.

Download Full-text

THE ASSOCIATION OF SOCIAL SUPPORT AND ACTIVITIES WITH HEALTH IN SOUTH KOREA: DIFFERENCES IN AGE AND GENDER

Journal of Biosocial Science ◽

10.1017/s0021932009990563 ◽

2010 ◽

Vol 42 (3) ◽

pp. 409-424 ◽

Cited By ~ 9

Author(s):

SANG-SIK MOON ◽

SANG-MI PARK ◽

SUNG-IL CHO

Keyword(s):

Social Support ◽

Older Adults ◽

South Korea ◽

Emotional Support ◽

Older Men ◽

Poor Health ◽

Instrumental Support ◽

Social Activities ◽

Age And Gender ◽

And Gender

SummaryThis study investigated gender difference in the effects of social support, including emotional support and instrumental support (such as help when sick and financial assistance), and social activities on perceived health of middle-aged and older adults in South Korea. Data were acquired from 3771 men and 4954 women aged 40 years and older who participated in the 2005 cross-sectional survey of the Seoul Citizens Health and Social Indicators Survey. Using multiple regression analysis, both age- and gender-specific differences related to social support and engagement in social activities and self-rated poor health were examined. Poor emotional support from close friends, relatives or someone with whom one could talk about worries was strongly associated with poor self-rated health in men, with the greatest effect in older men. Lack of engagement in social activities was associated with self-rated poor health in older adults, especially in older men. Poor instrumental support was associated with perceived poor health only in middle-aged women. As a health improvement strategy for men aged 65 years and older especially, emotional support should be considered. Measures should be considered for encouraging social activities by older adults, particularly older men.

Download Full-text

Reciprocal Effects Between Health and Social Support in Older Adults’ Relationships With Their Children and Friends

Research on Aging ◽

10.1177/0164027515611182 ◽

2016 ◽

Vol 39 (2) ◽

pp. 300-321 ◽

Cited By ~ 9

Author(s):

Jung-Hwa Ha ◽

Sang Kyoung Kahng ◽

Namkee Choi

Keyword(s):

Social Support ◽

Older Adults ◽

Social Relationships ◽

Negative Impact ◽

Poor Health ◽

Positive Interactions ◽

Reciprocal Effects ◽

The Impact ◽

Bidirectional Effects ◽

Negative Interactions

This study examined the longitudinal reciprocal effects between health and social support in older adults’ relationships with their children and friends. Data are from the 2006 and 2010 waves of the Health and Retirement Study ( N = 3,760). We focused on three specific aspects of social support: frequency of contact, positive interactions, and negative interactions. We used autoregressive cross-lagged models to examine the bidirectional effects between social support and health. When the bidirectional effects between health and social support were simultaneously examined, the longitudinal effect of social support on health was not significant. In contrast, older adults’ poor health was associated with decreased contact and decreased positive interactions with friends as well as with increased negative interactions with their adult children and friends. The findings suggest that older adults’ poor health has a negative impact on their social relationships and that such effect surpasses the impact of social relationships on health.

Download Full-text

Social Relationships and Salivary Telomere Length Among Middle-Aged and Older African American and White Adults

The Journals of Gerontology Series B ◽

10.1093/geronb/gbx049 ◽

2017 ◽

Vol 74 (6) ◽

pp. 1053-1061 ◽

Cited By ~ 6

Author(s):

Karen D Lincoln ◽

Donald A Lloyd ◽

Ann W Nguyen

Keyword(s):

Social Support ◽

Older Adults ◽

African American ◽

African Americans ◽

Telomere Length ◽

Social Relationships ◽

Biological Aging ◽

Common Mechanism ◽

Middle Aged ◽

Negative Interaction

Abstract Objectives A common mechanism underlying premature morbidity may be accelerated biological aging as reflected by salivary telomere length (STL). This study examined the extent to which social relationships, both positive and negative, can be protective or confer risk relative to biological aging. Method Data from the Health and Retirement Study and multiple regression were used to examine cross-sectional associations between STL, self-reported social support, and negative interaction (e.g., conflict, criticism) with family in a nationally representative sample of African American and non-Hispanic White middle-aged and older adults (N = 4,080). Results Social support from family was associated with shorter STL. Negative interaction with family had no main effect on STL but interactions characterized by high social support and more frequent negative interactions were associated with longer STL. Negative interaction with family was negatively associated with STL for African Americans and Whites but the magnitude of the effect was greater for African Americans. Discussion Study findings highlight the role of social relationships in physiological deterioration among middle-aged and older adults and identify a potential mechanism whereby race is linked to accelerated biological aging. Findings highlight the importance of considering positive and negative aspects of social relationships to understand the consequences of social connections for cellular aging in diverse populations.

Download Full-text