In Japan, the percentage of older adults living alone is steadily increasing. By 2050, it is estimated that 26.1% of men and 29.3% of women aged 65 years and older will live alone—more than one in four individuals.¹⁾ Older adults are more likely to live alone following life events such as the marriage of their children or the death of a spouse. Living alone is associated with frailty, and a meta-analysis of cross-sectional studies reported a significant association between living alone and physical frailty.^2-4) Factors such as isolation, loneliness, depression, and health issues associated with living alone may contribute to an increased risk of frailty.²⁾ Frailty, in turn, is associated with disability onset,^5,6) long-term care costs,⁷⁾ and mortality.⁸⁾ Therefore, strategies to prevent physical frailty among older adults living alone are urgently needed.

Infrequent social interaction is also a contributing factor to physical frailty.⁹⁾ Reduced social interaction may lead to fewer outings, decline in lower limb muscle strength and walking speed, decreased overall activity, and fatigue related to depression. Older adults living alone generally have fewer social interactions.¹⁰⁾ A previous study¹¹⁾ found that social network size is associated with loneliness, and frequent social interaction has been shown to reduce the risk of requiring care.¹²⁾ Consequently, the frequency of social interaction among older adults living alone has gained increasing attention. However, social interaction should be evaluated in terms of frequency and quality. A decline in satisfaction with social interaction is associated with depression.¹³⁾ In older adults, the quality of social interactions appears to be more strongly related than frequency to a reduction in physical symptoms¹⁴⁾ and loneliness.¹⁵⁾ Additionally, even among individuals who live with others, low satisfaction with social interactions can increase the risk of diabetes.¹⁶⁾ Conversely, high satisfaction with social interaction may positively influence health outcomes even among those living alone, highlighting the importance of social interaction satisfaction and its effect on health outcomes.

Therefore, for older adults living alone, high frequency and satisfaction with social interactions may help reduce the risk of frailty. However, no studies have examined the frequency and satisfaction of interaction in relation to frailty by living arrangement. Recently, the National Center for Geriatrics and Gerontology–Social Network Scale (NCGG-SNS) was developed to measure the quantity and quality (satisfaction) of interactions.¹⁷⁾ Accordingly, this study aimed to investigate whether the combined factors of interaction frequency and interaction satisfaction are associated with physical frailty among community-dwelling older adults, stratified by living arrangement, using NCGG-SNS data.

Demographic characteristics of the study participants are shown in Table 1. The median age of the 5,538 participants was 72 years (IQR 67–77), and 44.3% were men. A total of 696 participants (12.6%) lived alone, and 338 (6.3%) were classified as physically frail. Supplementary Table S1 shows participant characteristics stratified by living arrangements. Those living alone had a significantly higher proportion of physical frailty and a lower SNS score. Table 2 shows that when the association between physical frailty and the SNS score was examined by living arrangement, those with physical frailty in both the living-with-others and living-alone groups had lower SNS scores. Logistic regression analysis results with physical frailty as the dependent variable and the SNS score as the explanatory variable are shown in Fig. 2. The odds are presented per one-standard-deviation increase. In both the group living with others and those living alone, for every 1 SD (standard deviation) increase in the SNS score, the odds of physical frailty decreased significantly (odds ratio [OR]: 0.86, 95% confidence interval [CI]: 0.75–0.99 and OR: 0.66, 95% CI: 0.50–0.88, respectively). In the sensitivity analysis, participants were divided into two groups based on the median SNS score, then combined with living arrangements to form four groups, and the association with physical frailty was examined within each group.

Supplementary Table S2 compares the four groups by living arrangements and high/low SNS (classified using the median score of 31). The percentage of physical frailty was highest in the living-alone/low-SNS group (10.8%), followed by the living-with-others/low-SNS group (7.1%), the living-with-others/high-SNS group (4.8%), and the living-alone/high-SNS group (3.7%). The interaction effect between SNS status and living arrangements was significant (p=0.040). Logistic regression results using the four combined groups as explanatory variables are shown in Supplementary Fig. S1. Compared to the living-alone/low-SNS group, the living-alone/high-SNS group and living-with-others/high-SNS groups exhibited significantly lower odds of physical frailty (OR: 0.34, 95% CI: 0.15–0.75 and OR: 0.64, 95% CI: 0.44–0.94, respectively). No significant differences were observed in the living-with-others/low-SNS group (OR: 0.79, 95% CI: 0.55–1.14).

This study investigated whether the combined factors of interaction frequency and satisfaction with interaction were associated with physical frailty in community-dwelling older adults. The results showed that, regardless of living arrangements, higher levels of interaction frequency and satisfaction were associated with lower ORs for physical frailty. These findings underscore the importance of engaging in frequent interactions and deriving satisfaction from them.

The associations between social interaction and physical frailty have been reported in previous studies.⁹⁾ Although the underlying mechanisms remain unclear, associations with reduced activity levels,³¹⁾ changes in dietary patterns,³²⁾ and stress³³⁾ have also been proposed. Furthermore, a higher level of psychological resources, such as well-being, is suggested to help prevent the future onset of frailty.³⁴⁾ Research has found that higher levels of satisfaction with social networks are associated with better physical and mental health scores among older adults.³⁵⁾ In this study, the low-SNS groups reported poorer subjective health and a higher prevalence of depressive tendencies, suggesting that psychological factors associated with SNS decline may have influenced the results.

Even among individuals living alone, higher SNS was associated with lower ORs for physical frailty. One study found a high OR for functional limitations among individuals with limited social participation who did not live alone.³⁶⁾ Prior studies have reported that living alone without social isolation (assessed by the frequency of interactions with others) is not associated with all-cause mortality. However, when living alone and social isolation overlapped, the hazard ratio for all-cause mortality doubled.³⁷⁾ In the current study, the proportion of participants with low SNS was high among those living alone (Supplementary Table S2). Therefore, increased efforts to enhance social interaction frequency and satisfaction among individuals living alone are emphasized.

Prior research examining the relationship between social isolation (assessed by living arrangements, interaction frequency, and social participation) and loneliness, a subjective emotion, with physical frailty reported that only loneliness was a risk factor for physical frailty.³⁸⁾ However, emotional concerns may also be associated with physical frailty, as negative relationships with close relations have been linked to greater frailty risk.³⁹⁾ Therefore, even among those living with others, increasing satisfaction with social interactions may be essential to prevent frailty.

The relationship between social interaction frequency and physical frailty has been extensively studied. However, prior research suggests that obligatory participation in social activities is more strongly associated with lower mental health scores than voluntary participation. In some cases these scores can be lower than those observed with no participation, suggesting that it could have detrimental effects on the mental health of community-dwelling older adults.⁴⁰⁾ Additionally, the quality of social interactions is associated with physical symptoms in older adults.¹⁴⁾ Therefore, improving satisfaction with social interactions may be an important strategy for preventing frailty.

The originality of this study lies in its integrated assessment of both quantitative and qualitative aspects, simultaneously measuring the frequency and satisfaction of social interaction across different living arrangements. This demonstrates the importance of behavior that considers quality—a factor that cannot be fully captured by frequency alone. Generally, it is assumed that people living with others have more opportunities for daily interpersonal contact than those living alone; however, even in such circumstances, a high SNS score was associated with lower physical frailty. The fact that the SNS score was associated with frailty, even among those living with others, supports the possibility that it is not merely the quantity of interaction opportunities that matters, but also the qualitative aspects.

A strength of this study is the use of a large-scale cohort, which enabled a comprehensive evaluation of interaction frequency and satisfaction in relation to frailty, stratified by living arrangements. Nonetheless, several limitations should be noted. First, because this was a cross-sectional study, causal relationships between social interaction frequency and satisfaction and physical frailty could not be established. Physically frail individuals may experience reduced social interaction frequency and interaction satisfaction due to fewer outings or increased fatigue. Second, selection bias may have been present, as participants were relatively healthy, health-conscious older adults who voluntarily attended a health checkup conducted in a single city in Aichi Prefecture. This may limit generalizability. The prevalence of physical frailty in this study was 6.3%, which is relatively low compared with other studies.⁴¹⁾ This finding supports that the sample was healthier than average. Future longitudinal studies are needed to examine whether low SNS levels increase the risk of developing physical frailty. Finally, although the main covariates were adjusted for, variables not included—such as physical activity level, social participation or community engagement, socioeconomic status, and marital status—may have influenced the results through residual confounding. Furthermore, studies conducted across multiple regions with diverse populations are required to ensure generalizability. Overall, these findings suggest that enhancing the quality of interactions may be as important as increasing their frequency in efforts to prevent frailty among older adults.