6210 Week 8 Discussion

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=whsp20

Social Work in Public Health

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/whsp20

Risk and Protective Factors of Loneliness among Older Adults: The Significance of Social Isolation and Quality and Type of Contact

Barbra Teater, Jill M. Chonody & Nadia Davis

To cite this article: Barbra Teater, Jill M. Chonody & Nadia Davis (2021) Risk and Protective Factors of Loneliness among Older Adults: The Significance of Social Isolation and Quality and Type of Contact, Social Work in Public Health, 36:2, 128-141, DOI: 10.1080/19371918.2020.1866140

To link to this article: https://doi.org/10.1080/19371918.2020.1866140

Published online: 28 Dec 2020.

Submit your article to this journal

Article views: 2465

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Risk and Protective Factors of Loneliness among Older Adults: The Significance of Social Isolation and Quality and Type of Contact Barbra Teatera,b, Jill M. Chonodyc, and Nadia Davisa

aDepartment of Social Work, College of Staten Island, New York, New York, USA; bThe Graduate Center, City University of New York, New York, New York, USA; cSchool of Social Work, Boise State University, Boise, Idaho, USA

ABSTRACT Loneliness has a significant impact on the health and well-being of older people, including an increased risk of mortality. This cross-sectional study explored possible risk and protective factors that can help explain loneliness and emotional and social loneliness in a sample of community-dwelling older adults (N = 477). The survey incorporated a standardized scale of loneliness and items to assess type and quality of contact with others, community support, social isolation, physical health, cognitive health, and functional ability. Bivariate and multivariate analyses explored the factors that contributed to loneliness, emotional loneliness, and social loneliness. Results indicated overall quality of contact with others, use of phone contact, and social isolation was significant in all three regressions; other significant variables were different for each analysis. The findings support social work and public health recommendations for addressing loneliness, particularly within the current climate of “social distancing” under the COVID-19 pandemic.

KEYWORDS Loneliness; emotional loneliness; social loneliness; social isolation; older adults; public health emergency

Nearly 33% of adults aged 50 and older in the United States (US) reported being lonely (Wilson & Moulton, 2010), with over 42 million older adults having experienced chronic loneliness (Holt-Lunstad, 2017). Loneliness can be experienced at any age and has been found to progress nonlinearly across middle and old age with the highest prevalence of loneliness found among adults over 80 years of age (Demakakos, Nunn, & Nazroo, 2006; Dykstra, 2009; Wilson & Moulton, 2010). The effects of loneliness on older adults found in the literature indicate that loneliness predicts depression and psychological distress (Ng & Lee, 2019) and premature mortality (Cacioppo & Cacioppo, 2018; Valtorta & Hanratty, 2012). Social connections are viewed as both a preventative factor of loneliness and outcome of loneliness in that individuals who have strong social relationships are able to maintain independence, and maintaining independence allows one to engage in social relationships that prevent loneliness (Ten Bruggencate, Luijkx, & Sturm, 2019). This study builds on prior research by exploring the possible risk and protective factors associated with loneliness in older adults by examining how different types of social contact, quality of social interaction, a supportive community environment, social isolation, physical and cognitive health, and functional ability are associated with loneliness and emotional and social loneliness among community-dwelling older adults (aged 55 and older) residing in the US.

Literature review

Loneliness

Loneliness has both social and emotional components. Social loneliness is the loss or lack of a wider social network or circle of friends, family, people in the neighborhood, and/or acquaintances that

CONTACT Barbra Teater [email protected] 2800 Victory Blvd Staten Island, NY 10314.

SOCIAL WORK IN PUBLIC HEALTH 2021, VOL. 36, NO. 2, 128–141 https://doi.org/10.1080/19371918.2020.1866140

© 2020 Taylor & Francis Group, LLC

provide meaningful companionship and a sense of belonging, whereas emotional loneliness is the lack of an attachment figure who provides a more intimate relationship and emotional support, such as a partner (Weiss, 1973). Thus, a decrease in quantity of social networks may lead to social loneliness, and a decrease in quality of social networks may lead to emotional loneliness, which could include feelings of emptiness, abandonment, desolation, and insecurity (Olawa & Idemudia, 2019; Weiss, 1973). Moreover, loneliness is viewed as crucial for emotional and social well-being (Holt-Lunstad, 2017) and is viewed as the result of “situations in which the number of existing relationships is smaller than is considered desirable or admissible, as well as situations where the intimacy one wishes for has not been realized” (De Jong Gierveld, 1987, p. 120). Dykstra (2009) highlights that loneliness includes two elements: a subjective experience and a negative affect.

It is important to note the distinction between loneliness and social isolation although they both refer to a lack of social connection (Holt-Lunstad, 2017). Whereas loneliness is a negative, subjective experience, social isolation is more objective and refers to the absence of relationships or ties with other people (De Jong Gierveld & Van Tilburg, 2006; Dykstra, 2009). As Cohen-Mansfield and Perach (2015) report, “loneliness has been contrasted with belonging, whereas social isolation contrasts with social participation” (p. 109). According to the AARP Foundation (2020), an estimated 8 million adults age 50 and over are affected by social isolation with prolonged isolation and a lack of social connections having health risks equivalent to smoking 15 cigarettes a day (Holt-Lunstad, Smith, & Layton, 2010). Although they are two distinct concepts, examining their relationship with one another is important.

Prevalence and impact of loneliness on older adults

A study by Wilson and Moulton (2010) found 33% of a nationally representative sample of older adults (aged 50+) in the US to report being lonely, with age, income, and marital status significantly related to loneliness. Older adults, individuals with higher income, and those who were married reported less loneliness, yet no differences were found based on gender, education, or race/ ethnicity. In particular, a perceived lack of social support and a reduced network of friends were associated with loneliness, and individuals who were lonely were less likely to be engaged in social activities, such as attending religious services, volunteering, or participating in a community organization.

Many detrimental physical and psychological effects for older adults are linked to loneliness. For example, individuals who are lonely have an increased risk for mortality (Cacioppo & Cacioppo, 2018; Valtorta & Hanratty, 2012), have greater numbers of chronic illnesses and medical conditions (Theeke, 2010; Thurston & Kubzansky, 2009; Wilson & Moulton, 2010;), and have reduced ratings of subjective health (Cornwell & Waite, 2009; Theeke, 2010). A meta-analysis including 70 indepen- dent studies with over 3.4 million individuals reported social isolation, loneliness, and living alone have a significant and equivalent effect on risk for morality (Holt-Lunstad, Smith, Baker, Harris, & Stephenson, 2015). Loneliness has been found to impair activities of daily living, which in turn influences the individual’s functional status (Shanker, McMunn, Demakakos, Hamer, & Steptoe, 2017) and can lead to functional disability (Theeke, 2010). Additionally, loneliness has been linked to mental health in that it is associated with depression and lower levels of well-being (Golden et al., 2009; Routasalo, Savikko, Tilvis, Strandberg, & Pitkala, 2006; Tiikkainen & Heikkinen, 2005) and a predictor of suicide among older adults aged 65 and older (Waern, Rubenowitz, & Wilhelmson, 2003). Ten Bruggencate and colleagues conclude, “social needs do not change much with aging. Only the resources a person has seem to change during a lifetime” (p. 1845), with reduced resources being caused by health problems, reduced mobility, death of individuals in their social networks, and finances.

SOCIAL WORK IN PUBLIC HEALTH 129

Risk and protective factors of loneliness in older adults

Researchers have attempted to gain a better understanding of the risk and protective factors of loneliness in older adults to support ways in which to prevent and treat loneliness. One identified risk factor is living alone, particularly for older adults who have lost a spouse or partner, or is geographically distant from families who could provide social contact and support to cope with age- related difficulties (Holt-Lunstad, 2017; Zebhauser et al., 2015). Additionally, research has repeatedly established an association between depression and loneliness among older adults (Bath, Yang, & Nicholls, 2018; Courtin & Knapp, 2017; Domènech-Abella et al., 2017; Martinez, Baron, Largaespada, Ceruto, & Chaves, 2019; Zebhauser et al., 2015), but the causal direction of the relationship is unclear. For example, a cross-sectional study by Martinez et al. (2019) of community-dwelling older adults found lower social supports, lower social functioning, and higher loneliness were associated with higher levels of depression after controlling for chronic diseases and mobility difficulty, and a study including older adults who live alone found non-depressed participants were slightly over three times less likely to feel lonely when compared to depressed participants (Zebhauser et al., 2015). Finally, research has highlighted an association between physical limitations and health and loneliness. Wilson and Moulton (2010) found 55% of individuals who reported poor health also reported high levels of loneliness compared to 25% of individuals who reported excellent health, and Theeke (2010) found fewer age-related physical limitations were associated with less loneliness given that mobility and good health allowed the individuals to engage and maintain their social activities and social contacts.

Protective factors rest on a stable social network, meeting social needs, and the increasing use of social technology. A study including older adults who live alone found those participants who reported a stable social network were four times more likely not to report feeling lonely when compared to participants who reported a poor social network. When comparing the influence of social networks alongside other risk and protective factors, this study also identified social networks as the key factor while other factors, such as higher income, educational level, and being in good physical and mental health, were less likely to influence loneliness (Zebhauser et al., 2015). Olawa and Idemudia (2019) argue that a reduced social network can result in older adults experiencing both social and emotional loneliness, which, coupled with findings from the Pew Research Center (2009) that found the majority of adults in the US do not participate in any kind of social group, means that a significant number of older adults may be at risk for social isolation and loneliness. Continuing to examine the possible risk and protective factors of loneliness, as well as social and emotional loneliness, among older adults is crucial in helping to establish social work and public health strategies to address this public health concern. Therefore, this study aims to answer the following research question:

(1) To what extent do different types of social contact, quality of social interaction, a supportive community environment, social isolation, physical and cognitive health, and functional ability contribute to (a) loneliness; (b) emotional loneliness; and (c) social loneliness?

Method

Setting and sample

The data for this cross-sectional, exploratory study were drawn from an online survey of older adults in 2019 through Mechanical Turk (MTurk), an Amazon supported survey participant strategy that allows completion of tasks for monetary reward. This platform is popular for social scientists who in the past have relied on college students to complete surveys and experiments, and due to the large pool of potential respondents, researchers can target specific groups that are relevant to their study (e.g., older people) allowing larger groups to be reached (Paolacci & Chandler, 2014). Researchers are called “requesters” in this platform, the survey is called a “hit,” and the respondents are referred to as “workers” although the “work” in this case is small monetary payments for survey completion rather than actually being employed by Amazon. The use of MTurk is quite common in the empirical

130 B. TEATER ET AL.

literature (Paolacci & Chandler, 2014), and this platform has been found to be as reliable as traditional survey methods in terms of a demographically diverse population and reliability of data outcomes (Buhrmester, Kwang, & Gosling, 2011).

The recruitment of participants is conducted by Amazon MTurk based on the study inclusion and exclusion criteria, which are specified by the researcher. The inclusion criteria in this study were older adults aged 55+ who resided in the US and denoted as “master workers,” which, according to Amazon’s MTurk website, are those workers who have been identified by Amazon’s built technology that monitors workers and identifies those who “demonstrated excellence across a wide range of tasks.” Following best practices, when designing the hit, US residents only were selected and respon- dents were required to enter a code at the end of the survey, which prevents fake workers. “Master workers” were part of the inclusion criteria, and a monetary reward ($0.75 US) was deemed sufficient (Kees, Berry, Burton, & Sheehan, 2017). The study ceases when the requested sample size is reached at which point the data are transferred back to the researcher in an SPSS file. No identifying information (e.g., name, contact details) were recorded during survey completion. Ethical approval for the study was approved by the relevant Institutional Review Board. The participants were provided with a cover letter at the beginning of the survey, which described the purpose of the study, the confidential and voluntary nature of this study, and the contact information of the researcher. Completion of the online survey served as consent for participation. A total of 477 US residents 55+ completed the survey through the online MTurk platform.

Instrumentation

The author-constructed questionnaire consisted of 22 items that measured loneliness, types of social contact, quality of social interaction, the extent to which a community supports aging, social isolation, subjective physical health, subjective cognitive health, and functional ability. An additional six ques- tions were asked to capture the participants’ sociodemographics of identified gender, age, race/ ethnicity, sexuality, relationship status, and living arrangement.

Dependent variables The loneliness scale (De Jong Gierveld & Kamphuis, 1985; De Jong Gierveld & Van Tilbrug, 1999) measures the level of loneliness through two sub-scales of social loneliness and emotional loneliness, which served as the dependent variables in this study. The 11-item measure consists of five statements on social loneliness (e.g., “I miss having people around”) and six statements on emotional loneliness (e.g., “There are many people that I can trust completely”). The scale asks participants to indicate the extent to which they agree with the statements (1 = Strongly disagree – 6 = Strongly agree). After reverse scoring the five social loneliness items, the response to each item is summed to produce an overall loneliness score (11– 66) with higher scores indicating higher levels of loneliness. Additionally, the six items on social loneliness can be summed to produce an overall social loneliness score (6– 36), and the five items on emotional loneliness can be summed to produce an overall emotional loneliness score (5– 30). If a participant scored one or more missing values, the particular participant was deleted from the analysis. Cronbach’s alpha indicated a high level of internal consistency for the social loneliness subscale (α =.92), the emotional loneliness subscale (α = .90), and the overall loneliness scale (α = .93) in this study.

Independent variables Four different types of social contact were assessed individually by asking the following questions: “On average, how many days per week do you have face-to-face contact with others (family, friends)?”; “On average, how many days per week do you have phone contact (including FaceTime or Skype) with others (family, friends)?”; “On average, how many days per week do you text with others (family, friends)?” and “On average, how often do you have contact with groups or organizations (hobby groups, church) per week?” The participants were asked to rate the quality

SOCIAL WORK IN PUBLIC HEALTH 131

of their social contacts through the following question, “As a whole, how would you rate the quality of your interaction (spending time together, talking, and/or texting) with others?” (1 = Very poor – 6 = Excellent). The extent to which a community supports aging was measured by asking partici- pants to indicate their level of agreement (1 = Strongly disagree – 6 = Strongly agree) with the following statement, “My city/community supports me as I age (for example, places to rest in public, sidewalks, access to transportation)”. Social isolation was measured through a single-item indicator that asked participants to respond to the statement, “I feel socially isolated” (1 = Never – 7 = Always).

Additional independent variables related to health and functional ability were included as they have consistently been found to serve as both predictive and outcome factors to loneliness (Cornwell & Waite, 2009; Shanker et al., 2017; Theeke, 2010; Thurston & Kubzansky, 2009; Wilson & Moulton, 2010). Therefore, this study aimed to explore the extent to which health and functional ability contributed to loneliness and social and emotional loneliness alongside social isolation, community supports, and quality and type of contact. Physical health and cognitive health were measured through self-report by completing the following two statements: “My physical health is . . . ” and “My cognitive health (i.e., brain functioning) is . . . ” (1 = Very poor – 6 = Excellent). Finally, functional ability was asked through two questions that garnered respondents’ ability to engage in activities of daily living (ADLs), and instrumental activities of daily living (IADLs). ADLs were measured by the following question, “To what extent are you able to complete your activities such as cutting your toenails, completing your own bath, and going up and down the stairs?”, and IADLs were measured by the following question, “To what extent are you able to complete your activities such as grocery shopping, housework, and meal preparation?” (1 = I’m able to complete these activities without any difficulty – 4 = I have significant difficulty with these activities and need quite a bit of help”).

Data analysis

The data were analyzed in IBM SPSS, version 24, software using descriptive statistics to determine percentages, frequencies, and measures of central tendency for the sociodemographic variables and the items measuring loneliness, types and quality of social contact, community supports, social isolation, physical and cognitive health, and functional ability. Bivariate analyses were run to explore the relationship between two variables, for example, the relationship between loneliness and living arrangement, or the relationship between functional ability and loneliness. Before conducting ANOVAs, descriptive statistics were examined to ensure at least five cases within each category; if there were less than five cases in a specific category, then the specific categories of the variable were combined to create an “other” category. Post-hoc comparisons using the LSD test were preformed to assess specific differences between groups.

The variables found to be significant at the bivariate level were included in three separate ordinary least squares (OLS) regressions analyses to determine which factors contributed to overall loneliness, emotional loneliness, and social loneliness. Gender (0 = men; 1 = women) and age were included as control variables. The predictor variables were relationship status (1 = single, never married; 0 = all other relationship types), living arrangement (1 = lived alone; 0 = all other arrangements), face-to-face contact, phone contact, text contact, group/organization contact, quality of contact, supportive community environment for aging, social isolation, physical health, cognitive health, ADLs, and IADLs. The outcome variables were overall loneliness, emotional loneliness, and social loneliness. The control and predictor variables were entered simultaneously. Missing data were addressed through listwise deletion. Alpha was set at .05.

132 B. TEATER ET AL.

Results

Sample sociodemographics

A total of 468 respondents were included in the data analysis after removing nine cases due to missing items on the loneliness scale. The mean age of the respondents was 63.52 years with a range from 55 to 81 years. Over 68% identified as female, nearly 90% identified as White (non-Hispanic), and 96% identified as straight or heterosexual. Over 49% of respondents reported their relationship status as “married/partnered” with 20.30% reporting their relationship status as “divorced.” Finally, over 52% reported living with their spouse/partner and 27.30% reported living alone. Table 1 reports the sociodemographics of the sample.

Loneliness and potential risk and protective factors for loneliness

Table 2 reports the descriptive statistics for the loneliness, emotional loneliness, and social loneliness scales as well as other potential risk and protective factors for loneliness. The respondents had an

Table 1. Sample sociodemographics (N = 468).

Variable (n) M (SD) % (f)

Age (466) 63.52 (4.78) 55– 59 23.18% (108) 60– 64 39.27% (183) 65– 69 25.75% (120) 70– 74 8.80% (41) 75– 81 3.00% (14) Gender Female 68.60% (321) Male 31.40% (147) Race/Ethnicity African American/Black 4.90% (23) American Indian/Native American 0.20% (1) Asian American/Asian 0.90% (4) Biracial/Multiracial 0.90% (4) Chicano/Mexican-American 1.10% (5) Puerto Rican 0.20% (1) White (non-Hispanic) 89.70% (420) Another Racial/Ethnic Identity 2.10% (10) Sexuality Asexual 0.20% (1) Bisexual 1.30% (6) Lesbian, Gay, or Homosexual 2.40% (11) Straight or Heterosexual 95.90% (449) Don’t Know 0.20% (1) Relationship Status Divorced 20.30% (95) Divorced and Dating 2.10% (10) Divorced and Remarried/Partnered 4.10% (19) Married/Partnered 49.70% (232) Single, Never Married 11.80% (55) Single and Dating 0.90% (4) Widowed 7.10% (33) Widowed and Remarried/Partnered 1.10% (5) Other 3.00% (14) Living Arrangement Live alone 27.30% (127) Adult child 8.60% (40) Family member other than child 4.90% (23) Friend 0.20% (1) Roommate 1.30% (6) Spouse/Partner 52.60% (245) Other 5.20% (24)

SOCIAL WORK IN PUBLIC HEALTH 133

overall level of loneliness, emotional loneliness, and social loneliness that fell within the mid-range of the possible scores on these scales. Respondents reported spending more days per week having face-to- face contact, followed by text contact, phone contact, and group/organization contact, with a reported quality of contact as generally good to very good. The respondents tended to agree that their community supported them as they age, and had a lower level of social isolation. Overall physical and cognitive health was reported as good to very good, and respondents reported generally being able to complete ADLs and IADLs.

Relationship between sociodemographics, risk and protective factors, and loneliness

Demographics Bivariate analyses indicated that no relationship existed between age, race, or sexuality and overall loneliness, emotional loneliness, or social loneliness. Additionally, no difference in overall loneliness or emotional loneliness was found based on gender, but men were found to have a higher level of social loneliness (M = 14.96; SD = 6.51) compared to women (M = 13.57; SD = 6.81), t(466) = 2.08, p = .038.

Relationship status Table 3 reports the results from the bivariate analyses for risk and protective factors and overall loneliness, emotional loneliness, and social loneliness. There was a difference in relationship status and loneliness and post-hoc comparisons revealed: (a) single, never married respondents had higher levels of loneliness compared to respondents who were divorced and dating (p = .008), divorced and remarried/partnered (p = .002), and married/partnered (p < .001); and (b) divorced respondents had higher levels of loneliness compared to respondents who were divorced and dating (p = .035), divorced and remarried/partnered (p < .001), and married/partnered (p < .001). Likewise, there was a relationship between relationship status and emotional loneliness, and post-hoc comparisons revealed: (a) respondents who were single, never married had higher levels of emotional loneliness compared to respondents who were divorced and dating (p = .015), divorced and remarried/partnered (p = .009), and married/partnered (p < .001); (b) respondents who were divorced had higher levels of emotional loneliness compared to respondents who were divorced and dating (p = .036), divorced and remarried/partnered (p = .026), and married/partnered (p = .001); and (c) respondents who were widowed had higher levels of emotional loneliness compared to respondents who were married and partnered (p = .001). Finally, there was a relationship between relationship status and social loneliness, and post-hoc comparisons revealed: (a) respondents who were single, never married had higher levels of social loneliness compared to respondents who were divorced and dating (p = .020), divorced and remarried/partnered (p = .002), married/partnered (p < .001), widowed (p = .010), and widowed and remarried/partnered (p = .030); and (b) respondents who were divorced had higher levels of social

Table 2. Loneliness and potential risk and protective factors for loneliness (N = 468).

Variable (n) M (SD) Theoretical Range

Loneliness (468) 30.53 (13.10) 11– 66 Emotional Loneliness (468) 16.52 (7.66) 6– 36 Social Loneliness (468) 14.01 (6.74) 5– 30 Face-to-Face Contact (467) 5.79 (2.03) 0– 7 Phone Contact (465) 4.00 (2.54) 0– 7 Text Contact (464) 4.41 (2.67) 0– 7 Group/Organization Contact (467) 1.59 (1.92) 0– 7 Quality of Contact (466) 4.38 (1.21) 1– 6 Community Supports Aging (467) 4.05 (1.42) 1– 6 Social Isolation (467) 2.72 (1.66) 1– 7 Physical Health (466) 4.11 (1.11) 1– 6 Cognitive Health (467) 5.08 (0.89) 1– 6 ADLs (467) 1.19 (0.59) 1– 4 IADLs (468) 1.18 (0.59) 1– 4

134 B. TEATER ET AL.

loneliness compared to respondents who were divorced and remarried/partnered (p = .024), and married/partnered (p = .001).

Living arrangements Table 3 also reports the details of the respondents’ living arrangements and the relationship with overall loneliness, emotional loneliness, and social loneliness. There was a relationship between living arrangement and overall loneliness, and post-hoc comparisons indicated respondents who lived alone had higher levels of loneliness compared to respondents who lived with an adult child (p = .004), with a spouse/partner (p <.001), and those living in an “other” arrangement (p = .004). Likewise, there was a relationship between living arrangement and emotional loneliness, and post-hoc comparisons indicated respondents who lived alone had higher levels of emotional loneliness compared to

Table 3. Bivariate analyses: Risk and protective factors and loneliness, emotional loneliness, and social loneliness (N = 455).

Loneliness Emotional Loneliness Social Loneliness

Variable M (SD) F/r df p M (SD) F/r df p M (SD) F/r df p

Relationship Status

4.66 7, 459 <.001 3.62 7, 459 .001 4.38 7, 459 <.001

Single, never married

36.58 (13.51) 19.31 (7.65) 17.27 (6.64)

Divorced 33.92 (14.27) 18.24 (8.48) 15.67 (7.84) Widowed 31.73 (13.31) 18.18 (8.01) 13.55 (6.21) Married/

partnered 28.16 (11.96) 15.25 (7.00) 12.91 (6.11)

Widowed and remarried/

partnered 26.20 (9.31) 15.60 (6.35) 10.60 (4.10) Divorced and

remarried/ partnered 26.00 (9.72) 14.05 (6.82) 11.95 (5.21) Divorced and

dating 25.00 (9.72) 13.00 (6.91) 12.00 (6.27)

Other 29.72 (12.88) 16.22 (7.63) 13.50 (6.64) Living

Arrangement 6.69 5, 460 <.001 6.47 5, 460 <.001 4.53 5, 460 <.001

Lived alone 35.81 (13.52) 19.53 (8.03) 16.28 (7.19) Family member

(other) 32.48 (14.15) 18.04 (8.10) 14.43 (7.14)

Friend/ Roommate

31.00 (11.31) 16.86 (7.69) 14.14 (6.07)

Adult child 29.10 (12.41) 15.55 (6.96) 13.55 (6.74) Spouse/partner 28.09 (11.92) 15.19 (6.97) 12.90 (6.13) Other

arrangements 14.66 (2.99) 14.38 (8.41) 13.29 (7.15)

Face-to-Face Contact

−.40 <.001 −.37 <.001 −.36 <.001

Phone Contact −.46 <.001 −.38 <.001 −.46 <.001 Text Contact −.41 <.001 −.33 <.001 −.43 <.001 Group/

Organization Contact

−.31 <.001 −.28 <.001 −.28 <.001

Quality of Contact

−.69 <.001 −.58 <.001 −.68 <.001

Community Supports Aging

−.32 <.001 −.20 <.001 −.39 <.001

Social Isolation .75 <.001 .74 <.001 .61 <.001 Physical Health −.41 <.001 −.34 <.001 −.41 <.001 Cognitive Health −.33 <.001 −.33 <.001 −.27 <.001 ADLs .20 <.001 .17 <.001 .19 <.001 IADLs .23 <.001 .20 <.001 .22 <.001

SOCIAL WORK IN PUBLIC HEALTH 135

respondents who lived with an adult child (p = .003), with a spouse/partner (p <.001), and those living in an “other” arrangement (p = .002). Finally, there was a relationship between living arrangement and social loneliness, and post-hoc comparisons indicated respondents who lived alone had higher levels of social loneliness compared to respondents who lived with an adult child (p = .002), with a spouse/ partner (p <.001), and those living in an “other” arrangement (p = .042).

Type and quality of contact, and social and physical health The correlations among the social interaction, physical health, and social health variables are described in Table 3. The analyses revealed statistically significant relationships between all the potential risk and protective factors and loneliness, emotional loneliness, and social loneliness at p < .001. In particular, face-to-face contact, phone contact, text contact, group/organization contact, quality of contact, a supportive community, physical health, and cognitive health were negatively associated with lone- liness, emotional loneliness, and social loneliness. Social isolation, difficulty with ADLs, and difficulty with IALs were positively associated with loneliness, emotional loneliness, and social loneliness.

Risk and protective factors contributing to loneliness

Loneliness The results of the regression analysis for loneliness indicated that seven variables explained 70% of the variance. As Table 4 reports, being single/never married, living alone, less phone contact, lower quality of interaction, less community supports, higher levels of social isolation, and poorer subjective physical health were associated with higher levels of loneliness. Collinearity diagnostic tests indicated no problems with multicollinearity in this model (Durbin-Watson = 2.09, tolerance >.2, variance inflation factor <10; Field, 2013).

Emotional loneliness The results of the regression analysis indicated that four variables explained 60% of the variance. As Table 4 reports, living alone, less phone contact, and lower quality of interaction, and higher levels of social isolation were associated with higher levels of emotional loneliness. Collinearity diagnostic tests indicated no problems with multicollinearity in this model (Durbin-Watson = 2.00, tolerance >.2, variance inflation factor <10; Field, 2013).

Social loneliness The results of the regression analysis indicated that seven variables explained 61% of the variance. As Table 4 reports, being single/never married, less phone contact, lower quality of interaction, less community supports, higher levels of social isolation, poorer subjective physical health, and higher subjective cognitive health were associated with higher levels of social loneliness. Collinearity diag- nostic tests indicated no problems with multicollinearity in this model (Durbin-Watson = 2.06, tolerance >.2, variance inflation factor <10; Field, 2013).

Discussion

Results of this study support previous research regarding loneliness. First, this sample was moderately lonely overall, which is reflective of previous findings that found around 33% of older people are lonely (Wilson & Moulton, 2010). Subjective physical health was significant in the multivariate analyses for overall loneliness and social loneliness, which has been found in previous studies (e.g., Wilson & Moulton, 2010); however, social isolation, quality of contacts, and phone contact had greater impact in all three regression analyses based on their effect size. Nonetheless, subjective physical health played a larger role in explaining the variance for social loneliness, which echoes previous research that suggested that those who have fewer age-related physical impairments report less loneliness (Theeke, 2010). Relatedly, ability to complete ADLs and IADLs, while not significant in the regressions, were

136 B. TEATER ET AL.

reported to be high among this sample, which further supports this past research and points to the role of health and mobility as protective factors for loneliness.

Relationship status, particularly being single/never married, played a small role in the explanation of overall loneliness and social loneliness, which was also found in past research (Dahlberg, Andersson, McKee, & Lennartsson, 2015; Wilson & Moulton, 2010). Socializing both with a partner

Table 4. Factors contributing to loneliness, emotional loneliness, and social loneliness (N = 450).

Variable B SE B ß t p

Loneliness Gender −.19 .79 −.01 −.23 .815 Age .06 .08 .02 .81 .418 Relationship Status 2.51 1.14 .06 2.20 .028 Living Arrangement 2.01 .97 .07 2.07 .030 Face-to-Face Contact −.19 .22 −.03 −.89 .375 Phone Contact −.58 .17 −.11 −3.43 .001 Text Contact −.13 .17 −.03 −.77 .445 Group Contact −.01 .19 −.00 −.03 .974 Quality Contact −2.89 .43 −.27 −6.79 <.001 Community Supports −.61 .27 −.07 −2.27 .024 Social Isolation 3.86 .27 .49 14.21 <.001 Physical Health −.92 .41 −.08 −2.27 .024 Cognitive Health .28 .45 .02 .62 .533 ADLs −1.19 1.18 −.05 −1.01 .313 IADLs 1.47 1.17 −.07 1.26 .208 Adjusted R2 .67 F 66.60* Emotional Loneliness Gender .12 .53 .01 .23 .818 Age .04 .05 .26 .83 .409 Relationship Status 1.13 .77 .05 1.47 .141 Living Arrangement 1.30 .66 .08 1.98 .048 Face-to-Face Contact −.06 .15 −.02 −.41 .681 Phone Contact −.23 .11 −.08 −2.03 .043 Text Contact −.03 .12 −.01 −.27 .790 Group Contact −.09 .13 −.02 −.73 .464 Quality Contact −1.05 .29 −.16 −3.65 <.001 Community Supports .14 .18 .03 .80 .424 Social Isolation 2.66 .18 .57 14.49 <.001 Physical Health −.23 .27 −.03 −.84 .400 Cognitive Health −.28 .31 −.03 −.93 .356 ADLs −.54 .79 −.04 −.67 .501 IADLs .58 .79 .04 .04 .464 Adjusted R2 .59 F 43.43* Social Loneliness Gender −.31 .46 −.02 −.67 .506 Age .02 .04 .01 .43 .665 Relationship Status 1.38 .67 .07 2.07 .040 Living Arrangement .71 .57 .05 1.25 .213 Face-to-Face Contact −.14 .13 −.04 −1.04 .297 Phone Contact −.35 .09 −.13 −3.52 <.001 Text Contact −.10 .10 −.04 −1.00 .318 Group Contact .09 .12 .03 .79 .431 Quality Contact −1.84 −.25 −.33 −7.40 <.001 Community Supports −.75 .16 −.16 −4.79 <.001 Social Isolation 1.21 .16 .29 7.57 <.001 Physical Health −.69 .24 −.11 −2.91 .004 Cognitive Health .56 .27 .08 2.13 .034 ADLs −.66 .69 −.06 −.95 .343 IADLs .89 .69 .08 1.31 .191 Adjusted R2 .59 F 44.72*

B = unstandardized beta; SE B = standard error for the unstandardized beta. *p <.001

SOCIAL WORK IN PUBLIC HEALTH 137

and also as a couple may help prevent feelings of loneliness and also decrease the likelihood of social isolation. Likewise, living alone played a small role in the explanation for overall loneliness and emotional loneliness. When living alone, socializing can seem more daunting as these activities have to be endeavored solo, which may increase social isolation in an older person. At the bivariate level, single/never married participants had the greatest level of loneliness whereas those participants who were married or dating had the lowest levels of loneliness. The importance of relationships in older adulthood has been studied extensively, and research suggests that relationship quality has positive impact on both health (Umberson, Williams, Powers, Liu, & Needham, 2006) and well-being (Sherwood, Kneale, & Bloomfield, 2014). In a qualitative study of more than 1,500 participants 55 and older in an ongoing relationship, companionship and laughter were amongst the “best liked” elements of the relationship (Chonody & Gabb, 2019).

In addition to the quality of the contacts, which had a moderate effect size in the overall regression analyses, frequency of contact also played a small role, which has been found in other studies where frequency of contact with social networks was important in feelings of loneliness (Domènech- Abella et al., 2017; Rico-Uribe et al., 2016). Surprisingly, phone contact was significant in the multi- variate analyses, but face-to-face contact was not. Similarly, contact via texting and contact that occurs in groups and organizations were not significant either. Perhaps phone contact helps when there is a physical issue or a mobility problem, and emotional support can be garnered through phone contact with friends and loved ones. This type of contact also bridges geographical gaps, which in our increasingly mobile society could be a significant issue in older adulthood, not just in the case of adult children who have moved away, but also close friends. Equally, phone contact can be a way to reach adults who are at risk of experiencing loneliness in the time of public health crises, such as the novel corona disease 19 (COVID-19) pandemic that is taking place at the time of writing this manuscript. When the public is practicing “social distancing” during a pandemic, and older adults are being urged to socially isolate, the use of phone contact may serve as a protective factor to lessen the feelings of loneliness both by family and friends, but also volunteers through befriending services, or organized peer-support groups.

Limitations

The results of this study should be considered within the context of its limitations. First our sample was not representative of the older adult population, particularly in terms of race/ethnicity. The findings from this predominately White, heterosexual sample cannot be generalized to other groups. This may in part be explained by the research that indicates African American/Black and Latinx workers are underrepresented on MTurk (Berinsky, Huber, & Lenz, 2012). Relatedly, use of MTurk workers limits generalizability given that they may be more tech-savvy, which could have implications for negotiating loneliness in more modern ways, such as the use of social media and apps (e.g., Marco Polo, Groupme). Additionally, past studies have indicated that MTurk workers are likely to be in better health, have more education, and more likely to be unemployed (Goodman, Cryder, & Cheema, 2013). Nonetheless, these samples “are more representative of the US population than in-person convenience samples” (Berinsky et al., 2012, p. 351). Replication with a more diverse group of older people, including those who are not internet users, would be important to our understanding of risk and protective factors for loneliness.

Secondly, income was not included in the survey, which has been found to be associated with loneliness (Cohen-Mansfield, Hazan, Lerman, & Shalom, 2016; Niedzwiedz et al., 2016; Wilson & Moulton, 2010). Financial resources may play a key role in different types of loneliness given that it can increase accessibility, and future research should explore its role in this complex interplay of factors. Third, the measures for loneliness and social isolation may have influenced the significance of the risk and protective factors. Expanding the measurement strategy would be important for replication studies and would allow a greater assessment of these interrelated concepts. Similarly, while the explained variance was quite high across the three analyses, additional risk and protective factors

138 B. TEATER ET AL.

were not included in this study. Other variables, such as income and stressors, could provide additional insight into these predictors. Despite the limitations, this study has contributed to the knowledge base by highlighting the importance of type and quality of contact in potentially serving as protective factors to loneliness among older adults.

Implications for practice

The findings of this study have replicated findings from previous studies, but also add to this substantive literature base, particularly when considering them in the time of a public health emergency. The three significant variables common across the three regression analyses indicate social isolation explains loneliness, yet the use of phone contact and the quality of contact can serve as protective factors for loneliness. These findings, along with the significance of community supports for loneliness and social loneliness, provide the foundation for the following recommen- dations for future emergency preparedness that specifically recognizes the health needs of older adults in the time of a public health emergency. First, community supports explained a small but significant effect size in explaining loneliness, particularly social loneliness. This highlights the need for supportive and age-friendly communities that provide ways to increase the ability for older adults to leave their home and use the physical space of their community, particularly in times of social distancing when interacting with others is not a possibility. Participating in walks outside of their home or places to sit in the outdoors, even alone, may assist with physical and mental health, which in turn may serve as protective factors to loneliness. Second, there is a need to enhance the use of telehealth and video conferencing to meet the physical and mental health needs of older adults, which will also serve as a preventative measure to reduce social isolation which has been found to be highly correlated with loneliness. This study found phone contact to serve as a protective factor for loneliness, therefore, this type of contact in addition to the availability of telehealth and video conferencing need to become a usual way of reaching and communicating with older adults. This will require increasing the education and training of social work and public health workers in the use of telehealth, embedding telehealth into routine service delivery, and ensuring financial reimbursement for such services (Smith et al., 2020). Finally, older adults who are socially isolating during a public health emergency should be identified and contacted six months after release from isolation (Torales, O’Higgins, Castaldelli-Maia, & Ventriglio, 2020) to assess loneliness. Gardner, States, and Bagley (2020) argue that “isolation and loneliness is a medical problem in its own right” (p. 5), and understanding the risk and protective factors can serve as a starting point for prevention and intervention strategies to address this public health concern.

As we experience this public health emergency, fresh insights into how people fought feelings of loneliness will add to our understanding of how to mobilize both new and old connections in creative ways. Using apps, such as Marco Polo, or setting up dinners on Zoom are some of the ways that people are continuing to connect, and our research suggests that contact does not need to take place face-to-face to have an impact on loneliness. Programming for older adults, both those who are aging in place and those who may be living in a facility, can also take advantage of this technology to help keep social connections alive. For example, a global campaign emerged from the United Kingdom to “adopt a grandparent” to foster intergenerational friendships and cross- cultural learning experiences for older people living in care facilities (see https://chdliving.co.uk/ adopt-grandparent). Innovation coupled with technology can help combat the growing social disconnection that many older people feel.

Disclosure statement

No potential conflict of interest was reported by the authors.

SOCIAL WORK IN PUBLIC HEALTH 139

References

AARP Foundation. (2020, February 18). What is isolation? Author. Retrieved from https://connect2affect.org/about- isolation/

Bath, P., Yang, P. H., & Nicholls, J. (2018). Changes in loneliness and patterns of loneliness among older people. Innovation in Aging, 2(suppl_1), 480–481. doi:10.1093/geroni/igy023.1794

Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research: Amazon. com’s Mechanical Turk. Political Analysis, 20(3), 351–368. doi:10.1093/pan/mpr057

Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high- quality, data? Perspectives on Psychological Science, 6(1), 3–5. doi:10.1177/1745691610393980

Cacioppo, J. T., & Cacioppo, S. (2018). The growing problem of loneliness. The Lancet, 391(10119), 426. doi:10.1016/ S0149-6736(18)30142-9

Chonody, J. M., & Gabb, J. (2019). Understanding the role of relationship maintenance in enduring couple partnerships in later adulthood. Marriage & Family Review, 55(3), 216–238. doi:10.1080/01494929.2018.1458010

Cohen-Mansfield, J., Hazan, H., Lerman, Y., & Shalom, V. (2016). Correlates and predictors of loneliness in older adults: A review of quantitative results informed by qualitative insights. International Psychogeriatrics, 28(4), 557–576. doi:10.1017/S1041610215001532

Cohen-Mansfield, J., & Perach, R. (2015). Interventions for alleviating loneliness among older persons: A critical review. American Journal of Health Promotion, 29(3), e109–e125. doi:10.4278/ajhp.130418-LIT-182

Cornwell, E. Y., & Waite, L. J. (2009). Social disconnectedness, perceived isolation, and health among older adults. Journal of Health and Social Behavior, 50(1), 31–48. doi:10.1177/002214650905000103

Courtin, E., & Knapp, M. (2017). Social isolation, loneliness and health in old age: A scoping review. Health and Social Care in the Community, 25(3), 799–812. doi:10.1111/hsc.12311

Dahlberg, L., Andersson, L., McKee, K. J., & Lennartsson, C. (2015). Predictors of loneliness among older women and men in Sweden: A national longitudinal study. Aging & Mental Health, 19(5), 409–417. doi:10.1080/ 13607863.2014.944091

De Jong Gierveld, J. (1987). Developing and testing a model of loneliness. Journal of Personality and Social Psychology, 53 (1), 119–128. doi:10.1037/0022-3514.53.1.119

De Jong Gierveld, J., & Kamphuis, F. (1985). The development of a Rasch-Type loneliness scale. Applied Psychological Measurement, 9(3), 289–299. doi:10.1177/014662168500900307

De Jong Gierveld, J., & Van Tilbrug, T. (1999). Manual of the loneliness scale. Amsterdam, The Netherlands: Vrije Universiteit.

De Jong Gierveld, J., & Van Tilburg, T. (2006). A 6-item scale for overall, emotional, and social loneliness: Confirmatory tests on survey data. Research on Aging, 28(5), 582–598. doi:10.1177/0164027506289723

Demakakos, P., Nunn, S., & Nazroo, J. (2006). Loneliness, relative deprivation, and life satisfaction. In J. Banks, E. Breeze, C. Lessof, & J. Nazroo (Eds.), Retirement, health, and relationships of the older population in England: The 2004 English longitudinal study of aging (pp. 297–318). London, UK: Institute of Fiscal Studies.

Domènech-Abella, J., Lara, E., Rubio-Valera, M., Olaya, B., Moneta, M. V., Rico-Uribe, L. A., & Haro, J. M. (2017). Loneliness and depression in the elderly: The role of social network. Social Psychiatry and Psychiatry Epidemiology, 52 (4), 381–390. doi:10.1007/s00127-017-1339-3

Dykstra, P. A. (2009). Older adult loneliness: Myths and realities. European Journal of Ageing, 6(2), 91–100. doi:10.1007/ s10433-009-0110-3

Field, A. (2013). Discovering statistics using SPSS (4th ed.). London, UK: SAGE Publications, Inc. Gardner, W., States, D., & Bagley, N. (2020). The coronavirus and the risk to the elderly in long-term care. Journal of

Aging & Social Policy, 32(4–5), 310–315. doi:10.1080/08959420.2020.1750543 Golden, J., Conroy, R. M., Bruce, I., Denihan, A., Greene, E., Kirby, M., & Lawlor, B. A. (2009). Loneliness, social support

networks, mood, and wellbeing in community dwelling elderly. International Journal of Geriatric Psychiatry, 24(7), 694–700. doi:10.1002/gps.2181

Goodman, J. K., Cryder, C. E., & Cheema, A. (2013). Data collection in a flat world: The strengths and weaknesses of Mechanical Turk samples. Journal of Behavioral Decision Making, 26(3), 213–224. doi:10.1002/bdm.1753

Holt-Lunstad, J. (2017). The potential public health relevance of social isolation and loneliness: Prevalence, epidemiol- ogy, and risk factors. Public Policy & Aging Report, 27(4), 27–130. doi:10.1093/ppar/prx030

Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10(2), 227–237. doi:10.1177/ 1745691614568352

Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risks: A meta-analytic review. PLoS Medicine, 7(7), e1000316. doi:10.1371/journal.pmed.1000316

Kees, J., Berry, C., Burton, S., & Sheehan, K. (2017). An analysis of data quality: Professional panels, student subject pools, and Amazon’s Mechanical Turk. Journal of Advertising, 46(1), 141–155. doi:10.1080/ 00913367.2016.1269304

140 B. TEATER ET AL.

Martinez, I. L., Baron, A. C., Largaespada, V., Ceruto, J., & Chaves, P. H. (2019). Social correlates of depressive symptoms among Cuban and other Latino older adults in South Florida. Journal of Cultural Diversity, 26(4), 149–156. doi:10.1080/07317110802478024

Ng, S. M., & Lee, T. M. (2019). The mediating role of hardiness in the relationship between perceived loneliness and depressive symptoms among older. Aging & Mental Health. doi:10.1080/13607863.2018.1550629

Niedzwiedz, C. L., Richardson, E. A., Tunstall, H., Shortt, N. K., Mitchell, P. J., & Pearce, J. R. (2016). The relationship between wealth and loneliness among older people across Europe: Is social participation protective? Preventive Medicine, 91, 24–31. Retrieved from https://doi.10.1016/j.ypmed.2016.07.016

Olawa, B. D., & Idemudia, E. S. (2019). Satisfaction with adult children’s achievements is associated with depression and loneliness in later-life: The mediating roles of children’s support and gratitude. Educational Gerontology, 45(4), 269–282. doi:10.1080/03601277.2019.1612559

Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23(3), 184–188. doi:10.1177/0963721414531598

Pew Research Center. (2009). Social isolation and new technology: How the internet and mobile phones impact Americans’ social networks. Washington, DC: Pew Internet & American Life Project. Retrieved from https://www.pewresearch. org/internet/2009/11/04/social-isolation-and-new-technology/

Rico-Uribe, L. A., Caballero, F. F., Olaya, B., Tobiasz-Adamczyk, B., Koskinen, S., Leonardi, M., . . . Miret, M. (2016). Loneliness, social networks, and health: A cross-sectional study in three countries. PLoS ONE, 11(1), e0145264. doi:10.1371/journal.pone.0145264

Routasalo, P. E., Savikko, N., Tilvis, R. S., Strandberg, T. E., & Pitkala, K. H. (2006). Social contacts and their relationship to loneliness among aged people: A population-based study. Gerontology, 52(3), 181–187. doi:10.1159/000091828

Shanker, A., McMunn, A., Demakakos, P., Hamer, M., & Steptoe, A. (2017). Social isolation and loneliness: Prospective associations with functional status in older adults. Health Psychology, 36(2), 179–187. doi:10.1037/hea0000437

Sherwood, C., Kneale, D., & Bloomfield, B. (2014). The way we are now: The UK’s relationships. Edinburgh, Scotland: Relate.

Smith, A. C., Thomas, E., Snoswell, C. L., Haydon, H., Mehrotra, A., Clemensen, J., & Caffery, L. J. (2020). Telehealth for global emergencies: Implications for coronavirus disease 2019 (COVID-19). Journal of Telemedicine and Telecare, 26 (5), 309–313. doi:10.1177/1357633X20916567

Ten Bruggencate, T., Luijkx, K. G., & Sturm, J. (2019). When your world gets smaller: How older people try to meet their social needs, including the role of social technology. Ageing & Society, 39(8), 1826–1852. doi:10.1017/ s0144686X18000260

Theeke, L. A. (2010). Sociodemographic and health-related risks for loneliness and outcome differences by loneliness status in a sample of U.S. older adults. Research in Gerontological Nursing, 3(2), 113–125. doi:10.3928/19404921- 20091103-99

Thurston, R. C., & Kubzansky, L. D. (2009). Women, loneliness, and incident coronary heart disease. Psychosomatic Medicine, 71(8), 836–842. doi:10.1097/PSY.0b013e3181b40efc

Tiikkainen, P., & Heikkinen, R. L. (2005). Associations between loneliness, depressive symptoms and perceived togetherness in older people. Aging & Mental Health, 9(6), 526–534. doi:10.1080/13607860500193138

Torales, J., O’Higgins, M., Castaldelli-Maia, J. M., & Ventriglio, A. (2020). The outbreak of COVID-19 coronavirus and its impact on global mental health. International Journal of Social Psychiatry, 66(4), 317–320. doi:10.1177/ 0020764020915212

Umberson, D., Williams, K., Powers, D. A., Liu, H., & Needham, B. (2006). You make me sick: Marital quality and health over the life course. Journal of Health and Social Behavior, 47(1), 1–16. doi:10.1177/002214650604700101

Valtorta, N., & Hanratty, B. (2012). Loneliness, isolation and the health of older adults: Do we need a new research agenda? Journal of the Royal Society of Medicine, 105(12), 518–522. doi:10.1258/jrsm.2012.120128

Waern, M., Rubenowitz, E., & Wilhelmson, K. (2003). Predictors of suicide in the old elderly. Gerontology, 49(5), 328–334. doi:10.1159/000071715

Weiss, R. S. (1973). Loneliness: The experiences of emotional and social isolation. Cambridge, MA: MIT Press. Wilson, C., & Moulton, B. (2010). Loneliness among older adults: A national survey of adults 45+. Washington, DC:

AARP. Zebhauser, A., Baumert, J., Emeny, R. T., Ronel, J., Peters, A., & Ladwig, K. H. (2015). What prevents old people living

alone from feeling lonely? Findings form the KORA-Age-study. Aging & Mental Health, 19(9), 773–780. doi:10.1080/ 13607863.2014.977769

SOCIAL WORK IN PUBLIC HEALTH 141