Loneliness in Older Primary Care Patients and Its Relationship to Physical and Mental Health-Related Quality of Life

Abstract

Background:

Loneliness is a significant public health challenge in the United States, especially among older adults. The epidemiology of loneliness among older adults in primary care is lacking and specific research is needed on how loneliness impacts older primary patients’ physical, mental, and cognitive health. A large sample of older primary care patients were recruited for a trial during the COVID-19 pandemic to measure the relationship between loneliness and physical and mental quality of life (QOL).

Methods:

Baseline data come from the Caregiver Outcomes of Alzheimer’s Disease Screening (COADS) study, an ongoing randomized controlled trial evaluating benefits and risks of Alzheimer’s disease and related dementias screening among primary care patients ages 65 and older, collected April 2020 to September 2021. Loneliness was measured with the 5-item, Loneliness Fixed Form Ages 18+ from The NIH Toolbox Emotion Battery, physical and mental health-related QOL was measured with the SF-36v2, and depression and anxiety severity were measured with the PHQ-9 and GAD-7, respectively.

Results:

Spearman correlation analyses revealed that loneliness was moderately correlated with mental health QOL (r(601) = −.43, p<0.001), anxiety severity (r(601) =.44, p<0.001), and depression severity (r(601) = .42, p<0.001), while weakly correlated with physical health QOL (r(601) = −.15, p<0.001). After conducting unadjusted and adjusted linear regression models, we found that loneliness was significantly associated with both lower mental (p<0.001) and physical (p<0.001) QOL. Furthermore, loneliness remained significantly associated with worse mental QOL after adjusting for age, gender, race, ethnicity, educational level, perceived income status, neighborhood disadvantage, severity of comorbidities and comorbid depression and anxiety.

Conclusion:

Primary care providers should discuss loneliness with their older adult patients and provide resources to help patients develop and maintain meaningful social relationships.

Background

Loneliness, as a social construct, is defined as the difference between an individual’s desired and actual social relationships.¹ Loneliness is more than just a feeling; it is a “significant biopsychosocial stressor” with a mortality risk comparable to smoking more than 15 cigarettes a day and more deleterious than alcoholism, obesity, and physical inactivity.^2,3 Furthermore, loneliness in the US is common, with recent estimates of its prevalence at over one-quarter of older adults ≥65 years.^4,5 This is particularly worrisome as loneliness is linked to chronic physical health conditions including cardiovascular disease, stroke, hypertension, and all-cause mortality,^3,6 and its influence on mental and cognitive health outcomes, including depression and Alzheimer’s Disease and related dementias, is especially potent.^7–9 Given the profound consequences and prevalence of loneliness, the 2023 United States Surgeon General Advisory declared loneliness an urgent public health issue worthy of the “same investments” researchers and policy makers have made in addressing tobacco use, obesity, and the addiction crisis.¹⁰

Beyond the relationships between loneliness and physical and mental health conditions, loneliness is also a strong correlate of quality of life (QOL) among older adults, who are disproportionately impacted due to changes in social networks stemming from retirement¹¹ and partner loss through death or divorce.¹² The relationship between older adult loneliness and QOL in the general population has been established.^13,14 However, policy makers have identified a need to assess this relationship in public health¹⁵ and healthcare¹⁶ settings because of their access to and ability to treat those most vulnerable to loneliness.

In 2020, the National Academies of Sciences, Engineering, and Medicine (NASEM) released a consensus study report stressing the health effects of social isolation and loneliness.¹⁶ The committee concluded that loneliness was associated with increased contact with healthcare and therefore, highlighted it as an important setting to address the phenomenon. Currently, research assessing the relationship of loneliness and QOL in older adults focuses on community-dwelling older adults with less attention on clinical populations such as primary care patients. To our knowledge, there is only one study investigating the relationship between loneliness and multiple components of QOL among older adult primary care patients, but this study exclusively focused on sicker older adult patients. ¹⁴ The researchers found that severe and moderate loneliness was significantly associated with lower physical and mental components of QOL. Though this finding is insightful, this study is not generalizable to all (e.g., healthier) older adult primary care patients. Additionally, COVID-19 led to an increase in loneliness among older adults¹⁷, yet studies exploring the relationship between loneliness and QOL among this population during the pandemic are lacking. Finally, studies of loneliness and QOL often consider the conceptual overlap between these measures and mental health conditions^13,14, but we identified no studies which assess the individual and combined effects of both depression and anxiety. Using secondary data from a large randomized controlled trial, our objective was to determine the relationship between loneliness and QOL in older adult primary care patients during COVID-19.

Methods

Study Design & Participants

Data come from the ongoing Caregiver Outcomes of Alzheimer’s Disease Screening (COADS) study, a randomized controlled trial measuring the benefits, risks, and harms of Alzheimer’s disease and related dementia (ADRD) screening.¹⁸ Older primary care patients were recruited from Eskenazi Hospital and Indiana University (IU) primary care clinics. Eskenazi Health is the third largest safety net (essential) health care system in the United States and is responsible for the care of economically marginalized and uninsured/under-insured patients living in Indianapolis, Indiana. The IU Health system is an academic integrated care delivery network with 9 primary care clinics located throughout urban and rural Indiana. Patients were enrolled in person from their primary care office or via telephone, where they provided informed consent to participate in the study. Inclusion criteria for patients included: 1) 65 years or older, 2) had at least one visit to their primary care physician in the past 12 months, 3) could provide informed consent, 4) were able to communicate in English, and 5) had no previous or existing diagnosis of Alzheimer’s disease, related dementia, or mild cognitive impairment. Patients were excluded from the study if they were in a permanent nursing facility, had a serious mental illness (such as bipolar disorder or schizophrenia), or had been given a prescription for a cholinesterase inhibitor or memantine. The final analytical sample for this study includes 603 older adult primary care patients recruited during COVID-19 from March 11, 2020 through September 30, 2021. Study approval was granted by the Institutional Review Board at Indiana University.

Measures

Loneliness was measured using the validated 5-item, Loneliness Fixed Form Ages 18+ from The NIH Toolbox Emotion Battery.¹⁹ Patients were asked to report how often they experienced different feelings of loneliness within the past month via Likert scale (1 = never, 5 = always). Scores range from 5–25, with higher scores indicating higher perceived loneliness.

Quality of Life was assessed by the Short Form Health Survey-Version 2 (SF-36v2), a 36-item questionnaire measuring components of health-related QOL such as physical, mental, and social functioning.^20,21 Further, the SF-36 can be divided into eight domains of health-related QOL: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. The scores from the eight subscales were aggregated using standard scoring to create two summary scores: the physical component summary (PCS) and the mental component summary (MCS). Total scores range from 0–100, with higher scores indicating a more favorable state of health. The SF-36 exhibits psychometric properties and is a valid, reliable measure of older adult QOL.^22–24

Covariates

Anxiety and depression severity were measured by General Anxiety Disorder-7 (GAD-7) and the Patient Health Questionnaire-9 (PHQ-9), respectively. Among older adults, both the GAD-7 and PHQ-9 show evidence of reliability and validity.^25,26 The GAD-7 is a 7-item scale assessing anxiety severity, with scores ranging from 0–21.²⁷ The PHQ-9 is a 9-item scale assessing depression severity, with scores ranging from 0–27.²⁵ From these scales, we created two categorical measures of anxiety severity and depression severity where scores 0–4= none/minimal, 5–9= mild, 10–14=moderate, and 15+=severe. We also created a categorical measure of mental health comorbidity: neither (0–4 or “none-minimal” for both), anxiety only (5+ or “mild, moderate or severe”), depression only (5+ or “mild, moderate or severe”), or both.

Patient neighborhood disadvantage was measured using the Area Deprivation Index (ADI).^28–30 The ADI is a score calculated using 17 social determinants of health measurements such as income, employment, education, and housing information which were derived from 2009–2013 data from a large national study.³¹ Further, ADI was calculated for each neighborhood in the U.S. by a research team at the University of Wisconsin and is currently available for download from the Neighborhood Atlas.²⁸ Scores range from 1–100, with higher scores indicating more socioeconomic disadvantage.

Patient comorbidity was measured by the Charlson Comorbidity Index (CCI).³² The CCI calculates mortality risk by taking a weighted score from 19 medical conditions. Scores range from 0–16, with higher scores indicating greater number of medical comorbidities and higher mortality risk.

We also collected sociodemographic characteristics including age, gender, race, ethnicity, education, and perceived income.

Analysis

We used SAS 9.4 (SAS Institute, Cary, NC) for all analyses. We conducted a Spearman correlation assessing the relationship between physical (PHRQoL) and mental health-related (MHRQoL) quality of life (as measured by the SF-36 PCS and MCS, respectively), loneliness, anxiety severity, and depression severity. Accordingly, we ran unadjusted and adjusted linear regression models to assess the relationships between loneliness and patient PHRQoL and MHRQoL. All adjusted linear regression models controlled for patient age, gender, race, ethnicity, education, perceived income, neighborhood deprivation, and comorbidity. In addition, we conducted a second regression model adjusting for categorical measures of patient depression severity (PHQ-9) and anxiety severity (GAD-7), and a third regression model, instead adjusting for depression and anxiety comorbidity.

Results

Table 1 shows the baseline characteristics of the 603 older primary care patients in our analytic sample. Patients had an average loneliness score of 7.2 (range=5–21), with 52.9% reporting any loneliness (≥6) and 47.1% reporting “never” feeling lonely on all five items (=5). Patients reported an average PHRQoL of 47.0 (range=7.7–61.2) and MHRQoL of 55.3 (range=24–66.7).

Table 1.

Baseline Characteristics of the Sample (N = 603 Patients)

Variables	Range	N (%)	Mean (SD)
Age (years)	65–95		73.5 (5.1)
Gender
Male		270 (44.8)
Female		333 (55.2)
Race
White		551 (91.5)
Black		45 (7.5)
Other		6 (1.0)
Ethnicity
Hispanic		8 (1.3)
Non-Hispanic		594 (98.7)
Education
High School or Less		85 (14.1)
Some College/Degree		308 (51.1)
Post-graduate		210 (34.8)
Perceived Income
Just/not enough to make ends meet		507 (84.5)
Comfortable		93 (15.5)
Neighborhood Disadvantage a	7–99		44.2 (21.6)
Comorbidity b	0–16		2.3 (2.6)
Anxiety Severity c	0–17		1.7 (2.5)
None		536 (88.9)
Mild		54 (9.0)
Moderate		9 (1.5)
Severe		4 (0.7)
Depression Severity d	0–21		3.0 (3.4)
None		457 (75.8)
Mild		113 (18.7)
Moderate		25 (4.2)
Severe		8 (1.3)
Comorbid
Depression and Anxiety
Neither		442 (73.3)
Anxiety only		94 (15.6)
Depression only		15 (2.5)
Both		52 (8.6)
Loneliness	5–21		7.2 (3.0)
Quality of Life
Physical Health-related QOL	7.7–61.2		47.0 (10.1)
Mental Health-related QOL	24–66.7		55.3 (6.8)

^aNeighborhood Disadvantage as measured by the Area Deprivation Index (ADI)

^bComorbidity as measured by the Charlson Comorbidity Index (CCI)

^cAnxiety Severity as measured by the Generalized Anxiety Disorder-7 (GAD-7)

^dDepression Severity as measured by the Patient Health Questionnaire-9 (PHQ-9)

Among covariates, 55.2% were female; 91.5% White; 98.7% non-Hispanic; with an average age of 73.5 years old. Just over half reported completing at least some college or having a college degree (51.1%) and 84.5% of patients had a perception of their income as “comfortable,” with an average neighborhood disadvantage score of 44.2. The sample was relatively healthy with a comorbidity index of 2.6, and among mental health conditions, 88.9% had none or minimal anxiety, 75.8% had none or minimal depressive symptoms, and 73.3% had neither depression nor anxiety.

Spearman correlations of patient variables of interest are presented in Table 2. We found a moderate negative correlation between patient PHRQoL and depression severity (r(df) = −0.41, p<0.001). Patient MHRQoL was found to be negatively correlated with patient anxiety severity (r(df) = −0.51, p<0.001), depression severity (r(df) = −0.49, p<0.001), and loneliness (r(df) = −0.43, p<0.001). Patient anxiety severity was found to be positively correlated with depression severity (r(df) = 0.57, p<0.001) and loneliness (r(df) = 0.44, p<0.001). Further, patient depression severity was found to be positively correlated with loneliness (r(df) = 0.42, p<0.001).

Table 2.

Spearman Correlations of Variables of Interest

	PHRQoL	MHRQoL	Anxiety	Depression	Loneliness
Physical Health Related Quality of Life a	1.00	−0.10*	−0.15***	−0.41***	−0.15***
Mental Health Related Quality of Life b		1.00	−0.51***	−0.49***	−0.43***
Anxiety Severity c			1.00	0.57***	0.44***
Depression Severity d				1.00	0.42***
Loneliness e					1.00

^*p<0.05

^**p<0.01

^***p<0.001; weak correlation=0.1–0.3, moderate correlation=0.4–0.6

^aPhysical Health Related Quality of Life (PHRQoL) as measured by the Short Form Health Survey-Version 2 (SF-36v2) Physical Component Score (PCS)

^bMental Health Related Quality of Life (MHRQoL) as measured by the Short Form Health Survey-Version 2 (SF-36v2) Mental Component Score (MCS)

^cAnxiety Severity as measured by the Generalized Anxiety Disorder-7 (GAD-7)

^dDepression Severity as measured by the Patient Health Questionnaire-9 (PHQ-9)

^eLoneliness as measured by the 5-item, Loneliness Fixed Form Ages 18+ from The NIH Toolbox Emotion Battery

Three adjusted linear regression models demonstrate the association between patient loneliness and PHRQoL as measured by SF-36 PCS (Table 3). All three regression models control for patient age, gender, race, ethnicity, perceived income status, neighborhood disadvantage, and comorbidity. We found that loneliness was associated with lower PHRQoL (p=0.001) when adjusting for only patient sociodemographic characteristics. However, this significance between patient loneliness and PHRQoL was not observed in model two when adjusting for depression and anxiety severity (p=0.166), and in model three when adjusting for depression and anxiety comorbidity (p=0.104). Across models, no demographic factors (age, gender, race, ethnicity) were associated with PHRQoL, but having a post graduate degree (p=0.005), reporting a comfortable income (p=0.015, 0.048, 0.036, respectively), and comorbidity (p<.001) were positively associated with PHRQoL in all three models. In model two, mild (p<0.001), moderate (p<0.001), and severe (p=0.001) depression was associated with lower PHRQoL, while mild anxiety was associated with higher PHRQoL (p<0.001). In model three, patients with depression only (p<0.001) or both depression and anxiety (p=0.015) were associated with having a lower PHRQoL.

Table 3.

Linear Regression Results for Physical Health Related Quality of Life

	Model 1		Model 2		Model 3
	b (SE)	p-value	b (SE)	p-value	b (SE)	p-value
Loneliness	−0.45 (0.13)	0.001	−0.19 (0.14)	0.166	−0.23 (0.14)	0.104
Age	−0.05 (0.08)	0.510	−0.07 (0.07)	0.347	−0.06 (0.07)	0.398
Female	−0.90 (0.81)	0.263	−0.81 (0.77)	0.294	−0.56 (0.77)	0.470
Race
White	Ref.		Ref.		Ref.
Black	0.21 (1.52)	0.888	−0.13 (1.45)	0.931	−0.11 (1.45)	0.940
Other	0.71 (3.91)	0.856	0.35 (3.71)	0.926	0.14 (3.73)	0.971
Hispanic	−2.90 (3.35)	0.388	−2.89 (3.19)	0.364	−2.55 (3.21)	0.427
Education
≤HS or Less	Ref.		Ref.		Ref.
College Degree	0.81 (1.19)	0.498	0.84 (1.14)	0.460	0.90 (1.14)	0.432
Post-graduate	3.59 (1.28)	0.005	3.45 (1.22)	0.005	3.49 (1.23)	0.005
Comfortable Income	2.78 (1.14)	0.015	2.16 (1.09)	0.048	2.29 (1.09)	0.036
Neighborhood Disadvantage a	−0.02 (0.02)	0.399	−0.01 (0.02)	0.780	−0.01 (0.02)	0.688
Comorbidity b	−0.95 (0.16)	<0.001	−0.82 (0.15)	<0.001	−0.83 (0.15)	<0.001
Anxiety c
None			Ref
Mild			4.98 (1.45)	0.001
Moderate			5.29 (3.28)	0.107
Severe			−2.32 (5.44)	0.670
Depression d
None			Ref
Mild			−7.45 (1.05)	<0.001
Moderate			−8.92 (2.07)	<0.001
Severe			−13.73 (3.97)	0.001
Comorbid Depression and Anxiety
Neither					Ref.
Depression Only					−8.03 (1.08)	<0.001
Anxiety Only					2.67 (2.39)	0.265
Both					−3.53 (1.45)	0.015
R 2	15.6%		24.6%		23.6%

^aNeighborhood Disadvantage as measured by the Area Deprivation Index (ADI)

^bComorbidity as measured by the Charlson Comorbidity Index (CCI)

^cAnxiety Severity as measured by the Generalized Anxiety Disorder-7 (GAD-7)

^dDepression Severity as measured by the Patient Health Questionnaire-9 (PHQ-9)

Three adjusted linear regression models demonstrate the association between patient loneliness and MHRQoL as measured by SF-36 MCS (Table 4). All three regression models control for patient age, gender, race, ethnicity, perceived income status, neighborhood disadvantage, and comorbidity. We found that loneliness was associated with lower MHRQoL across all models, even when controlling for both severity of (Model 2) and comorbid depression and anxiety (Model 3). None of the sociodemographic characteristics were associated with MHRQoL across all three models, and comorbidity was only associated with MHRQoL in Model 1 (p=0.137), before adjusting for depression and anxiety severity in Model 2 (p=0.137) and depression and anxiety comorbidity in Model 3 (p=0.134). We found that patient depression and anxiety severity were associated with lower MHRQoL in Model 2 (p<0.001 for both). When removing depression and anxiety severity as covariates in Model 3, we found that having depression only (p<0.001), anxiety only (p=0.002), or both depression and anxiety (p<0.001) were associated with lower MHRQoL.

Table 4.

Linear Regression Results for Mental Health Related Quality of Life

	Model 1		Model 2		Model 3
	b (SE)	p-value	b (SE)	p-value	b (SE)	p-value
Loneliness	−1.01 (0.08)	<0.001	−0.46 (0.07)	<0.001	−0.53 (0.08)	<0.001
Age	0.06 (0.05)	0.239	0.05 (0.04)	0.174	0.05 (0.04)	0.199
Female	−0.04 (0.51)	0.936	−0.47 (0.41)	0.253	−0.22 (0.44)	0.616
Race
White	Ref.		Ref.		Ref.
Black	−0.97 (0.97)	0.318	−0.79 (0.78)	0.130	−1.02 (0.83)	0.220
Other	0.91 (2.49)	0.714	0.61 (2.00)	0.813	0.40 (2.13)	0.852
Hispanic	−1.87 (2.13)	0.381	−2.39 (1.71)	0.164	−2.30 (1.84)	0.210
Education
≤HS or less	Ref.		Ref.		Ref.
College Degree	0.59 (0.76)	0.436	−0.08 (0.61)	0.894	−0.14 (0.65)	0.827
Post-graduate	−0.18 (0.82)	0.821	−0.92 (0.66)	0.163	−1.02 (0.70)	0.146
Income Comfortable	0.20 (0.72)	0.786	−0.63 (0.59)	0.281	−0.64 (0.62)	0.303
Neighborhood Disadvantage a	−0.01 (0.01)	0.540	−0.01 (0.01)	0.337	−0.02 (0.01)	0.141
Comorbidity b	−0.21 (0.10)	0.037	−0.12 (0.08)	0.137	−0.13 (0.09)	0.134
Anxiety c
None			Ref.
Mild			−5.20 (0.78)	<0.001
Moderate			−8.43 (1.76)	<0.001
Severe			−12.82 (2.92)	<0.001
Depression d
None			Ref.
Mild			−2.53 (0.56)	<0.001
Moderate			−9.12 (1.11)	<0.001
Severe			−13.44 (2.13)	<0.001
Comorbid Depression and Anxiety
Neither	-	-	-	-	Ref.
Depression Only	-	-	-	-	−3.12 (0.62)	<0.001
Anxiety Only	-	-	-	-	−4.35 (1.37)	0.002
Both	-	-	-	-	−11.96 (0.83)	<0.001
R 2	22.8%		50.9%		43.6%

^aNeighborhood Disadvantage as measured by the Area Deprivation Index (ADI)

^bComorbidity as measured by the Charlson Comorbidity Index (CCI)

^cAnxiety Severity as measured by the Generalized Anxiety Disorder-7 (GAD-7)

^dDepression Severity as measured by the Patient Health Questionnaire-9 (PHQ-9)

Discussion

Our findings build upon the literature by assessing the relationship between loneliness and quality of life in a primary care setting. More specifically, our study provides insights into the prevalence of loneliness among older adult primary care patients during COVID-19, and the implications loneliness has on their physical and mental health-related of quality of life. Among our sample, 52.9% of patients reported experiencing any loneliness, just slightly lower than the 55% prevalence reported in other primary care settings.¹⁴

We found that loneliness was associated with both lower PHRQoL and MHRQoL, echoing findings by others that both components of quality of life are significantly reduced by loneliness.¹⁴ Specifically, our study shows that loneliness reduces PHRQoL when adjusting for sociodemographic factors and patient comorbidity, but the relationship is no longer significant when adjusting for mental health conditions. Similarly, Musich et al.¹⁴ found that both severe and moderate loneliness reduces physical components of QOL by 9% and 5%, respectively. Though their study showed an enduring relationship in fully adjusted models, their sample focused on older, sick adults (as defined by care management program eligibility) and did not adjust for anxiety symptomology. Our study is unique in that it assessed the relationship between loneliness and PHRQoL among a general population of older adult primary care patients, but others have documented the effects of loneliness on related outcomes such as functional health^6,33 and sedentariness.^34,35

Though loneliness was associated with both physical and mental health related quality of life, the strength of the relationship was stronger for loneliness and MHRQoL—the relationship endures even when adjusting for comorbidity and mental health conditions. Others have reported similar findings, reporting between 9% and 24% reductions in mental health components of QOL due to loneliness.^14,36 Though few studies have investigated the relationship between loneliness and MHRQoL, others have documented an association with similar outcomes such as psychological well-being.^37,38

Among covariates, we found that depression and anxiety were consistently associated with lower MHRQoL, and comorbidity was associated with PHRQoL across all models. Although these results were anticipated, we also found that mild anxiety was also associated with higher PHRQoL. It is likely that this is a spurious relationship due, instead, to the relationship between mild anxiety and personality traits, such as neuroticism and conscientiousness, and in turn, their influence on health. Some individuals with neurotic tendencies can be anxious and especially watchful of their health, and conscientiousness has been linked to health-protective behaviors, including exercise, abstention from substances, and consumption of a lower fat diet.^39,40 Indeed, in their study of older, primary care patients, Chapman and colleagues,³⁹ found that higher conscientiousness was associated with higher SF-36 role physical scores and freedom from IADLs. Therefore, being both conscientious and “vigilantly anxious” may be especially health protective for physical health related quality of life.⁴⁰

Finally, we also found that having a post-graduate degree and reporting a comfortable level of income was associated with higher PHRQoL in older primary care patients. Similarly, Shankar et al.⁴¹ found that loneliness was associated with functional status, as measured by gait speed, and that the relationship was stronger among those with lower objective wealth. Studies have also found a relationship between loneliness and more subjective measures of SES, including income discomfort.⁴² Researchers posit that lower SES can have both direct and indirect influences on loneliness.⁴³ Those with more education tend to have better and more diverse opportunities for social interaction and tend to have higher income; those with lower actual or perceived income are less likely to utilize commercial social opportunities and are less able to return support provided by others.^42,43 Furthermore, low SES can be associated with reduced self-esteem, which can inhibit the search for new social contacts.⁴³ Our findings, echoed by others, suggest that older adults with lower socioeconomic status could especially benefit from targeted interventions at reducing loneliness.

Several limitations to our study need to be acknowledged. First, our sample was limited to one geographic region of the United States. Although we recruited from two demographically diverse health systems across 14 sites statewide, our results are not generalizable to the US population. Another limitation to our study is the lack of racial and ethnic diversity of our sample, which mirrors a similar trend in health care utilization. For primary care, in particular, evidence suggests that Black and Hispanic older adults are more likely than their White counterparts to delay routine medical care until more acute care becomes necessary.⁴⁴ Though premature mortality and access factors may play a small role, Dunlop and colleagues⁴⁵ recently found that older, minority adults have less contact with health care providers, even when adjusting for predisposing factors, measures of need, and economic access. It is likely that factors such as patient beliefs about the medical system, care-seeking behaviors, and bias among health care providers contribute to unequal primary care utilization. Finally, as a cross sectional study, we cannot establish causation between loneliness and quality of life—it is possible that lower quality of life precedes loneliness. Indeed, studies have demonstrated a significant reciprocal relationship between loneliness and QOL related measures including subjective well-being⁴⁶ and physical functioning.⁶ Similarly, studies have demonstrated cyclical relationships between mental health symptomologies, comorbidities, and quality of life that are difficult to disentangle³³. Though we considered these factors, reverse causality is plausible.

This topic is more relevant than ever given the May 2023 U.S. Surgeon General’s call to action to tackle the loneliness epidemic. Feeling lonely in 2023 is now less common than during the height of the pandemic, but rates remain substantially higher than during pre-pandemic times with over one third of older adults reporting loneliness as of January 2023.⁴⁷ Studies hypothesize that this may be due, in part, to factors such as smaller social networks due to “pruning” during the pandemic and to post-pandemic health-related social anxiety.⁴⁸

Addressing loneliness in the primary care setting is necessary because it provides an important point of contact for older adults, especially for those who are lonely—older adults visit primary care between 5 and 6 times a year, on average, and those who are lonely average one additional visit per year.⁴⁹ Further, studies have found that older adults generally appreciate discussing social risks with their doctor⁵⁰ and report no discomfort with such screenings.⁵¹

Identifying patients who report higher levels of loneliness will indicate those in need of future intervention, but loneliness is rarely diagnosed and can be difficult to intervene on. In their meta-analysis of interventions to reduce loneliness, Masi and colleagues⁵² identified four primary intervention strategies: improving social skills, enhancing social support, increasing opportunities for social contact, and addressing maladaptive social cognition (e.g., cognitive behavioral therapy), the latter proving most successful. One effective intervention, The Circle of Friends, is a three-month group-based, psychosocial rehabilitation model was aimed to enhance interaction and friendships between participants that has been shown to be effective in both reducing loneliness and improving health outcomes including subjective health, lower health care costs, cognition, and mortality.^53–55 Primary Care Physicians should be aware of local resources and community partners available for referral for similar programs.

Conclusions

Loneliness is a common phenomenon among older adults, and even more so since the onset of the global COVID-19 pandemic. This study provides empirical evidence that loneliness is associated with both physical and mental aspects of quality of life. Despite the deleterious effects of loneliness, interventions to address the discrepancy between desired and actual social interaction are few and limited. Given the potential to increase quality of life and other outcomes including countless physical health conditions, ADRD, and early mortality, assessing loneliness in primary care settings is necessary and is a promising first step in providing targeted interventions to address this significant public issue.

Key Points:

• Nearly 53% of older, primary care patients reported loneliness during the COVID-19 pandemic.

• Loneliness is associated with both physical and mental health related quality of life and the association with the latter remains after adjusting for both anxiety and depression.

• Primary care practitioners have increased contact with lonely older adults and are uniquely poised to measure and help address the lingering loneliness pandemic in the United States that is associated with a host of poor health outcomes.

Why Does this Paper Matter?

The United States Surgeon General Advisory recently declared loneliness an urgent public health issue and have urged researchers and policy makers into action. Older adults frequently visit their primary care provider for routine health check-ups; therefore, monitoring older adult loneliness in primary care is pertinent for understanding how loneliness is related to overall health-related quality of life. This study illustrates the prevalence of loneliness among older, primary car patients and delineates it’s influence on both physical and mental health-related quality of life.