The Role of Daily Calls to Telephone Companionship Lines for Older Adults Experiencing Chronic Loneliness

Abstract

This study uses mixed-methods to understand the experiences of daily callers to a telephone-based companionship line, the Institute on Aging Friendship Line, for loneliness. Baseline interviews of Friendship Line participants (n=63) were conducted between October 1st, 2020 to April 30th, 2021, with follow-up interviews at 3 and 6 months. We tested the association of call volume (daily vs non-daily) with chronic loneliness, defined as scoring 6+ points on 3-item UCLA Scale (Range: 3-9 points) at two or more time points. In addition, we analyzed qualitative interviews of a purposive sample (n=23) thematically. Daily callers were more likely to be chronically lonely than non-daily callers (61% vs 18%, χ² (1) = 11.5, p<0.001) and had similar rates of persistent anxiety (33% vs 13%, χ² (1) = 3.3, p=0.07) and depression (22.2% vs 8.9%, χ² (1) = 2.1, p=0.15). Although daily callers reported high levels of chronic loneliness, they reported high satisfaction with and meaningful relationships through the Friendship Line (daily: 83% vs non-daily: 61%, χ² (1) = 2.9, p=0.09). Daily callers described the importance of a long-term, evolving relationship with the Friendship Line, whereas non-daily callers described the Friendship Line as an outlet for acute social needs, similar to an “urgent care.” Taken together, results suggest the Friendship Line is an important source of support for older adults managing both chronic loneliness and occasional loneliness.

Graphical Abstract

In this mixed-methods study of a telephone-based support line for loneliness among older adults, participants who called the line daily were more likely to be experiencing chronic loneliness than non-daily callers. Despite high levels of loneliness, daily callers high satisfaction and the importance of long-term support from the Friendship Line, suggesting that benefits of telephone-based support may extend to people experiencing chronic loneliness.

INTRODUCTION

Chronic loneliness is defined as a state of emotional distress due to feeling alone or isolated that persists for an extended period, typically months to years.¹ Unlike transient or occasional loneliness, which can often motivate people to reconnect with their social lives,^2,3 chronic loneliness is thought to more profoundly impact both mental and physical health through long-term effects on the immune system, sleep, and inflammatory pathways.^4,5 Indeed, the longer an individual experiences loneliness, the greater their risk of cardiovascular diseases, cognitive decline,⁶ depressive disorders,⁷ and, ultimately, mortality.^4,8,9 Consequently, there is a need for interventions which address chronic loneliness.

The Friendship Line, created in 1973, is a nationally available and accredited telephone support service for older adults. Interventions like the FL were originally designed to address occasional loneliness of older adults through brief, 10-minute phone calls.¹⁰ However, in the course of a larger study on the Friendship Line, a subset of participants were identified who were calling the Friendship Line daily and tended to have persistently high loneliness levels on standardized loneliness scales over time. Our goal was to use mixed-methods to report on this group of daily callers experiencing chronic loneliness and explore how FL was supporting their needs in comparison to those who did not require daily calls.

METHODS

A convenience sample of N=63 participants aged over 55 who called into the Friendship Line were surveyed between October 1st, 2020, and April 30th, 2021 with a baseline interview and follow-ups at 3-months and 6-months. Trained research volunteers called interested participants over the phone to track psychosocial metrics, use of the Friendship Line, and satisfaction with the service. The overall goal of the study was to understand how telephone-based lines were supporting older adults in a time of heightened loneliness over multiple months of the COVID-19 pandemic. The study received approval from the University of California, San Francisco Institutional Review Board (IRB: # 20-30420).

Participants were categorized as daily or non-daily callers based on self-reported frequency of calling. Loneliness was assessed using the 3-item UCLA Loneliness Scale (Range 3-9 points),¹¹ with “chronic loneliness” defined as scoring 6+ points at two or more time points. Symptoms of depression and anxiety were assessed using the Patient Health Questionnaire 2-item (PHQ-2) Depression Screen and Generalized Anxiety Disorder 2-item (GAD-2) Anxiety Screen, respectively.^14,15 We used bivariate statistics (chi-square tests and student t-tests) to describe differences between daily versus non-daily callers across experiences of chronic loneliness, psychological well-being, and demographics. Analysis was conducted using Stata 16 statistical software.

In addition, 23 participants were recruited from the above study for semi-structured, in-depth interviews between February 18, 2022 to May 17, 2022 for a parent study to understand how the structure and implementation of the Friendship Line uniquely impacts social well-being. Interviewees were purposively selected from the parent study and additional participants were recruited based on age, sex, race and ethnicity, and primary language (English or Spanish), with a goal to include diverse perspectives since prior literature on telephone companionship lines had primarily included perspectives of English-speaking or White/Caucasian participants.^16–20 Thematic analysis was initially conducted for the parent study by three researchers using an inductive approach; thematic saturation was achieved when no new codes emerged.²¹ A secondary qualitative analysis of the interviews was then conducted using a primarily deductive approach focused on the study goals of understanding differences in perspectives on the Friendship Line among daily and non-daily callers. First, two researchers (DE and AK) read through all interview transcripts to ensure understanding of the context of stories and to check for additional codes relevant to the secondary analysis not previously captured. Next, the frequency of codes was visualized for daily versus non-daily callers to determine major thematic differences in experiences. In particular, we focused on groups of codes that were absent for one group versus common for the other. Illustrative quotes are shown in Table 2.

Table 2.

Open-Ended Interviews on Friendship Line’s Role in Participants’ Lives

Theme(s)	Illustrative Quotations
Friendship Line offers support through longitudinal relationships	Shift from support with specific need to general social support: “I rely on it, like I rely on food and water … You can’t believe how much Friendship Line helped me through my grieving process, which took me at least five years … now, I look more forward to it for like just like I called up one of my friends for real. <laughs> You know, I may not know them, but it feels like a friendship”. (age 69, Daily)a
	Support with persistent social stressor: “But I call the friendship line because I have nobody to talk to. I try to talk to different people because I only have one son with me that lives with me right now. He’s 64, but he’s temporary and I live by myself most of the time since my wife left” (age 87, Daily)
Friendship Line offers accessible support as needs arise	Urgent social needs or crises: “There have been nights when I’ve been here by myself and I’ve prayed and I said to God, “I’ve got to talk to somebody,” so then that’s when I call … I mean if I didn’t have (FL) I would go (to the ER) more often.” (age 69, Non-Daily) “I mean if I didn’t have (FL) would I go (to the ER) more often(…) Whereas with the Friendship Line, I can just get on the phone if somebody talks to me, terrific…” (age 71, Non-Daily)
	Addressing occasional needs by complementing existing coping strategies: “It used to be every day or sometimes about twice a day, but now I just use it every month, every other month… I just felt I didn’t need to call them as often. I didn’t need to talk to anyone else, because when I am in despair now what I do is I pray” (age 63, Non-Daily) “I mean I try not to lean heavily on anything, but it does help because I’m literally pretty much alone, and it’s my choice in many ways because I don’t want to go out … but it’s like having somebody to talk to even if it’s about the weather … you know, anything. Like I’ll call my older relatives way out in different areas like Arizona, Texas, New Mexico (…) But sometimes I just don’t want to call them at a weird hour or, you know, it’s good to have somebody to talk to” (age 64, Non-Daily)

RESULTS

Sample characteristics stratified by daily versus non-daily callers within the parent (Quantitative) study and the in-depth interviews (Qualitative) are presented in Table 1. Of the participants who called daily at any time-point (n=18), 55% identified as female, 50% had regular access to a computer, and 66% reported having at least three medical conditions. Amongst non-daily callers (n=45), 60% identified as female, approximately 29% had regular access to a computer, and 60% reported at least three medical conditions. Between the two populations, daily callers had non-significantly higher rates of self-reported fair/poor vision (daily: 39% vs non-daily: 24%, χ² (1) = 1.3, p=0.25) and significantly higher self-reported fair/poor hearing (daily: 44% vs non-daily: 13%, χ² (1) = 7.2, p=0.007). Individuals who received in-depth qualitative interviews (N=23) were 65% female, 35% White, 26% Non-White Hispanic/Latino, 17% Black, 4% Native American, and 17% multi-ethnic or of another race or ethnicity. Of these participants, 35% called the FL daily.

Table 1.

Characteristics of study participants

		Quantitative		Qualitative
		Non-Daily (45)	Daily (18)	Non-Daily (15)	Daily (8)
Characteristics		N (%)	N (%)	N (%)	N (%)
Age Groups	55-65	17 (37.8%)	6 (33.3%)	4 (26.7%)	4 (50.0%)
	66-75	17 (37.8%)	10 (55.6%)	7 (46.7%)	3 (37.5%)
	75+	11 (24.4%)	2 (11.1%)	4 (26.7%)	1 (12.5%)
Gender	Female	27 (60.0%)	10 (55.6%)	13	2 (25.0%)
	Male	16 (35.6%)	8 (44.4%)	2 (13.3%)	6 (75.0%)
	Identity not listed	2 (4.4%)	-	-	-
Race/Ethnicity	Black/African American	-	1 (5.6%)	2 (13.3%)	2 (25.0%)
	Latinx	3 (6.7%)	-	3 (20.0%)	3 (37.5%)
	Native American	1 (2.2%)	1 (5.6%)	1 (6.6%)	-
	White/Caucasian	36 (80.0%)	15 (83.2%)	6 (40.0%)	2 (25.0%)
	Multi-ethnic	1 (2.2%)	1 (5.6%)	2 (13.3%)	-
	Other	4 (8.9%)	-	1 (6.6%)	1 (12.5%)
Relationship Status	Married/Living with a partner	7 (15.6%)	2 (11.1%)	1 (6.6%)	1 (12.5%)
	Divorced/Separated	16 (35.6%)	8 (44.4%)	5 (33.3%)	3 (37.5%)
	Widowed	5 (11.1%)	1 (5.6%)	2 (13.3%)	2 (25.0%)
	Never married	16 (35.6%)	6 (33.3%)	7 (46.7%)	2 (25.0%)
Financesa	Usually some money left over	9 (20.0%)	6 (33.3%)	4 (26.7%)	2 (25.0%)
	Just enough to make ends meet	23 (51.1%)	7 (38.9%)	4 (26.7%)	5 (62.5%)
	Not enough to make ends meet	9 (20.0%)	5 (27.8%)	5 (33.3%)	-
	Not sure	4 (8.9%)	-	2 (13.3%)	1 (12.5%)
Education	<High School	5 (11.1%)	3 (16.7%)	2 (13.3%)	1 (12.5%)
	High School/GED	7 (15.6%)	1 (5.6%)	2 (13.3%)	2 (25.0%)
	Some college or post-high school vocational school	14 (31.1%)	7 (38.9%)	2 (13.3%)	2 (25.0%)
	College degree or more	19 (42.2%)	7 (38.9%)	9 (60.0%)	2 (25.0%)
Access to the Internet?	No	15 (33.3%)	5 (27.8%)	-	-
Regular Use of Devices	Computer	13 (28.9%)	9 (50.0%)	-	-
	Smartphone	16 (35.6%)	8 (44.4%)	-	-
	Tablet	8 (17.8%)	5 (27.8%)	-	-
	Cell phone	22 (48.9%)	6 (33.3%)	-	-
Multi-morbidity (self-reported)	+3 medical conditions	27 (60.0%)	12 (66.7%)	-	-
	1-2 medical conditions	13 (28.9%)	4 (22.2%)	-	-
	No medical conditions reported	5 (11.1%)	2 (11.1%)	-	-
Self-rated Hearingb	Fair/Poor	6 (13.3%)	8 (44.4%)	-	-
Self-rated Visionb	Fair/Poor	11 (24.4%)	7 (38.9%)	-	-
Difficulty with Activities of Daily Living	Difficulty with +2 ADLs	12 (26.7%)	5 (27.8%)	-	-
	Difficulty with 1 ADL	9 (20.0%)	5 (27.8%)	-	-
	No difficulty with ADLs reported	24 (53.3%)	8 (44.4%)	-	-

^aFinancial stress was determined by asking “In general, how do your finances usually work out at the end of the month?”

^bSelf-rated hearing and vision impairment was rated in 5-categories: Excellent, Very good, Good, Fair, and Poor.

We next compared psychosocial health measures between daily and non-daily callers. Daily callers had higher rates of screening positive for depression (56% versus 33%, χ² (1) = 2.7, p=0.1), loneliness (89% vs 73%, χ² (1) = 1.8, p=0.18), and anxiety (78% vs 44%, χ² (1) = 5.75, p=0.02) at any time point compared to non-daily callers. In examining participants’ psychosocial states over time (Figure 1), daily callers were more likely to be chronically lonely than non-daily callers (61% vs 18%, χ² (1) = 11.5, p<0.001) and had similar rates of persistent anxious symptoms (33% vs 13%, χ² (1) = 3.3, p=0.07) and persistent depressive symptoms (22.2% vs 8.9%, χ² (1) = 2.1, p=0.15). Although daily callers reported high levels of loneliness, they more often reported that “by participating in the Friendship Line, I have made a meaningful relationship or friendship” (daily: 83% vs non-daily: 61%, χ² (1) = 2.9, p=0.09).

Figure 1. Psychosocial Measures of Non-Daily and Daily Callers.

^a Persisting depression was defined by screening positive for depression on the PHQ-2 at any two timepoints

^b Persisting anxiety was defined by screening positive for anxiety on the GAD-2 at any two timepoints

^c Chronic loneliness was defined by screening positive for loneliness on the 3-item UCLA Loneliness Scale at any two timepoints

^d P-values were determined from chi-square tests between psychosocial measures and frequency of calling

Qualitative findings are presented in Table 2. Our overarching finding concerned a fundamentally different view of the Friendship Line between daily and non-daily callers. Daily callers described the importance of a long-term, evolving relationship with the Friendship Line. Callers were able to process both unexpected distressing events and challenges inherent to daily life. For example, one daily caller explained a shift from seeking help with a specific mental health challenge to a general need to talk with someone who cares:

“When I first started calling … it was more like about my experience with [my mental health challenge]… but now things are… like a need <laughs>, a need to talk to somebody, and now it’s like it’s excitement about the future, and upcoming events” (64M, daily)

This need for daily calls could also stem from a chronic or persistent stressor, typically the loss or complete lack of social network ties. For example, one daily caller noted the loss of family, friends and neighbors, and that daily calls made a difference:

“It gave me somebody to talk to you for 10 minutes every afternoon. And it made a difference (…) And the main thing was just having a human voice to communicate with. I mean I would talk to my daughter once or twice, but that was it. There was nobody else.” (81F, Daily)

In contrast, non-daily callers emphasized the Friendship Line’s use as an outlet for acute needs or crises. For example, one caller likened the service to an urgent care:

“Friendship Line … was like … going to an urgent care where right now you got to talk to someone … especially when I was living with-- in my brother’s house. I was … so stressed, so overwhelmed and really traumatized … I had nothing. I had just a suitcase. … And so at that time I needed urgent care and that’s what Friendship Line was for me.” (72M, Non-Daily)

Callers also noted that interactions with the Friendship Line may have replaced their need to go to physical urgent care clinic or emergency department (Table 2). Occasional, non-urgent needs were also addressed, with callers noting that it helped to be able to reach out at any time of day, even for a few minutes, as a way of complementing existing coping strategies and social support:

“From what I noticed through my experience with the Friendship Line, the primary purpose is to provide a space for a fellow human to report, or request, a few minutes-- oh my god, I’m going to cry. Report or request a few minutes of time from another human” (83F, Non-Daily)

DISCUSSION

In this mixed-methods study, we demonstrated that individuals with persistently high or chronic loneliness who frequently call the Friendship Line had high satisfaction and perceived benefits from the service. To our knowledge, this study is the first to report on how interventions may differ in their benefit for acute versus chronic loneliness. As depicted in our conceptual framework (Figure 2), loneliness can arise from an interplay of risk factors (e.g socioeconomic status, health, environment) and stressors (e.g. bereavement, relocation), which may trigger acute episodes. These episode may or may not develop into chronic loneliness through persistent stressors or a cycle of maladaptive social cognition. In both cases, findings are suggestive that the Friendship Line’s dual structure as both a support line and crisis hotline allow the service to uniquely tackle loneliness when it is an acute crisis and when it becomes a chronic experience—providing immediate relief for some and sustained psychological and relational support for others.

Figure 2.

Conceptual model for telephone companionship line addressing acute or short-term loneliness and chronic or long-term loneliness

We draw attention to two major study findings and their implications for future practice and research. First, results demonstrate that the Friendship Line was able to meet a variety of social needs of its callers, despite not being specifically designed for high-frequency callers and those experiencing chronic loneliness. This complements prior literature suggesting short-term benefits for telephone lines, while adding to the body of work by showing the longitudinal support callers may gain through these interventions.^10,16 Participants had a fundamentally different view of the Friendship Line based on their call frequency. Less frequent callers likened it to an accessible source of support for urgent needs or occasional needs to complement existing coping strategies. Some even noted that using the line helped in avoiding unnecessary emergency department visits. In contrast, daily callers described a shift from receiving targeted help for specific issues (e.g. grief) to experiencing the line as a daily source of general emotional connection. Notably, our data is suggestive that the Friendship Line is highly accessible, in comparison to virtual platforms which may not be available to all older adults due to barriers such as economic constraints, lack of training, or limited access to the internet. For example, a large proportion of daily callers had sensory impairment (vision and hearing loss), who may struggle with virtual platforms. We hypothesize that the ease of engaging with telephone lines, particularly for an older population experiencing significant multimorbidity and functional impairment, allows telephone lines to meet at least part of the social needs of chronically lonely older adult participants. Future trials of loneliness interventions using telephone companionship might specifically target older adults experiencing chronic loneliness.

Second, participants who called daily had higher rates of chronic loneliness, yet also reported more frequently that they have made a meaningful relationship through the Friendship Line. Consequently, findings suggest that a mixed methods approach is critical to loneliness intervention evaluations. If evaluations only include quantitative measures of loneliness, one may miss identifying benefits for people experiencing chronic or persistent loneliness where quantitative scales remain high over time. Relatedly, measures such as the UCLA 3-point Loneliness Scale are useful for screening purposes, but may have ceiling effects which limit their ability to capture changes in loneliness, which have variable inputs from partnership, friendship, community engagement, and broader experiences of belongingness and meaning. Longer scales would be beneficial in addressing different dimensions of loneliness, and further research is needed on sensitive methods or scales to identify changes in severity or disruptions to daily life over time.

Our study has limitations that warrant consideration. First, we used a convenience sample that included fewer participants of different ethnicities, of oldest old ages, or of the LGBTQ+ community which limits generalizability. Larger and more diverse population in future work can help validate these findings or be used in future studies of interventions. Second, use of a brief quantitative measure of loneliness as opposed to multiple or longer measures limited our ability to measure change over time due to ceiling effects. Future studies could measure additional dimensions of loneliness using longer or varied scales, or measure the impacts on other domains of health such as physical health, psychological health, or health care use. Third, while we categorized participants as daily versus non-daily callers based on self-reported call frequency, our response options did not capture exact call counts, and so individuals calling near-daily may have interpreted response options differently. This may have introduced some misclassification and limits our ability to explore finer gradations of call frequency.

In conclusion, we provide preliminary evidence that the Friendship Line was able to serve older adults experiencing loneliness stemming from both acute and chronic sources. Our data suggest that studies of interventions should account for the persistent nature of loneliness in some communities which may not necessarily indicate an absence of benefit for social interventions.

Data Availability:

The authors confirm that the data supporting the findings of this study are available within the article.