Association between social isolation and increased arterial stiffness in older adults: a cross-sectional study

Abstract

Social isolation is associated with adverse physical and mental health outcomes. However, its relationship with arterial stiffness, a marker of vascular aging and cardiovascular risk, remains unclear. This study aimed to determine the association between social isolation and arterial stiffness in patients aged 55 years or older attending a cardiovascular outpatient clinic. This cross-sectional study included 357 patients (mean age: 72.0 ± 7.6 years; 204 men). Social isolation was assessed using the Lubben Social Network Scale-6 (LSNS-6), and arterial stiffness was assessed using brachial-ankle pulse wave velocity (baPWV). The LSNS-6 total score was significantly and negatively correlated with baPWV (r = -0.19, p < 0.001). This association remained significant after adjustment for potential confounders (β = -0.146, p = 0.002). Patients with an LSNS-6 total score < 12 had a significantly higher prevalence of arterial stiffening (baPWV ≥ 1800 cm/s) than did those with an LSNS-6 total score ≥ 12 (46.7% vs. 34.5%; p = 0.04). Logistic regression analysis revealed that an LSNS-6 total score < 12 was significantly associated with increased risk of arterial stiffening (OR, 2.29; 95% CI, 1.21–4.33; p = 0.01). Although no significant sex interaction was observed, the association appeared stronger in men. These findings suggest that addressing social isolation may serve as a potential strategy to reduce cardiovascular risk.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-49495-y.

Introduction

Social isolation generally refers to an objective lack of social contact with others¹. The risk of social isolation increases with age due to life course factors such as retirement and widowhood¹. Therefore, the prevalence of social isolation is increasing in developed countries with aging populations. In addition to population aging, other societal changes, such as shifts in family structure, urbanization, weakening of community ties, and changes in employment patterns, may also contribute to the increasing prevalence of social isolation. In Japan, a repeated cross-sectional study showed that social isolation among older adults became more prevalent from 2010 to 2016, accompanied by an increase in the proportion of those lacking interaction with relatives². Recent research has shown that approximately one-quarter of community-dwelling older adults aged 65 years or older in the United States are socially isolated¹. Social isolation is associated with an increased risk of adverse physical and mental health outcomes, including premature mortality, depression, and cognitive decline^1,3. As a result, social isolation has emerged as a critical public health issue with significant implications for physical and mental health.

Arterial stiffness, which refers to reduced elasticity of the arterial wall, is a key indicator of vascular aging and is an established marker of cardiovascular risk⁴. Arterial stiffness increases with the progression of arteriosclerosis and contributes to target organ damage and the development of cardiovascular disease by decreasing the buffering effect on pulsatile flow generated by cardiac contraction⁵. Recently, measurement of brachial-ankle pulse wave velocity (baPWV) has been widely used in clinical practice due to its simplicity, non-invasiveness, and high reproducibility, and baPWV may serve as a predictive vascular marker for future cardiovascular events^6,7. Although carotid-femoral PWV (cfPWV) is considered as the reference standard for assessing central arterial stiffness, baPWV has been shown to significantly correlate well with cfPWV and is accepted as a valid surrogate marker⁸.

In addition to its association with adverse physical and mental health outcomes, social isolation has also been shown to be associated with an increased risk of cardiovascular disease, including coronary heart disease (CHD) and stroke⁹. Given that arteriosclerosis is a key pathophysiological process underlying cardiovascular disease, elucidating the relationship between social isolation and arterial stiffness may help clarify the mechanisms through which social factors influence the development and progression of cardiovascular disease. Arterial stiffness is a key contributor to the pathogenesis of cardiovascular diseases. However, there is little information on the association between social isolation and arterial stiffness. Social isolation is commonly assessed using structured questionnaires that evaluate an individual’s contact and interactions with family and friends. Among these, the Lubben Social Network Scale-6 (LSNS-6) is a validated screening tool specifically developed to assess social networks and identify individuals at risk for social isolation among older adults^10,11. Therefore, the aim of this study was to determine the association between social isolation, as assessed using the LSNS-6, and arterial stiffness, as assessed using baPWV, in patients aged 55 years or older who were attending a cardiovascular outpatient clinic.

Results

Baseline clinical characteristics

The baseline clinical characteristics of the subjects are summarized in Table 1. Of the 357 subjects, 325 (91.3%) had hypertension, 250 (70.2%) had dyslipidemia, 73 (20.5%) had diabetes mellitus, 155 (44.0%) had chronic kidney disease (CKD), 42 (11.9%) had CHD, 32 (9.0%) had cerebrovascular disease, and 173 (48.7%) were smokers (20 current smokers and 153 former smokers). The mean LSNS-6 total score was 15.0 ± 5.6 (median, 15; interquartile range, 11 to 19; range, 0 to 30) (Supplementary Fig. 1). Of the 357 subjects, 90 (25.2%) had an LSNS-6 total score of less than 12. The percentage of subjects with such a low score was significantly higher in men than in women (30.9% vs. 17.7%, p = 0.004) (Table 1). The mean baPWV was 1722 ± 332 cm/s (range, 1089 to 2986 cm/s).

Table 1.

Clinical characteristics of the subjects.

Variables	All	Men	Women	p value
	(n = 357)	(n = 204)	(n = 153)
Age, y	72.0 ± 7.6	71.5 ± 8.1	72.5 ± 6.8	0.24
Body mass index, kg/m2	23.7 ± 3.4	24.2 ± 3.2	23.0 ± 3.4	< 0.001
Systolic blood pressure, mm Hg	128.1 ± 16.1	127.6 ± 15.8	128.8 ± 16.5	0.48
Diastolic blood pressure, mm Hg	75.3 ± 10.1	76.2 ± 10.1	74.2 ± 10.1	0.07
Heart rate, bpm	67.0 ± 10.3	67.6 ± 10.7	66.2 ± 9.8	0.21
Total cholesterol, mmol/L	4.97 ± 0.96	4.77 ± 0.93	5.25 ± 0.93	< 0.001
Triglycerides, mmol/L	1.49 ± 1.05	1.51 ± 1.04	1.46 ± 1.06	0.65
HDL-cholesterol, mmol/L	1.65 ± 0.44	1.59 ± 0.43	1.74 ± 0.44	0.003
LDL-cholesterol, mmol/L	2.76 ± 0.77	2.61 ± 0.73	2.94 ± 0.78	< 0.001
Glucose, mmol/L	6.14 ± 1.62	6.18 ± 1.59	6.07 ± 1.66	0.55
HbA1c, %	5.8 ± 0.7	5.8 ± 0.7	5.8 ± 0.7	0.65
eGFR, mL/min/1.73 m2	62.0 ± 16.9	62.1 ± 17.8	61.9 ± 15.7	0.92
Smoking, n (%)	173 (48.7)	157 (77.0)	16 (10.6)	< 0.001
Current smoking, n (%)	20 (5.6)	18 (8.8)	2 (1.3)	0.002
Comorbidities
Hypertension, n (%)	325 (91.3)	190 (93.1)	135 (88.8)	0.15
Dyslipidemia, n (%)	250 (70.2)	134 (66.0)	116 (75.8)	0.045
Diabetes mellitus, n (%)	73 (20.5)	48 (23.5)	25 (16.5)	0.10
Coronary heart disease, n (%)	42 (11.9)	38 (18.9)	4 (2.7)	< 0.001
Cerebrovascular disease, n (%)	32 (9.0)	25 (12.4)	7 (4.6)	0.01
Chronic kidney disease, n (%)	155 (44.0)	85 (41.9)	70 (47.0)	0.34
Medication use
Antihypertensive drug, n (%)	304 (85.4)	182 (89.2)	122 (80.3)	0.02
Lipid-lowering drugs, n (%)	183 (51.4)	99 (48.5)	84 (55.3)	0.21
Antidiabetic drugs	55 (15.4)	43 (21.1)	12 (7.9)	< 0.001
baPWV, cm/s	1722 ± 332	1710 ± 314	1737 ± 356	0.46
baPWV ≥1800 cm/s, n (%)	134 (37.5)	73 (35.8)	61 (39.9)	0.43
LSNS-6 total score	15.0 ± 5.6	14.1 ± 5.8	16.1 ± 5.1	< 0.001
LSNS-6 total score < 12, n (%)	90 (25.2)	63 (30.9)	27 (17.7)	0.004

HDL indicates high-density lipoprotein; LDL, low-density lipoprotein; eGFR, estimated glomerular filtration rate; baPWV, brachial-ankle pulse wave velocity; LSNS-6, Lubben Social Network Scale-6.

Relationship between LSNS-6 total score and baPWV

Univariate regression analysis showed that the LSNS-6 total score was negatively correlated with baPWV (r = -0.19, p < 0.001) (Fig. 1). Multiple linear regression analysis also showed that the LSNS-6 total score was negatively associated with baPWV (β = -0.146, p = 0.002) (Table 2). In sex-stratified analyses, there was a significant negative association between the LSNS-6 total score and baPWV in men (β = -0.195, p = 0.001) but not in women (β = -0.002 p = 0.98) (Table 2), although no significant interaction between men and women was observed in the association between LSNS-6 total score and baPWV (p for interaction = 0.19).

Fig. 1.

Scatter plots showing the relationship between the Lubben Social Network Scale-6 (LSNS-6) total score and brachial-ankle pulse wave velocity (baPWV).

Table 2.

Multiple linear regression analyses of the relationships between the Lubben Social Network Scale-6 (LSNS-6) total score and brachial-ankle pulse wave velocity (baPWV).

LSNS-6 total score	baPWV
	All					Men					Women
	β	VIF	Standard error	t value	P value	β	VIF	Standard error	t value	P value	β	VIF	Standard error	t value	P value
Unadjusted	-0.18694	1	3.098494	-3.59	< 0.001	-0.24002	1	3.715943	-3.51	<0.001	-0.14434	1	5.575324	-1.79	0.08
Model 1	-0.20088	1.0334878	3.091919	-3.86	< 0.001	-0.24188	1	3.646517	-3.61	<0.001	-0.1424	1	5.504112	-1.79	0.08
Model 2	-0.1442	1.1840	2.876035	-2.91	0.004	-0.19967	1.195225	3.406166	-3.16	0.002	0.011041	1.208512	5.042329	0.15	0.88
Model 3	-0.14634	1.186839	2.765508	-3.06	0.002	-0.19526	1.1949816	3.139249	-3.35	0.001	-1.00188	1.2211615	5.143272	-0.02	0.98

Unadjusted: The adjusted r² were 0.03 for the analysis including all subjects, 0.06 for men, and 0.01 for women.

Model 1: adjusted for age and sex. The adjusted r² were 0.07 for the analysis including all subjects, 0.09 for men, and 0.04 for women.

Model 2: adjusted for age, sex, body mass index, mean arterial pressure, heart rate, antihypertensive medication use, dyslipidemia, diabetes mellitus, smoking, chronic kidney disease, and a history of cardiovascular disease (coronary heart disease and cerebrovascular disease). The adjusted r² were 0.32 for the analysis including all subjects, 0.38 for men, and 0.37 for women.

Model 3: adjusted for age, sex, body mass index, systolic blood pressure, heart rate, antihypertensive medication use, dyslipidemia, diabetes mellitus, smoking, chronic kidney disease, and a history of cardiovascular disease (coronary heart disease and cerebrovascular disease). The adjusted r² were 0.35 for the analysis including all subjects, 0.44 for men, and 0.31 for women.

P for interaction for men and women was 0.19.

Association between an LSNS-6 total score < 12 and arterial stiffening (baPWV ≥ 1800 cm/s)

The subjects were divided into two groups based on LSNS-6 total scores of ≥ 12 and < 12 (Table 3). The prevalence of arterial stiffening was significantly higher in individuals with an LSNS-6 total score < 12 than in those with an LSNS-6 total score ≥ 12 (46.7% vs. 34.5%, p = 0.04) (Fig. 2). Multiple logistic regression analyses revealed that an LSNS-6 score < 12 was significantly associated with a higher risk of arterial stiffening (odds ratio [OR], 2.29; 95% confidence interval [CI], 1.21 to 4.33; p = 0.01) (Table 4). In sex-stratified analyses, an LSNS-6 total score < 12 tended to be associated with a higher risk of arterial stiffening in both men (OR, 2.27; 95% CI, 0.99 to 5.15; p = 0.05) and women (OR, 1.99; 95% CI, 0.60 to 6.60; p = 0.26), although these associations did not reach statistical significance (Table 4). No significant interaction between men and women was observed in the association between an LSNS-6 total score < 12 and arterial stiffening (p for interaction = 0.62).

Table 3.

Clinical characteristics of the subjects based on a Lubben Social Network Scale-6 (LSNS-6) total score of 12.

Variables	LSNS-6 total socre ≥ 12	LSNS-6 total socre < 12	p value
	(n = 267)	(n = 90)
Age, y	72.0 ± 7.7	71.8 ± 7.2	0.80
Men, n (%)	141 (52.8)	63 (70.0)	0.004
Body mass index, kg/m2	23.7 ± 3.4	23.6 ± 3.3	0.74
Systolic blood pressure, mm Hg	127.6 ± 15.8	129.5 ± 17.0	0.34
Diastolic blood pressure, mm Hg	74.5 ± 9.8	77.7 ± 10.9	0.01
Heart rate, bpm	65.5 ± 9.7	71.4 ± 10.9	< 0.001
Total cholesterol, mmol/L	4.98 ± 0.92	4.96 ± 1.07	0.86
Triglycerides, mmol/L	1.39 ± 0.92	1.77 ± 1.32	0.004
HDL-cholesterol, mmol/L	1.68 ± 0.43	1.58 ± 0.46	0.07
LDL-cholesterol, mmol/L	2.76 ± 0.74	2.75 ± 0.84	0.91
Glucose, mmol/L	6.13 ± 1.69	6.15 ± 1.40	0.91
HbA1c, %	5.8 ± 0.7	5.8 ± 0.8	0.44
eGFR, mL/min/1.73 m2	61.9 ± 15.7	62.2 ± 20.2	0.92
Smoking, n (%)	118 (44.4)	55 (61.8)	0.004
Current smoking, n (%)	14 (5.3)	6 (6.7)	0.60
Comorbidities
Hypertension, n (%)	246 (92.5)	79 (87.8)	0.17
Dyslipidemia, n (%)	184 (69.2)	66 (73.3)	0.46
Diabetes mellitus, n (%)	58 (21.7)	15 (16.9)	0.32
Coronary heart disease, n (%)	36 (13.7)	6 (6.7)	0.08
Cerebrovascular disease, n (%)	29 (11.0)	3 (3.4)	0.03
Chronic kidney disease, n (%)	112 (42.6)	43 (48.3)	0.35
Medication use
Antihypertensive drug, n (%)	236 (88.7)	68 (75.6)	0.002
Lipid-lowering drugs, n (%)	140 (52.6)	43 (47.8)	0.43
Antidiabetic drugs	46 (17.3)	9 (10.0)	0.10
baPWV, cm/s	1691 (322)	1815 (347)	0.002
baPWV ≥1800 cm/s, n (%)	92 (34.5)	42 (46.7)	0.04
LSNS-6	17.4 ± 3.9	7.7 ± 3.1	NA

HDL indicates high-density lipoprotein; LDL, low-density lipoprotein; eGFR, estimated glomerular filtration rate; baPWV, brachial-ankle pulse wave velocity; NA, not applicable.

Fig. 2.

Bar graphs showing the prevalence of arterial stiffening (brachial-ankle pulse wave velocity ≥ 1800 cm/s) in individuals with a Lubben Social Network Scale-6 (LSNS-6) total score ≥ 12 and those with an LSNS-6 total score < 12.

Table 4.

Associations between brachial-ankle pulse wave velocity (baPWV) ≥1800 cm/s and the Lubben Social Network Scale-6 (LSNS-6).

Variable		baPWV ≥1800 cm/s
		Odds ratio (95% Confidence interval); p value
		Unadjusted	Model 1a	Model 2b	Model 3c
LSNS-6 total score < 12*	All	1.66 (1.02–2.70); 0.04	1.80 (1.09–2.98); 0.02	1.82 (0.98–3.38); 0.058	2.29 (1.21–4.33); 0.01
	Men	1.71 (0.93–3.15); 0.09	1.77 (0.95–3.30); 0.07	1.97 (0.89–4.38); 0.09	2.27 (0.99–5.15); 0.05
	Women	1.81 (0.78–4.22); 0.16	1.83 (0.77–4.38); 0.17	1.21 (0.37–3.93); 0.75	1.99 (0.60–6.60); 0.26

^aModel 1: adjusted for age and sex.

^bModel 2: adjusted for age, sex, body mass index, mean arterial pressure, heart rate, antihypertensive medication use, dyslipidemia, diabetes mellitus, smoking, chronic kidney disease, and a history of cardiovascular disease (coronary artery disease and stroke).

^cModel 3: adjusted for age, sex, body mass index, systolic blood pressure, heart rate, antihypertensive medication use, dyslipidemia, diabetes mellitus, smoking, chronic kidney disease, and a history of cardiovascular disease (coronary artery disease and stroke).

*P for interaction for men and women was 0.62.

Discussion

In the present study, we demonstrated that the LSNS-6 total score, an established measure for screening social isolation, was significantly and negatively associated with baPWV as a continuous variable. This association remained significant even after adjusting for potential confounders. In addition, an LSNS-6 total score < 12, indicating a high risk of social isolation, was significantly associated with a higher risk of arterial stiffening (baPWV ≥ 1800 cm/s). Although no statistically significant interaction by sex was observed, the association between LSNS-6 total score and baPWV tended to be stronger in men than in women. To our knowledge, this study is the first study to demonstrate a significant association between social isolation, as assessed using LSNS-6, and arterial stiffness, as assessed using baPWV.

LSNS-6 was developed as an abbreviated version of the original Lubben Social Network Scale to efficiently assess social networks and screen for social isolation, particularly in older adults¹⁰. A lower LSNS-6 total score indicates a greater degree of social isolation, with a total score of less than 12 proposed as a cutoff value for identifying individuals at high risk of social isolation¹⁰. This cutoff value was validated in community-dwelling older adult populations in three European countries. The prevalence of social isolation identified by LSNS-6 in these populations was consistent with the expected rates of 10% − 15% previous reported for community-dwelling older adults. In addition, the cutoff value showed good convergent validity with other established measures of social support and social isolation. A Japanese version of LSNS-6 has also been developed and validated in community-dwelling Japanese individuals aged 55 years or older, supporting its utility in screening for social isolation in older Japanese adults¹¹. Based on these findings, the LSNS-6 was adopted in this study to assess social isolation in individuals aged 55 years or older.

Previous studies have shown that a small social network and insufficient social support are associated with increased incidence and mortality of CHD and stroke^12–20. In the Japan Public Health Center-based Prospective Study (JPHC study) with an average follow-up of 10.7 years, low social support was associated with a higher risk of stroke mortality in men¹². Similarly, in the Atherosclerosis Risk in Communities (ARIC) study with a median follow-up of 18.6 years, a small social network was associated with an increased risk of incident stroke even after adjustment for traditional risk factors²⁰. In the Health Professionals Follow-up Study in the United States with a 10-year follow-up, social isolation was associated with a higher risk of fatal CHD in men¹⁸. Consistent results have been reported in European cohorts; low social support was associated with an increased risk of incident CHD in Swedish men¹⁷. Furthermore, a systematic review showed that limited social networks and low social support were associated with an increased risk of morbidity and mortality from CHD¹⁹. Taken together, these findings support the notion that social isolation contributes to adverse cardiovascular outcomes. Consistent with this concept, the results of the present study showed that social isolation, as assessed using LSNS-6, was significantly associated with increased arterial stiffness, as measured by baPWV. Given that increased baPWV has been shown to be independently associated with an increased risk of cardiovascular events⁷, these findings suggest that social isolation may contribute to an increased risk of cardiovascular events, at least in part, through its adverse effects on arterial stiffness. In addition, our findings suggest that interventions targeting arterial stiffness may help reduce cardiovascular risk in socially isolated individuals. Although the exact mechanisms underlying the association between social isolation and arterial stiffening remain unclear, several potential pathways have been proposed. Behavioral factors such as smoking, physical inactivity, poor diet, and inadequate sleep, as well as psychological factors such as depression, may contribute to vascular changes. Physiological responses to chronic stress, including increased allostatic load and systemic inflammation, are also implicated^21–23. A clinical study demonstrated that individuals with smaller social networks had elevated serum levels of interleukin-6, a marker of inflammation²¹. Experimental studies have shown that social isolation enhances stress-related neuroendocrine and inflammatory responses, which can lead to endothelial dysfunction and vascular injury^22,23. These processes may impair post-stroke recovery and worsen clinical outcomes. Taken together, these findings suggest that socially isolated individuals are more likely to experience unhealthy lifestyle patterns and heightened psychosocial stress, leading to systemic inflammation, sympathetic nervous system activation, increased platelet reactivity, and endothelial dysfunction, all of which may contribute to arterial stiffening²⁴. Further studies are needed to elucidate these mechanisms and to determine whether interventions targeting social isolation, such as community-based programs or enhanced social support networks, can mitigate its adverse effects on arterial stiffness and cardiovascular risk.

Although a recent study by Arcidiacono et al. demonstrated a significant association between social isolation and cfPWV in a sample of 58 healthy individuals²⁵, our study offers several novel contributions. First, our study included a substantially larger sample of 357 patients from a cardiovascular outpatient clinic, representing a clinically high-risk population. Second, we used baPWV, a validated and widely used marker of arterial stiffness that is more practical for routine clinical use. Third, social isolation was assessed using the LSNS-6, a brief and efficient six-item instrument suitable for implementation in real-world clinical setting. Fourth, our findings highlight a significant association between social isolation and arterial stiffness in a population with existing cardiovascular risk, thereby, offering practical implications for patient management and intervention strategies.

We investigated whether the association between social isolation and baPWV differed by sex. Previous epidemiological studies have primarily focused on men^16–18, and one cohort study reported a stronger association between social isolation and cardiovascular risk in men than in women¹², whereas significant associations have also been observed in women¹³. A systematic review, however, found no consistent sex-specific differences in this association²⁶, highlighting inconsistencies in the existing evidence. Given the established sex differences in psychosocial profiles including stress perception, and coping strategies, as well as vascular biology, such as endothelial function and arterial stiffness, we conducted sex-stratified analyses in the present study. In our analyses, no significant sex-specific interaction was observed. However, the association between social isolation and baPWV was statistically significant in men but not in women. This discrepancy may reflect underlying biological and psychosocial differences between sexes. Nonetheless, the lack of statistical significance in women should be interpreted with caution, as it may be partly due to the smaller sample size and lower prevalence of social isolation among women in our cohort, which could have limited the statistical power to detect a meaningful association.

This study has several limitations. First, as this was a cross-sectional study, causality cannot be established. Second, the generalizability of our findings may be limited to similar populations, as this study was conducted at a single center. Third, there is a potential for selection bias, as the participants were not selected randomly or consecutively, and all participants were recruited from a cardiology outpatient clinic. Fourth, a formal sample size calculation was not performed before the study because of its exploratory nature. Fifth, although the observed associations were statistically significant, the effect sizes were modest and may not directly inform clinical decision-making. Therefore, the practical implications of our findings should be interpreted with caution. Sixth, other unmeasured confounders, including health-related behaviors, socioeconomic status and mental health conditions, may have influenced the observed associations. We cannot deny the possibility that these unmeasured potential confounders could have contributed to a spurious association between social isolation and increased arterial stiffness. Seventh, current and former smokers were grouped together as “smokers” in this study, although these groups may differ in their effects on arterial stiffness. Due to the relatively small number of current smokers (n = 20), as shown in Table 1, stratified multivariable analysis was not feasible. Further studies involving large and more diverse populations are needed to investigate these potential sex-specific mechanisms and to improve the generalizability of our findings.

Conclusions

Social isolation, as assessed using LSNS-6, is significantly associated with increased arterial stiffness, as assessed by baPWV, in individuals aged 55 years or older. Social isolation may be associated with cardiovascular risk partly through its adverse effects on arterial stiffness. These findings in this study highlight the importance of identifying and addressing social isolation to reduce cardiovascular risk.

Methods

Subjects

This cross-sectional study retrospectively analyzed data from patients aged ≥ 55 years who visited the cardiovascular outpatient clinic at Hiroshima University Hospital between October 2018 and December 2022. Participants were not randomly selected. The inclusion criteria were age ≥ 55 years, attendance at the cardiovascular outpatient clinic, completion of the LSNS-6 questionnaire, and availability of baPWV measurements. The exclusion criteria were secondary hypertension, missing baPWV data, and a bilateral ankle-brachial index (ABI) of < 0.9. Of the 446 eligible patients, 89 were excluded (50 with secondary hypertension, 28 with missing baPWV data, and 11 with a bilateral ABI of < 0.9). Finally, 357 participants (mean age: 72.0  ±  7.6 years; 204 men and 153 women) were included in the final analysis. Hypertension was defined as treatment with oral antihypertensive drugs or systolic blood pressure of more than 140 mm Hg and/or diastolic blood pressure of more than 90 mm Hg in a sitting position on at least 3 different occasions without medication²⁷. Diabetes was defined according to the American Diabetes Association recommendation²⁸. Dyslipidemia was defined according to the third report of the National Cholesterol Education Program²⁹. We defined smokers as those who had ever smoked. Cardiovascular disease included CHD and cerebrovascular disease. CHD included angina pectoris, myocardial infarction, and unstable angina. Cerebrovascular disease included ischemic stroke, hemorrhagic stroke, and transient ischemic attack. Mean arterial pressure was calculated as diastolic blood pressure plus one-third of the pulse pressure. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. The estimated glomerular filtration rate (eGFR) was calculated using the Japanese eGFR equation: eGFR (mL/min/1.73 m²) = 194 × serum creatinine^-1.094 × age^-0.287 (× 0.739 for women)³⁰. CKD was defined as eGFR < 60 mL/min/1.73 m²³¹. Measurement of baPWV was performed without withholding medications. This study was conducted in accordance with the 1975 Declaration of Helsinki. The study protocol was approved by the Hiroshima University Hospital Institutional Review Board on February 5, 2020 (approval no. E2019-1902). As this was a retrospective study, informed consent was obtained through an opt-out process. The protocol was registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN000039512).

Study protocol

Subjects fasted the previous night for at least 8 h and abstained from consuming caffeine, alcohol, and antioxidant vitamins and from smoking on the day of the examination. The subjects were kept in the supine position in a quiet, dark, air-conditioned room (constant temperature, 23─26 °C) throughout the study. A 23-gauge polyethylene catheter was inserted into the left deep antecubital vein to obtain blood samples. At least 20 min after maintaining the supine position, baPWV measurements were performed by skilled and trained physicians without knowledge of the baseline clinical characteristics of the subjects³².

LSNS-6

LSNS-6 was used to assess social networks and screen for social isolation among participants^10,11. LSNS-6 is constructed from a set of three questions that evaluate kinship ties (family relationships) and a comparable set of three questions that evaluate nonkin ties (friendships). Each item is scored on a scale from 0 to 5, yielding a maximum score of 15 for kinship-related items and a maximum score of 15 for nonkin-related items, with a total score ranging from 0 to 30. Higher scores indicate stronger social connections. Based on prior validation studies, a score of less than 12 has been identified as a threshold for defining individuals at risk of social isolation¹⁰.

baPWV measurement

baPWV was measured using a volume-plethysmographic apparatus (Form PWV/ABI, Omron Health Care Co., Kyoto, Japan). Detailed information on the study protocol and baPWV measurement is provided in the online-only Data Supplement. In brief, four blood pressure cuffs were applied to both the upper arms and ankles. baPWV was automatically calculated using the formula: (D1 - D2)/T, where D1 is the distance from the suprasternal notch to the ankle, which was estimated using the equation D1 = 0.8129 × height (cm) + 12.328, and D2 is the distance from the suprasternal notch to the brachium, which was estimated using the equation D2 = 0.2195 × height (cm) − 2.0734. T represents the time interval between the wave front of the brachial waveform and that of the ankle waveform. In patients with bilateral ABI ≥ 0.9, the higher baPWV value between the two sides was used for the analysis. In patients with an ABI < 0.9 on one side and an ABI ≥ 0.9 on the other, the baPWV value from the side with an ABI ≥ 0.9 was used. Arterial stiffening was defined as baPWV ≥ 1800 cm/s³³.

Methodological quality assessment

To enhance methodological transparency, we evaluated the quality of this study using the National Institutes of Health (NIH)/National Heart, Lung, and Blood Institute (NHLBI) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies (https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools). A summary of the assessment is presented in the online-only Data Supplement (Supplementary Table 1).

Statistical analysis

All reported probability values were two-sided, with a p-value of < 0.05 considered statistically significant. Continuous variables are summarized as means  ±  standard deviation (SD) and were compared using the unpaired Student’s t-test. Categorical variables are presented as frequencies and percentages and were analyzed using the chi-squared test. The Pearson product-moment correlation coefficient was calculated to examine the relationship between baPWV and LSNS-6 total score. Multiple linear regression analyses were performed to determine whether the LSNS-6 total score, treated as a continuous variable, was significantly associated with baPWV. The model included LSNS-6 total score, age, sex, BMI, systolic or mean arterial pressure, heart rate, antihypertensive medication use, dyslipidemia, diabetes mellitus, smoking, CKD, and a history of cardiovascular disease (CHD and cerebrovascular disease). Adjusting variables were selected according to previously reported determinants of baPWV³⁴, which were considered as potential confounders and therefore included in the multivariable models. Multiple logistic regression analyses were performed to identify independent variables associated with arterial stiffening (baPWV ≥ 1800 cm/s). The model included LSNS-6 total score < 12, age, sex, BMI, systolic or mean arterial pressure, heart rate, antihypertensive medication use, dyslipidemia, diabetes mellitus, smoking, CKD, and a history of cardiovascular disease. Analyses were performed using complete-case data. Participants with missing values for any variable included in a given analysis were excluded from that analysis. No imputation was performed. All data were analyzed using JMP version 18 (SAS Institute, Cary, NC).

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Yukihito Higashi and Tatsuya Maruhashi drafting the article and conception of this study; Masato Kajikawa, Shinji Kishimoto, Takayuki Yamaji, Takahiro Harada, Aya Mizobuchi, Shunsuke Tanigawa, Farina Mohamad Yusoff, and Chikara Goto measuring baPWV; Yukiko Nakano and Ayumu Nakashima revising the article critically for important intellectual content.

Funding

This study was supported in part by Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (18590815 and 21590898 to Y.H.) (16K19408 and 19K17565 to T.M.) and a Grant in Aid of Japanese Arteriosclerosis Prevention Fund (to Y.H.).

Data availability

The data that support the findings of this study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Ethical approval

This study was conducted in accordance with the 1975 Declaration of Helsinki. The study protocol was approved by the Hiroshima University Hospital Institutional Review Board on February 5, 2020 (approval no. E2019-1902).