Loneliness and Mode of Social Contact in Late Life

Abstract

Objectives

Social contact may alleviate loneliness, but little is known about within-person daily fluctuations in loneliness and social encounters. Older adults who feel lonely may engage in different modes of social contact (in-person, phone, digital). This study asked how different forms of contact are associated with loneliness throughout the day.

Methods

Participants were 313 community-dwelling older adults (aged 65–90). They completed ecological momentary assessments reporting on their social encounters (e.g., type of social partner, mode of contact) and their loneliness every 3 hr for 5–6 days. We differentiated close social ties from ties not identified as close (i.e., weak ties).

Results

We examined within-person effects using multilevel models. Findings revealed that momentary loneliness predicted a greater likelihood of phone contact in the next 3 hr. However, only in-person contact was associated with lower levels of loneliness. Regarding close and weak ties, momentary loneliness was associated with more in-person and phone contact with close ties, yet fewer in-person contacts with weak ties. In-person contact with both close and weak ties predicted lower levels of loneliness.

Discussion

Although older adults engage in both in-person and phone contact when they feel lonely, it appears that only in-person contact may reduce loneliness. Digital contact was not widely adopted as a response to momentary loneliness among these older adults. Findings underscore older adults’ willingness to maintain regular contact with close ties. Interventions addressing older adults who are lonely may consider innovative approaches to increase in-person contact.

Loneliness is an aversive experience that arises from a perceived discrepancy between desired and actual interpersonal experiences (Perlman & Peplau, 1982). Loneliness is a subjective feeling and is different from social isolation which is an objective state of lacking contact (Pinquart & Sörensen, 2001). A study synthesized results from 75 longitudinal studies covering different life stages suggested that loneliness is relatively stable across the lifespan (Mund et al., 2020), yet a recent coordinated analysis of 9 longitudinal studies found a curvilinear pattern for loneliness across adulthood, with an uptick in late life (Graham, 2024). In fact, in one survey, 43% of community-dwelling older Americans (aged 65+) reported feeling lonely (National Academies of Sciences, Engineering, and Medicine, 2020), likely due to a shrinking social network and reduced mobility in later life (Antonucci et al., 2014; Moeyersons et al., 2022).

Loneliness may elicit a desire to seek social contact. Older adults engage in different modes of contact to help assuage their loneliness including in-person, phone, and digital contact (e.g., text, email, social media; Faverio, 2022). Phone contact and digital contact may help individuals overcome geographical barriers and facilitate social engagement when in-person contact is not possible (e.g., physical conditions that constrain mobility; Lin & Lachman, 2021; Xie et al., 2021). Yet, it is not known whether phone or digital contact is associated with older adults’ loneliness in similar ways as in-person contact (Antonucci et al., 2017; Fingerman, Birditt, K. S., & Umberson, 2020). Some studies reveal that only in-person contact meets older adults’ emotional needs and improves their well-being (Fingerman et al., 2021; Macdonald et al., 2021), whereas other studies find that digital contact also may lessen older adults’ loneliness (Czaja, 2017; Quintana et al., 2018).

The benefits of different forms of contact may vary depending on relationship closeness. Socioemotional Selectivity Theory suggests that older adults prioritize emotional fulfillment in their relationships, typically favoring close social ties over less close/weak ties (Carstensen, 2021). Digital communication may be suitable for developing new ties and expanding social networks, but in-person and phone contact may allow for greater intimacy and maintaining close ties (e.g., families and friends; Haythornthwaite, 2002; Hülür & Macdonald, 2020).

Loneliness may arise throughout the day (i.e., momentary loneliness) and dissipate with certain modes of contact with different social partners (Zhaoyang et al., 2022). In sum, this study asks whether (a) momentary feelings of loneliness lead to more in-person, phone, and/or digital social contacts, (b) different modes of social contact mitigate loneliness in older adults’ daily lives, and (c) such associations vary by the type of social ties (close vs weak).

Mode of Contact and Loneliness

Momentary Loneliness and Preferences for Different Modes of Contacts

Momentary loneliness may motivate people to seek social contact to fulfill a desired state of social involvement (Archer Lee et al., 2022), and research has documented this phenomenon in young adults (Reissmann et al., 2021). On the other hand, loneliness may elicit self-doubt and fear of social rejection, which leads to less social contact (Layden et al., 2018). A study of adults of all ages found that those who felt lonely were more likely to be solitary and less likely to engage in social activities at subsequent assessments (Arpin et al., 2015). Another study using ecological momentary assessments (EMA; surveys throughout the day) reported the within-person level effect that older adults reported fewer social interactions following feelings of momentary loneliness at a prior time point (Zhaoyang et al., 2022).

Importantly, the inconsistent findings in prior studies may result from a mix of different modes of contact. For older adults, whether to engage in social contact via certain modes may depend on the availability of social ties and comfort with different modes of contact. The Social Compensation Hypothesis suggests that phone and online communication may provide an alternative mode of contact when in-person contact is not available (Toma, 2022). Older adults may experience chronic conditions, mobility issues, or geographically distant social contacts that make in-person contact difficult. Phone and digital communication may help overcome the physical barriers that make it difficult to stay connected with social partners.

Familiarity and comfort with different technologies are important factors to consider as well. The Senior Technology Acceptance Model emphasizes perceived ease of use and perceived value as major factors to consider for technology adoption (Chen & Chan, 2014). Some older adults are not confident with their internet skills and are concerned about the lack of deep conversations online (Hargittai et al., 2019). Compared to digital communication, older adults may find phone communication more comfortable, yet some older adults with hearing loss have difficulty with phone communication (Shukla et al., 2020; Vlachantoni et al., 2023). As such, most older adults prefer and feel most comfortable with in-person contact (Gell et al., 2015), though phone contact remains a viable option. Importantly, although theories regarding older adults’ modes of social contact offer valuable context in understanding older adults’ perceptions and ease of use of various modes of contact, these theories mostly did not explicitly address intraindividual (i.e., within-person) fluctuations throughout the day. The current study would contribute to the literature by examining the co-occurrence of loneliness and social contact for a given individual over time. We hypothesize that when older adults experience momentary loneliness, they may be more likely to engage in in-person and phone contact than when they are less lonely, but not more digital contact (H1).

Different Modes of Contacts and Benefits for Momentary Loneliness

Different modes of contact may be more or less effective in combatting momentary loneliness. Media Richness Theory suggests that in-person contact carries more verbal and nonverbal (e.g., physical contact, facial expression) information compared to other forms of contact (Ishii et al., 2019). In-person contact provides opportunities for emotionally meaningful communication (Keltner et al., 2019). In contrast, phone and digital communication are limited to a certain set of cues that are either verbal or language based (Ishii et al., 2019). The lack of cues and information may impede individuals from deep conversations that are important for older adults’ needs for emotional fulfillment and connectedness which would combat loneliness. Further, although high-quality social contact benefits older adults’ emotional well-being and reduces loneliness, it is also necessary for individuals to spend time themselves (i.e., solitude) for self-compassion (Hoppmann & Pauly, 2022). Undesirable social contact and social contact that are not emotionally meaningful may not contribute to reduced loneliness. Empirical studies found that the emotional benefits from phone and digital contact may be minimal compared with in-person contact (Hülür et al., 2023; Zhaoyang et al., 2022). In other studies, phone and digital contact have even been associated with worse emotional well-being or increased loneliness (Fingerman, Birditt, & Umberson, 2021; Lin & Lachman, 2021).

As such, we hypothesize that more in-person contact would be associated with lower loneliness throughout the day, but more phone and digital contact would not (H2). See Supplementary Figure 1.

Close and Weak Social Ties

Notably, Media Multiplexity Theory underlines the importance of the type and strength of social ties in the context of loneliness (Haythornthwaite, 2002). Specifically, new communication technologies may be more appropriate for enlarging social networks, developing new social ties, and reconnecting with dormant ties. In contrast, older adults may engage in traditional methods (i.e., in-person, phone) to stay connected with close ties (Sumner et al., 2021), as these traditional modalities encompass more visual and vocal cues and are well suited for emotional, expressive, and complicated communications with close ties. We hypothesize that the association between momentary loneliness and in-person and phone contact described in H1 would be stronger for contact with close ties, compared to weak ties (H3a).

Similarly, as in-person contact with close ties (vs weak ties) may be more emotionally beneficial for older adults’ well-being, we hypothesize that the association between in-person contact and momentary loneliness in H2 would be stronger for contact with close ties, compared to contact with weak ties (H3b).

The Current Study and Other Factors Associated With Mode of Contact and Loneliness

Retrospective reports of social contact among older adults are subject to memory errors (Yu et al., 2016). In this study, older adults completed EMA surveys regarding social contact and loneliness at regular intervals throughout the day.

This study adjusted for other factors associated with modes of contact and loneliness: gender, age, marital status, education, health, ethnic or racial minority status, social network size, and time of day. Older men are more likely to use the internet than older women (Gell et al., 2015), yet women report larger network sizes, more social interactions, and lower loneliness (National Academies of Sciences, Engineering, and Medicine, 2020). Among older adults, the usage of new technologies decreases with age, and the association between online communication and psychological well-being depends on age (Yu et al., 2016). Compared to unmarried older adults, married older adults tend to have larger social networks and more in-person contact (Ng et al., 2022). Older adults with higher levels of educational attainment have better digital literacy and use of digital communication (Oh et al., 2021). People with worse physical conditions face barriers to in-person contact, as well as digital contact, and feel lonelier (Yang & Lee, 2022). Ethnic and racial minoritized older adults are less likely to engage in digital communication (Gell et al., 2015). Older adults with larger social networks report less loneliness yet may be sensitive to momentary isolation (Goldman & Compernolle, 2023). On the assessment level, time of day may be related to older adults’ daily social activities (e.g., more likely to have social contact in the afternoon versus morning; Stallings et al., 2022).

The current study explores different modes of social contact in late life, how these modes of contact are associated with older adults’ loneliness, and whether such links differ by the type of social ties. We hypothesized that:

   

Method

Samples and Procedures

The study will use data from The Daily Experiences and Well-being Study (DEWS) collected in 2016–2017 in the greater Austin area and involving 333 community-dwelling older adults aged 65 and older. Participants were recruited via listed landline numbers and random digital dialing using city area codes. The study oversampled in areas with a high density of racial/ethnic minorities and lower socioeconomic status (SES) populations; the overall sample was racially and ethnically diverse and included a wide range of SES. Although 56% of the participants had a college degree or higher, Austin is a highly educated city, where 43% of older adults aged 65 and older had a college degree in 2017 (U.S. Census Bureau, 2017).

Participants initially completed a 2-hr face-to-face baseline interview in which they reported on their social network, physical health, and demographic characteristics. Participants then completed the EMA in which they reported their social contacts, mode of contact, and loneliness every 3 hr for 5–6 days. The time of EMA survey is contingent on participants’ initial finishing time of the morning survey (e.g., if a participant finished the morning survey at 8 a.m., the first EMA survey would be at 11 a.m.). As such, the survey time was not the same across the participants. The morning survey asks different questions and was not included in the current study. Supplementary Figure 2 illustrates the study procedures. EMA information was recorded offline on the Android device that was provided to the participants. Participants returned the Android device after the EMA period and submitted the surveys.

Of the 333 older adults who participated in the initial interview, 313 of them completed the EMA and were eligible for the current study, and 20 were excluded from the current analysis due to lack of the EMA information. Participants were provided with a $50 honorarium for completing the baseline interview and $100 for the EMA data collection. The analytic sample included 313 older adults (M_age = 73.94, standard deviation [SD]_age = 6.38). They completed 5.31 days of EMA on average, with 20.01 EMA assessments per person on average. Compared to the excluded participants (n = 20), the current sample was less likely to be racial or ethnic minorities (X² = 7.19, p = .007), but did not differ regarding other demographic characteristics (Table 1).

Baseline Interview Measures

Social partners

Participants listed up to 30 close social partners using the social convoy measure (Antonucci, 1986; Birditt et al., 2020). The measure asked participants to name people who are close and important to them in three concentric circles. Participants listed 15.10 social partners on average (SD = 6.96, range = 0–30). We then transferred up to the top 10 closest social partners’ names to EMA surveys, during which participants answered questions about their social contact with these close social partners.

Participants’ characteristics

Participants reported their age in years. We coded gender as 1 (female) and 0 (male). Participants reported their education level, recoded as 1 (high school or less), 2 (some college school), and 3 (college or more). Participants reported their marital status as married, cohabitating/living with a partner, divorced, separated, widowed, and never married, recoded as 1 (married or cohabitating) and 0 (not married). Self-reported physical health was rated 1 (poor), 2 (fair), 3 (good), 4 (very good), and 5 (excellent; Idler & Kasl, 1995). Participants indicated their race as White, Black or African American, Asian, American Indian/Alaska Native, and native Hawaiian/other Pacific Islander. They also indicated their ethnicity as Hispanic/Latino and not Hispanic/Latino. Minority status was coded as 1 (ethnic or racial minorities) and 0 (non-Hispanic White).

EMA Measures

Social contacts

The top 10 closest social partners listed in the social convoy were transferred into a handheld device used for the EMA. Participants reported whether they had social contact with each of the 10 closest partners every 3 hr throughout the day. These social encounters were counted as contact with close ties. The study also asked participants if they had contact with up to six additional people every 3 hr which was coded as contact with weak ties.

Regarding mode of contact, if participants indicated an encounter, they also reported if the encounter was (a) in-person, (b) by phone, or (c) by text/email/social media. Numbers of social encounters using each type of contact with each type of tie (close, weak, combined) in the prior 3 hr were generated, resulting in nine continuous variables (e.g., number of in-person contact with close ties, in-person contact with weak ties, and in-person contact with any ties). Because participants could have more than one type of contact at each 3-hr assessment, we treated these as distinct variables rather than as one categorical variable. Participants on average completed 3.83 assessments on social contact each day (SD = 1.40, range = 1–6).

Loneliness

Participants rated their loneliness using one item from 1 (not at all) to 5 (a great deal) from the Positive and Negative Affect Schedule (Watson et al., 1988) every 3 hr throughout the day. This single-item measure of loneliness has been used in previous EMA studies (Fingerman et al, 2021; Reissmann et al., 2021). Participants on average completed 3.74 assessments on momentary loneliness each day (SD = 1.43, range = 1–6).

Assessment-level covariates

We generated dummy variables indicating if the assessment happened in (a) morning (6 a.m. to 11:59 a.m.), (b) afternoon (noon to 6:59 p.m.), and (c) evening and bedtime (7 p.m. to 5:59 a.m.).

Analytic Strategy

We examined the descriptive statistics and bivariate correlations between the variables of interest. To test whether the momentary feeling of loneliness predicts different types of social contacts (H1), we estimated multilevel models with 3-hr assessments (Level 1) nested within participants (Level 2). Three-level models (assessments nested within days, within participants) did not yield different results with two-level models and thus were not reported for parsimonious reasons. We examined the lagged association between momentary loneliness at one 3-hr assessment and the social contacts in the next 3-hr assessment. Models for in-person, phone, and digital contacts were estimated separately. Specifically, we substituted the missing values in each person’s first measurement occasion of the lagged variable with zero and created a dummy variable indicating if the measurement was the first occasion or not. This method allowed us to retain the first measurement occasion and avoid unnecessary listwise deletion (McNeish & Matta, 2020). To isolate the within-person effect from the between-person effect, we calculated the person mean of the lagged loneliness and the lagged loneliness centered on the person mean. The models controlled for time of day (morning vs afternoon vs evening, Level 1), age, gender, race and ethnicity, marital status, education, health, and social network size (Level 2). All models described later used the same approach to separate within- and between-person effects and controlled for these background variables. Results from random intercept only and random slope models showed the same pattern. As such, we reported the results from the random intercept-only models.

Next, we estimated multilevel models again with 3-hr assessments (Level 1) nested within participants (Level 2) to examine the associations between different types of social contacts on loneliness in the same assessment (H2). We estimated separate models for the three types of social contacts and then estimated a combined model with all three types of social contacts together. The models controlled for loneliness at the prior assessment (i.e., lagged loneliness) to account for the autoregression of loneliness.

Finally, to investigate the associations between different types of social contacts with close and weak ties and loneliness (H3), we estimated multilevel models with 3-hr assessments (Level 1) nested within participants (Level 2). The time-varying predictors were the number of encounters that participants had, including (a) in-person contact with close social partners; (b) phone contact with close social partners; (c) digital contact with close social partners; (d) in-person contact with weak social partners; (e) phone contact with weak social partners; (f) digital contact with weak social partners, respectively, in the prior 3 hr. Equations for all models are included in Supplementary Materials. All analyses were performed using Stata 17.

Results

On average, participants completed 20.01 EMA assessments (SD = 6.03, range = 1–32), with 3.83 surveys each day (SD = 1.40, range = 1–6). Participants’ average score on loneliness was 1.15 (SD = .47). On average, participants reported having in-person, phone, and digital contact among 76%, 32%, and 15% assessments, respectively. Participants had 2.03, 0.59, and 0.27 in-person (SD = 2.19, range = 0–16), phone (SD = 1.21, range = 0–14), and digital (SD = .79, range = 0–10) contact, respectively, on average at each assessment. Participants were more likely to report contact with close ties than weak ties for all three modes of contact at each assessment. Table 1 summarizes the descriptives. See Supplementary Table 1 for bivariate associations.

Bidirectional Associations Between Momentary Loneliness and Social Contact

Multilevel models revealed that, on the within-person level, momentary loneliness was associated with more phone contact in the next 3 hr (B = 0.08, p < .001) but not in-person (B = 0.01, p = .86) and digital contact (B = 0.02, p = .15). Yet after controlling for time of day (morning vs afternoon vs evening), the association between momentary loneliness and phone contact was not significant (B = 0.05, p = .053; Supplementary Table 2).

Multilevel models predicting loneliness revealed more in-person contact was associated with lower levels of loneliness at the 3-hr assessment (i.e., within-person) level (B = −0.01, p < .001), yet phone (B = 0.01, p = .20) and digital contact (B = −0.00, p = .66) were not (Table 2). That is, older adults who had more in-person contact in the prior 3 hr felt less lonely.

Close and Weak Ties

Regarding types of social ties (close vs weak), momentary loneliness was linked to more in-person (B = 0.10, p < .001) and phone (B = 0.06, p < .001) contact with close ties, yet was associated with less in-person contact with weak ties (B = −0.09, p = .01; Supplementary Table 3). The associations were not significant anymore after controlling for time of day. In-person contact with close (B = −0.01, p = .02) and weak ties (B = −0.01, p < .001) were both associated with lower loneliness (Table 3). Yet phone and digital contact with either type of social ties were not associated with loneliness in the same 3 hr.

Sensitivity Tests

In-person contact oftentimes happens outside the home and via outdoor activities, which have also been shown to mitigate loneliness. As such, we estimated the models that examined the bidirectional associations between social contact and loneliness, controlling for whether participants left home or not. Results showed that, on the within-person level, leaving home was associated with more in-person contact (B = 2.05, p < .001), yet higher loneliness was associated with fewer in-person contact (B = −0.11, p = .034). In other words, participants had more in-person contact if they left home but reported fewer in-person contact than their average if they felt lonelier in the prior 3 hr. There were no within-person associations between momentary loneliness and phone or digital contact. Additionally, in the models predicting loneliness, leaving home was not associated with lower loneliness. The models suggested the same pattern as the main analyses, such that only in-person contact was associated with lower momentary loneliness (B = −0.01, p < .001), but not phone and digital contact.

Discussion

This study examined the bidirectional associations between momentary loneliness and social contact. Momentary loneliness was associated with more phone contact in the next 3 hr, yet not with in-person and digital contact. But with regard to the benefits of contact, only in-person contact was associated with lower loneliness; phone and digital contact were not. These patterns differed by the type of relationship. Older adults engaged in more in-person and phone contact in response to loneliness but benefited only from more in-person contact.

Frequency of Different Modes of Social Contact

Older adults reported more phone contact if they felt lonelier in the previous 3 hr, yet not more in-person and digital contact. The Social Compensation Theory suggests that when in-person contact is not available, phone and digital contact may be used as an alternative (Toma, 2022). Additionally, older adults feel relatively comfortable with phone contact, yet sometimes hesitate to use digital media (Vlachantoni et al., 2023). A recent study examining older adults’ attitudes toward modes of contact found that older adults did not rate in-person contact as more favorable compared to phone contact with regard to valence (pleasant vs unpleasant), social relatedness (connection), and calmness. In some cases, they rated phone contact as more meaningful than in-person contact. Yet digital contact was rated lower on meaning, valence, and social relatedness consistently (Hülür et al., 2023). Frameworks for cultivating trust in emerging technologies suggest that adoption and trust thrive from individuals’ affective and analytic reactions toward the technology (Lee & See, 2004). As such, with doubts of the deepness and meaningfulness of digital contact, older adults may not feel comfortable having digital contact even if they feel lonely.

Notably, associations between momentary loneliness and social contact did not remain consistent after adjusting for time of day (morning, afternoon, and evening). This result suggests that individuals’ daily activities may strongly adhere to preset schedules and routines (Stallings et al., 2022) with momentary feelings contributing only a minor portion of fluctuations in daily social contact.

In sum, findings illustrate the possibility of using phone contact to overcome barriers to in-person contact in late life, especially when people feel lonely and have the need for social connection. Yet digital contact was not commonly adopted among older adults in response to momentary loneliness.

Different Modes of Contact and Benefits for Momentary Loneliness

Different modes of contact showed different degrees of effectiveness in mitigating loneliness. Older adults who had more in-person contact in the previous 3 hr felt less lonely compared to their personal average. These findings are consistent with literature examining how different modes of contact combat loneliness in late life (Macdonald et al., 2021; Zhaoyang et al., 2022), by highlighting the unique effect of in-person contact on older adults’ loneliness. A study collected data from older participants for 21 days and found that older adults reported higher positive affect, lower negative affect, and lower loneliness on days that they had more in-person contact (Macdonald et al., 2021). Similarly, another study tracked older adults’ social contact for 14 days and found that having in-person contact was associated with lower loneliness, yet digital contact was not (Zhaoyang et al., 2022). Likewise, the current study emphasizes the importance of in-person contact in later adulthood. By examining participants’ loneliness and social contact at 3-hr intervals, we captured the vicissitudes of loneliness that occur concurrently with social contact throughout each day.

Combining the results, although phone contact is available at most times and it provides older adults with opportunities for social connections when they feel lonely, it appears that older adults may not benefit from phone contact with regard to loneliness. Importantly, within-person analyses highlighted that only in-person contact was shown to be associated with reduced levels of loneliness. Research has been discussing whether phone and digital contact would be able to substitute in-person contact (Antonucci et al., 2017; Fingerman, Birditt, & Umberson, 2020) and our study showed that phone and digital contact do not provide older adults with the same emotional closeness and comfort as in-person contact.

Close and Weak Social Ties

When older adults were lonely at one time, they tended to report more in-person and phone contact with close ties at the next time point. Conversely, they reported less in-person contact with weak ties when they felt lonely than when they felt less lonely. Socioemotional Selectivity Theory posits that older adults value emotionally close social ties more than weak ties (Carstensen, 2021). Our findings are consistent with this theory, such that older adults appear to prioritize close ties when they wish to reach a desired state of social connectedness and alleviate loneliness. Although weak ties also serve important roles in individuals’ daily lives and may provide opportunities for various activities (Fingerman, 2009; Fingerman, Huo et al., 2020; Huxhold et al., 2020), close ties may be more important for older adults’ emotional well-being, especially when they feel lonely and need companionship.

Results also showed that in-person contact with both close and weak ties was associated with lower loneliness. Considering the bidirectional associations, it is interesting that although weak in-person contact was linked to lower loneliness, older adults themselves still prefer interacting with close ties when they feel lonely. Given the distinctive role that in-person contact with close ties plays on emotional well-being, the findings offer practical implications such that it is critical for older adults to cultivate close local ties that are readily available in their daily lives (York Cornwell & Goldman, 2021). Those lacking access to in-person contact may face increased risks of loneliness and social isolation (National Academies of Sciences, Engineering, and Medicine, 2020). Community-based programs targeting loneliness in later life should consider facilitating emotionally meaningful contact and strengthening existing local social ties (Greenhalgh et al., 2016).

Limitations and Future Directions

The current study did not differentiate age and cohort effect. During the coronavirus disease 2019 (COVID-19) pandemic, people relied on technology to stay connected with each other. Older adults in their 60s and 70s increased the use of digital media (Freedman et al., 2022), though older adults in their 80s decreased the use of these media (perhaps due to lack of technical support; Vlachantoni et al., 2023). The current sample born in 1930s to 1950s grew up in times when digital contact was not widely adopted.

Due to the study design and participant burden, it was not possible for the current study to assess the effects of social contact from a prior 3-hr assessment without accounting for social contact during the concurrent 3 hr. Future studies may consider implementing an event-based study design or a more finely tuned time structure during intensive data collection to examine the casual association between social contact and loneliness.

Additionally, we need to be cautious when generalizing the current results to other older adults from different backgrounds and different living situations (e.g., education, mobility). Specifically, participants were from the greater Austin area with higher education levels than the national average (U.S. Census Bureau, 2017). Over half of the participants had college degrees or more and may have had more positive views of digital media (Oh et al., 2021). Given the lack of access to internet in rural areas (Reddick et al., 2020), findings from this study conducted in an urban area may involve greater use of digital media than in other populations. Importantly, the sample in the current study reported a relatively low level of loneliness, yet the results may differ in other populations that face greater risks of loneliness such as older adults with fewer resources or with limited mobility. An internet-based intervention was effective in improving well-being among low-income and home-bound older adults (Choi et al., 2020), though additional research is warranted given findings in this study only supported in-person contact to mitigate loneliness. Finally, this study was conducted prior to the 2020 COVID pandemic, and rates of digital communication, particularly video conferencing, may have increased among older adults since that time (Xie et al., 2021).

In conclusion, the current study points to the bidirectional associations between loneliness and social contact, with evidence these patterns differ for contact with close and weak ties. Findings revealed older adults’ preference of in-person and phone contact with close ties and the distinct benefits of in-person contact in diminishing loneliness. This study also found that although digital device use has been more prevalent recently, older adults did not seek digital contact in response to loneliness. Further, digital contact may not be an effective approach to alleviate loneliness in late life. Together, this study provides practical insights for interventions and programs focused on loneliness to consider strategies that facilitate meaningful social contact to offer emotional benefits.

Funding

This research was supported by grants from the National Institute on Aging: (R01AG046460 to K. L. Fingerman, PI; and P30AG066614 to the Center on Aging and Population Sciences at The University of Texas at Austin), as well as a grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (P2CHD042849 to the Population Research Center at The University of Texas at Austin). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflict of Interest

None.

Data Availability

The current study was not preregistered. Data and information about the study are available at https://www.icpsr.umich.edu/web/NACDA/studies/38570.

References