Hostname: page-component-848d4c4894-xm8r8 Total loading time: 0 Render date: 2024-07-01T02:02:42.151Z Has data issue: false hasContentIssue false

Association between social network and dietary variety among community-dwelling older adults

Published online by Cambridge University Press:  03 July 2023

Miyuki Yokoro
Affiliation:
Department of Dietary Life and Food Sciences, Junior College Division, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan Research Institute for Nutrition Sciences, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan
Naoto Otaki*
Affiliation:
Research Institute for Nutrition Sciences, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan Department of Food Sciences and Nutrition, School of Food Sciences and Nutrition, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan
Tomomi Imamura
Affiliation:
Research Institute for Nutrition Sciences, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan Department of Innovative Food Sciences, School of Food Sciences and Nutrition, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan
Norikazu Tanino
Affiliation:
Department of Food Sciences and Nutrition, School of Food Sciences and Nutrition, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan
Keisuke Fukuo
Affiliation:
Research Institute for Nutrition Sciences, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan Department of Food Sciences and Nutrition, School of Food Sciences and Nutrition, Mukogawa Women’s University, 6-46 Ikebiraki-cho, Nishinomiya, Hyogo 663-8558, Japan
*
*Corresponding author: Email otk_nao@mukogawa-u.ac.jp
Rights & Permissions [Opens in a new window]

Abstract

Objective:

Social networks are critical social health factors for older adults. This study examined the association between social networks and dietary variety among community-dwelling older adults.

Design:

A cross-sectional study, using the dietary variety score (DVS) developed for older Japanese people to assess dietary variety and the Lubben Social Network Scale (LSNS-6) to assess social networks.

Setting:

N City, H Prefecture, Japan.

Participants:

Community-dwelling older adults aged ≥ 65 years (n 1229).

Results:

The LSNS-6 score in the low DVS group was lower than that in the middle and high DVS groups (12·2 ± 5·6 v. 13·4 ± 5·4 and 14·4 ± 5·7, P < 0·001). The population of social isolation (LSNS-6, < 12) in the low DVS group was higher than that in the middle and high DVS groups (43·5 % v. 35·8 % and 31·0 %, P = 0·005). Multivariate linear regression analysis showed that the LSNS-6 score was positively correlated with DVS (standardised coefficient, 0·092; P = 0·002). Social isolation was also significantly associated with a low DVS in the multivariate-adjusted logistic analysis model (OR, 1·30; 95 % CI 1·00, 1·68; P = 0·048). Stratified analysis results revealed the LSNS-6 and DVS were significantly associated in participants with the following characteristics: younger age (< 75 years), women and those living with someone.

Conclusions:

Social networks were associated with dietary variety; social isolation was related to poor dietary variety among community-dwelling older adults. An association between social networks and dietary variety was observed among young-old older adults, women and those living with someone.

Type
Research Paper
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of The Nutrition Society

A social network refers to the structural aspects of social ties, such as the size and frequency of contact with others, and is considered an indicator of social integration and social engagement(Reference Holtzman, Rebok and Saczynski1,Reference Kurimoto, Awata and Ohkubo2) . Small social networks result in social isolation(3). Social isolation can lead to malnutrition, depression, lower cognitive function and frailty among older adults(Reference Boulos, Salameh and Barberger-Gateau4Reference Gale, Westbury and Cooper7). Furthermore, social isolation increased the risks of CVD and mortality(Reference Holt-Lunstad and Smith8,Reference Holt-Lunstad, Smith and Baker9) . Research suggests that social networks could be related to health outcomes through several pathways, including their influence on health-related behavioural patterns such as smoking, alcohol use or medical help-seeking(Reference Cohen and Wills10,Reference Kawachi and Berkman11) . However, limited studies have investigated the relationship between social networks and dietary habits among older adults(Reference Delerue Matos, Barbosa and Cunha12,Reference Yamaguchi, Yamada and Hapsari13) .

Dietary variety, defined as the number of different foods or food groups consumed during a given period, is among the significant indicators of a balance of nutrients(Reference Bernstein, Tucker and Ryan14). Higher dietary variety improves the adequacy of nutrient intake(Reference Foote, Murphy and Wilkens15). The dietary variety score (DVS) was developed by Kumagai et al. to evaluate dietary variety among older Japanese adults(Reference Kumagai, Watanabe and Shibata16). DVS validity has been adequately confirmed for energy, protein and the intake of some nutrients(Reference Narita, Kitamura and Takemi17). A low DVS is related to malnutrition, depression, poor sleep quality, lower physical performance and frailty among older adults(Reference Tsuji, Yamamoto and Yamasaki18Reference Yokoro, Otaki and Yano25). DVS is affected by various factors, such as age, gender, chewing ability, intellectual activity, living arrangements, economic status and food accessibility(Reference Narita, Kitamura and Takemi17,Reference Yokoro, Otaki and Yano25Reference Harada, Masumoto and Okada28) . A previous study reported that perceived social support for diet, such as having someone to help with food shopping and receiving any food from neighbours or relatives, acted as a buffer against low DVS among older adults who live alone(Reference Fukuda, Ishikawa and Yokoyama27). However, the association between social networks and DVS is unclear.

Therefore, this study aimed to examine the association between social networks and DVS among community-dwelling older adults. Since social networks have been suggested to influence health behaviours, we hypothesise that poor social networks lead to deficient dietary variety among older adults. Furthermore, we consider the influence of age, gender and living arrangements on social networks among older adults based on prior related literature. For instance, older adults’ social networks have shown a reverse progressive sequence with an increase in age(Reference Zheng and Chen29). Furthermore, social network size in old age has been found to expand more for women than men(Reference Schwartz and Litwin30). Additionally, lower social network scores have been reported among older adults living alone than those living with someone(Reference Kurimoto, Awata and Ohkubo2). Therefore, further analyses were conducted to determine the association between social networks and DVS after stratifications based on age, gender and living arrangements.

Methods

Participants and study period

This cross-sectional study was conducted using a follow-up survey from our previous study(Reference Yokoro, Otaki and Yano25,Reference Otaki, Yokoro and Yano31) . The study participants were community-dwelling older people from the overall older population aged ≥ 65 years in N City, H Prefecture, Japan, excluding hospitalised older people and those who resided in a nursing home. The participants included respondents who agreed to receive our follow-up survey at the time of the baseline survey. Of the 2794 participants in the baseline survey, 1635 responded to the prospective survey (58·5 %). The survey forms, including research descriptions such as the aim of this study, were sent to these participants by mail in mid-August 2021. Of the 1635 participants, 1243 participants (76·0 %) responded to the survey questionnaire. However, twenty-four respondents were excluded owing to missing data on age and DVS. Finally, 1229 participants were included in the study.

Social networks

Social networks were assessed using the Japanese version of the abbreviated Lubben Social Network Scale (LSNS-6)(Reference Kurimoto, Awata and Ohkubo2,Reference Lubben, Blozik and Gillmann32) . The LSNS-6 covers emotional and instrumental support and comprises three items each related to family networks and friendship networks. The number of people in each network was determined on a 6-point scale. The LSNS-6 score ranged from 0 to 30 points; higher scores indicated larger social networks and scores < 12 indicated social isolation.

Dietary variety

Dietary variety was assessed using the DVS developed for older Japanese people(Reference Kumagai, Watanabe and Shibata16). Participants selected the intake frequencies of ten food groups, which included fish and seafood, meat, eggs, dairy products, soyabean and its products, green and yellow vegetables, seaweed, potatoes, fruits and fat or oils, in the last week from the following choices: ‘eat almost every day’, ‘eat once every 2 d’, ‘eat 1 or 2 d a week’ and ‘hardly eat at all’. The choice of ‘eat almost every day’ led to a score of 1, while the other choices led to a score of 0. The sum of the ten food group scores was used to express the DVS. The DVS ranged from 0 to 10 points, with a higher score indicating higher dietary variety. Scores of < 4, 4–6 and ≥ 7 points indicated low, middle and high DVS, respectively. The validity of DVS has been previously confirmed(Reference Narita, Kitamura and Takemi17).

Other variables

Demographic data were obtained on variables including comorbidities, living arrangements and subjective economic status. The number of comorbidities was obtained by counting self-reported comorbidities; in particular, self-reported cancer, cardiac disease, stroke, hypertension, diabetes and dyslipidaemia were assessed as comorbidities. Living arrangement was assessed from participants’ responses to whether they lived alone or with someone. Referring to our previous study, subjective economic status was assessed using a five-point scale ranging from very secure to very insecure and was divided into two groups: secure and insecure(Reference Otaki, Tanino and Yokoro33). Secure included the following responses: ‘very secure’, ‘secure’, and ‘neither’; insecure included the following responses: ‘insecure’ and ‘very insecure’. Since DVS is affected by chewing ability, the oral frailty score was assessed using the oral frailty checklist proposed by the Japan Dental Association, to determine oral function(Reference Kwon, Suzuki and Kumagai26,Reference Nomura, Ishii and Chiba34) . The oral frailty score was calculated as the sum of points of the eight questions regarding oral functions, and the oral frailty risk was divided into low risk, risk and high risk according to previous reports(Reference Nomura, Ishii and Chiba34).

Statistical analysis

The participants were divided into low, middle and high DVS groups. Age was calculated from the date of birth obtained in the baseline survey. Gender details were obtained from the baseline survey data. Participant characteristics, as quantitative variables, were expressed as mean ± sd (median). Categorical variables were expressed as numbers (percentages). The Kruskal–Wallis and chi-squared tests for quantitative and categorical variables, respectively, were used to compare the participants among the three DVS groups. Multivariable linear regression analysis was used to analyse crude, age- and gender-adjusted and multivariable-adjusted correlations between the LSNS-6 score and DVS. The covariates of the multivariable-adjusted model included age, gender (women), number of comorbidities, subjective economic insecurity and oral frailty score (continuous variable). The assumptions of linearity and constant variance of error in the models were checked by scatter plots. The normality in the models was checked by determining whether the residuals have a normal distribution with histograms(Reference Casson, Franzco and Dm35). All models met these assumptions. The association between social isolation and low DVS was tested using logistic regression analysis to further calculate OR and 95 % CI. Covariates of the logistic analysis included the following: ≥ 75 years of age, gender (women), living alone, hypertension, diabetes, subjective economic insecurity and oral frailty risk (low risk, risk and high risk). Variables that showed significant differences in the comparisons among the DVS groups (P < 0·05) were chosen as covariates in the linear regression and logistic regression analyses. Living alone was also included as a covariate since a previous study reported an association between this variable and DVS(Reference Yokoro, Otaki and Yano25). The correlations between the LSNS-6 and DVS were examined through stratified analysis by age, gender and living arrangements, which have been reported to influence social networks(Reference Kurimoto, Awata and Ohkubo2,Reference Zheng and Chen29,Reference Schwartz and Litwin30) . A sensitivity analysis was performed using mean imputation to examine the possible influence of missing data on the results. IBM SPSS Statistics 26.0 (IBM Corp.) was used to analyse all the statistical data. Two-tailed P < 0·05 was considered statistically significant.

Results

Participant characteristics

Table 1 shows the overall participant characteristics and those based on DVS groups. The average ± sd (median) age of the participants was 75·2 ± 6·0 (74·0) years. Of the total participants, 45·6 % (n 561) were men and 54·4 % (n 668) were women. The average ± sd of the LSNS-6 score was 13·1 ± 5·6 (13·0); 38·1 % of the participants were considered socially isolated (LSNS-6 score < 12). Populations with cancer, heart disease, stroke, dyslipidaemia and those that lived alone differed insignificantly among the three DVS groups. However, the three DVS groups exhibited significant differences in age, gender, number of comorbidities, population with hypertension and diabetes, subjective economic status, oral frailty score and oral frailty risk. Compared with the low DVS group, the age and population of women were significantly higher in the middle and high DVS groups. In the low DVS group, the number of comorbidities and the economically insecure population were high. Additionally, the oral frailty score and high-risk population with oral frailty in the low DVS group were significantly higher than those in the middle and high DVS groups. The LSNS-6 score among the three DVS groups differed significantly. The low DVS group had a low LSNS-6 score compared to the middle and high DVS groups (12·2 ± 5·6 v. 13·4 ± 5·4 and 14·4 ± 5·7, P < 0·001). Furthermore, the socially isolated population in the low DVS group was higher than that in the middle and high DVS groups (43·5 % v. 35·8 % and 31·0 %, P = 0·005).

Table 1 Overall participant characteristics and those according to DVS groups (n 1229)

DVS, dietary variety score; LSNS, Lubben Social Network Scale.

Kruskal–Wallis test.

X2 test.

Associations between the Lubben Social Network Scale-6 score and social isolation with DVS

The association between the LSNS-6 score and DVS was analysed using multivariate linear regression analysis (Table 2). The LSNS-6 score positively correlated with DVS in the crude (standardised coefficient, 0·148; P < 0·001) and the age- and gender-adjusted (Model 1; standardised coefficient, 0·132; P < 0·001) and multivariate-adjusted models (Model 2; standardised coefficient, 0·108; P < 0·01). Model 3, in which the oral frailty score was added as a covariate, displayed a constant significant association (standardised coefficient, 0·092; P = 0·002). Furthermore, the association between social isolation and low DVS was examined using logistic analysis (Table 3). Social isolation (LSNS-6 score < 12) was found to be significantly associated with low DVS (score < 4) in the crude (OR, 1·45; 95 % CI 1·14, 1·86; P = 0·003) and the age- and gender-adjusted (Model 1; OR, 1·39; 95 % CI 1·08, 1·79; P = 0·010) and multivariate models (Model 2; OR, 1·36, 95 % CI 1·05, 1·75; P = 0·019). The significant association between social isolation and low DVS also remained constant in Model 3, which included the oral frailty score as a covariate (OR, 1·30; 95 % CI 1·00, 1·68; P = 0·048). These results showed that social networks were associated with dietary variety among community-dwelling older adults.

Table 2 Liner regression analysis between the DVS and LSNS-6 (n 1148)

DVS, dietary variety score; LSNS, Lubben Social Network Scale.

Table 3 Logistic regression analysis between low DVS (< 4) and social isolation (n 1148)

DVS, dietary variety score; LSNS, Lubben Social Network Scale.

Stratified analysis of the association between the Lubben Social Network Scale-6 and dietary variety score using liner regression analysis

The association between the LSNS-6 and DVS was stratified by age, gender and living arrangement (Table 4). In the multivariate-adjusted models, the results show that the LSNS-6 and DVS were significantly associated in participants of a younger age (age < 75 years; standardised coefficient, 0·122; P = 0·001), women respondents (standardised coefficient, 0·103; P = 0·011) and those living with someone (standardised coefficient, 0·085; P = 0·010). In contrast, the LSNS-6 scores were insignificantly related to DVS in participants with the following characteristics: older age (≥ 75 years), men and living alone.

Table 4 Stratified analysis of the association between the LSNS-6 and DVS using liner regression analysis (n 1148)

DVS, dietary variety score; LSNS, Lubben Social Network Scale.

* Adjusted for age, gender, number of comorbidities, living arrangements, subjective economic status and oral frailty score.

Adjusted for age, number of comorbidities, living arrangements, subjective economic status and oral frailty score.

Adjusted for age, gender, number of comorbidities, subjective economic status and oral frailty score.

Sensitivity analysis

Mean imputations were conducted to account for missing data on the number of diseases requiring therapy, hypertension, diabetes and subjective economic status. The mean imputation results ascertained similar relationships between the LSNS-6 and social isolation with DVS.

Discussion

The results of this study revealed social networks to be positively associated with dietary variety among randomly selected community-dwelling older adults. Additionally, social isolation showed a significant relationship with low dietary variety, indicating low dietary quality. These associations were found to be independent of age, gender, living arrangements, economic insecurity and oral frailty. Significant associations between social network and dietary variety were observed among those aged < 75 years, women and those living with someone, rather than among those aged ≥ 75 years, men and those living alone. Overall, poor social networks may lead to worse dietary variety among community-dwelling older adults. Moreover, the association between social networks and dietary variety was found to be influenced by age, gender and living arrangements. To the best of our knowledge, the association between social isolation and dietary variety has not been described before. A low DVS causes malnutrition, depression, lower physical performance and frailty among older adults(Reference Tsuji, Yamamoto and Yamasaki18,Reference Yokoyama, Kitamura and Yoshizaki19,Reference Yokoyama, Nishi and Murayama22Reference Otaki, Yano and Yokoro24) . Accordingly, our results, which show the importance of social networks for DVS, have implications for supporting healthy ageing.

Several studies support our results. Kwon et al. reported that loss of a spouse is a significant predictor of the decline in DVS(Reference Kwon, Suzuki and Kumagai26). Frequent friend contact is associated with a better variety of vegetables among older adults who are widowed or have limited family contact(Reference Conklin, Forouhi and Surtees36). Although the mechanisms of the association between social networks and dietary variety were unclear in this study, Kurimoto et al. showed that larger social networks facilitated access to social support among older adults(Reference Kurimoto, Awata and Ohkubo2). A related qualitative study(Reference Bukman, Ronteltap and Lebrun37) notes that the limited use of social support for grocery shopping and cooking is among the determinants of eating behaviours among older adults, while food-related social support availability, such as help with grocery shopping and/or food exchange, was observed as a related factor for DVS(Reference Fukuda, Ishikawa and Yokoyama27,Reference Bukman, Ronteltap and Lebrun37) . Moreover, several mechanisms, such as social influence/social comparison, social control, role-based purpose and meaning, self-esteem, sense of control, belonging and companionship, might be related to the association between social networks and dietary behaviour(Reference Thoits38).

Differences were observed in the association between social networks and DVS according to age, gender and living arrangement. This study notes an association between social networks and DVS among oldest-old adults rather than young-older adults. According to the multivariate linear regression analysis, the number of comorbidities was significantly associated with DVS among the oldest-old adults (standardised coefficient, –0·107; P = 0·012), whereas this association was not observed among young-old adults (standardised coefficient, 0·005; P = 0·890). The number of comorbidities among oldest-old and young-old adults was 1·7 ± 1·3 (1·0) and 1·3 ± 1·2 (1·0), respectively (P < 0·001, Mann–Whitney U test). Several diseases, including hypertension, have been reported to be associated with reduced DVS(Reference Narita, Kitamura and Taniguchi39). The present results also indicated that the hypertension and diabetes rates in the low DVS group were higher than those in the high DVS group (Table 1). The association between social networks and DVS among the oldest-old adults may have weakened owing to an increase in diseases that affect eating habits. The gender differences in the influence of social networks on older adults’ diets are consistent with those of other studies(Reference Fukuda, Ishikawa and Yokoyama27,Reference Bloom, Edwards and Jameson40) . One of the pathways through which social networks influence health behaviours is the provision of social support. A previous study reported that the association between social support and vegetable and fruit intake was found only in women and not in men, suggesting that the social aspects of meals are crucial for maintaining an adequate diet among women(Reference Rugel and Carpiano41). A possible reason for the gender difference in the association between social ties and health behaviours is that women tend to maintain more emotionally intimate relationships than men(Reference Kawachi and Berkman11). Therefore, women may be more likely to receive support from the members of their social networks than men. In addition, a qualitative study also reported that the role of family and friends in motivating women is particularly important(Reference Bloom, Lawrence and Barker42). Hence, this study observes an association between social networks and dietary variety among women rather than men. Regarding living arrangements, our results were inconsistent with those of a previous study. This study shows that social networks and DVS are associated in those living with someone, while Conklin et al. reported that frequent contact with friends increases vegetable variety among older adults who live alone. However, no association was found among those who live with others(Reference Conklin, Forouhi and Surtees36). Our previous study showed that the difference in living arrangements in relation to the DVS was influenced by gender(Reference Yokoro, Otaki and Yano25). Differences in living arrangements in the association between social networks and DVS might be complicated owing to the involvement of multiple factors. Further research must include other factors related to living alone, such as the period of living alone and life events that lead to living alone (e.g. divorce, death of spouse or children).

The association between social networks and dietary variety was adjusted by the oral frailty score in this study, as chewing ability is a critical influencing physical factor for eating behaviour among older adults(Reference Kwon, Suzuki and Kumagai26,Reference Kimura, Ogawa and Yoshihara43) . The independent relationship between social networks and dietary variety was determined in our multivariate analysis, which included the oral frailty score as a covariate. This result suggests that physical and social care are important for dietary management among older adults.

The strengths of this study are as follows. First, to the best of our knowledge, it was the first to identify an association between social networks and DVS among older adults. Second, it involved stratified analysis with almost comparable age and gender ratios. However, the number of participants living alone was smaller than the number of participants living with someone; thus, the statistical power was small in the living alone group. This study had a few other limitations. First, it was a cross-sectional study. The results need to be interpreted with caution since we could not discuss causal relationships. Second, 58·5 % of the respondents accepted the follow-up study at the baseline survey. Additionally, the response rate in the follow-up survey was 76·0 %. This might have resulted in sampling bias and reduced generalisability. In the baseline survey, while no differences were observed in age, number of diseases or percentage living alone according to the presence or absence of acceptance for the follow-up survey, men had a higher percentage of acceptance than women did. The DVS was significantly higher for those who accepted compared to those who did not (data not shown). The study population may have been biased towards those with a high level of health awareness. Third, all variables were self-reported data obtained from the mail survey. Thus, a social desirability response bias may affect the DVS and the social network scores in this study. Fourth, the survey was conducted during the COVID-19 pandemic when almost no community activities intended for older adults occurred. This may have influenced the self-evaluation of the social networks. Fifth, we did not assess quantitative food or nutrient intake, whereas the DVS was validated to relate to nutrient intake(Reference Narita, Kitamura and Takemi17). Finally, although depression is a key influencing factor of dietary variety, we could not assess such psychological conditions in this study(Reference Yokoyama, Kitamura and Yoshizaki19).

In conclusion, this study found that poor social networks and social isolation are related to reduced dietary variety among community-dwelling older adults. Particularly, the association between social networks and dietary variety was shown in the following subgroups: young-old adults, women and those living with someone. These results suggest that screening older adults who have poor social networks and increased social support may contribute to improved food variety and health. However, further research is needed to identify effective social support for maintaining healthy dietary behaviour among socially isolated older adults.

Acknowledgements

We thank the city employees at Nishinomiya City Office for their help in distributing survey questionnaires.

Financial support

This research was supported by the Department of Planning and Prefectural Citizens, Hyogo Prefecture under Grant Number 10 and JSPS KAKENHI Grant Number 20K19730 and 21K11435.

Conflict of interests

There are no conflicts of interest.

Authorship

N.O., N.T. and K.F. designed this study. M.Y., N.O., T.I. and N.T. participated in the data collection. M.Y. analysed and interpreted the data, and prepared the draft manuscript. N.O., T.I., N.T. and K.F. revised the manuscript for important intellectual content and provided critical revision of the draft. All the authors approved the final version of the article.

Ethics of human subject participation

This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving research study participants were approved by the Ethics Committee of Mukogawa Women’s University approved this study (Approval Number: 20–53). The details of the study were explained in writing to the subjects, and the return of a completed survey form was considered to obtain informed consent for participation in the study.

References

Holtzman, RE, Rebok, GW, Saczynski, JS et al. (2004) Social network characteristics and cognition in middle-aged and older adults. J Gerontol B Psychol Sci Soc Sci 59, 278284.CrossRefGoogle ScholarPubMed
Kurimoto, A, Awata, S, Ohkubo, T et al. (2011) Reliability and validity of the Japanese version of the abbreviated Lubben Social Network Scale. Nihon Ronen Igakkai Zasshi 48, 149157.CrossRefGoogle ScholarPubMed
World Health Organization (2021) Advocacy Brief: Social Isolation and Loneliness among Older People. https://www.who.int/publications/i/item/9789240030749 (accessed March 2023).Google Scholar
Boulos, C, Salameh, P & Barberger-Gateau, P (2017) Social isolation and risk for malnutrition among older people. Geriatr Gerontol Int 17, 286294.CrossRefGoogle ScholarPubMed
Courtin, E & Knapp, M (2017) Social isolation, loneliness and health in old age: a scoping review. Health Soc Care Community 25, 799812.CrossRefGoogle ScholarPubMed
Evans, IEM, Martyr, A, Collins, R et al. (2019) Social isolation and cognitive function in later life: a systematic review and meta-analysis. J Alzheimer’s Dis 70, S119S144.CrossRefGoogle ScholarPubMed
Gale, CR, Westbury, L & Cooper, C (2018) Social isolation and loneliness as risk factors for the progression of frailty: the English Longitudinal Study of Ageing. Age Ageing 47, 392397.CrossRefGoogle ScholarPubMed
Holt-Lunstad, J & Smith, TB (2016) Loneliness and social isolation as risk factors for CVD: implications for evidence-based patient care and scientific inquiry. Heart 102, 987989.CrossRefGoogle ScholarPubMed
Holt-Lunstad, J, Smith, TB, Baker, M et al. (2015) Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspect Psychol Sci 10, 227237.CrossRefGoogle ScholarPubMed
Cohen, S & Wills, TA (1985) Stress, social support, and the buffering hypothesis. Psychol Bull 98, 310357.CrossRefGoogle ScholarPubMed
Kawachi, I & Berkman, LF (2001) Social ties and mental health. J Urban Health 78, 458467.CrossRefGoogle ScholarPubMed
Delerue Matos, A, Barbosa, F, Cunha, C et al. (2021) Social isolation, physical inactivity and inadequate diet among European middle-aged and older adults. BMC Public Health 21, 924.CrossRefGoogle ScholarPubMed
Yamaguchi, Y, Yamada, M, Hapsari, ED et al. (2020) The influence of social isolation on the preventive behaviors for non-communicable diseases in community-dwelling older adults in Japan. Int J Environ Res Public Health 17, 8985.CrossRefGoogle ScholarPubMed
Bernstein, MA, Tucker, KL, Ryan, ND et al. (2002) Higher dietary variety is associated with better nutritional status in frail elderly people. J Am Diet Assoc 102, 10961104.CrossRefGoogle ScholarPubMed
Foote, JA, Murphy, SP, Wilkens, LR et al. (2004) Dietary variety increases the probability of nutrient adequacy among adults. J Nutr 134, 17791785.CrossRefGoogle ScholarPubMed
Kumagai, S, Watanabe, S, Shibata, H et al. (2003) Effects of dietary variety on declines in high-level functional capacity in elderly people living in a community. Nihon Koshu Eisei Zasshi 50, 11171124.Google ScholarPubMed
Narita, M, Kitamura, A, Takemi, Y et al. (2020) Food diversity and its relationship with nutrient intakes and meal days involving staple foods, main dishes, and side dishes in community-dwelling elderly adults. Nihon Koshu Eisei Zasshi 67, 171182.Google ScholarPubMed
Tsuji, T, Yamamoto, K, Yamasaki, K et al. (2019) Lower dietary variety is a relevant factor for malnutrition in older Japanese home-care recipients: a cross-sectional study. BMC Geriatr 19, 197.CrossRefGoogle ScholarPubMed
Yokoyama, Y, Kitamura, A, Yoshizaki, T et al. (2019) Score-based and nutrient-derived dietary patterns are associated with depressive symptoms in community-dwelling older Japanese: a cross-sectional study. J Nutr Heal Aging 23, 896903.CrossRefGoogle ScholarPubMed
Yamamoto, K, Motokawa, K, Yoshizaki, T et al. (2021) Dietary variety is associated with sleep efficiency in urban-dwelling older adults: a longitudinal study. Clin Nutr ESPEN 41, 391397.CrossRefGoogle ScholarPubMed
Yokoro, M, Otaki, N, Yano, M et al. (2021) Lower dietary variety is associated with worse sleep quality in community-dwelling elderly Japanese women. Asia Pac J Clin Nutr 30, 275282.Google ScholarPubMed
Yokoyama, Y, Nishi, M, Murayama, H et al. (2017) Dietary variety and decline in lean mass and physical performance in community-dwelling older Japanese: a 4-year follow-up study. J Nutr Health Aging 21, 1116.CrossRefGoogle ScholarPubMed
Motokawa, K, Watanabe, Y, Edahiro, A et al. (2018) Frailty severity and dietary variety in Japanese older persons: a cross-sectional study. J Nutr Health Aging 22, 451456.CrossRefGoogle ScholarPubMed
Otaki, N, Yano, M, Yokoro, M et al. (2020) Relationship between dietary variety and frailty in older Japanese women during the period of restriction on outings due to COVID-19. J Gerontol Ser B 76, e256e262.CrossRefGoogle Scholar
Yokoro, M, Otaki, N, Yano, M et al. (2023) Associations between dietary variety and frailty in community-dwelling older people who live alone: gender differences. J Frailty Aging 12, 3036.Google ScholarPubMed
Kwon, J, Suzuki, T, Kumagai, S et al. (2006) Risk factors for dietary variety decline among Japanese elderly in a rural community: a 8-year follow-up study from TMIG-LISA. Eur J Clin Nutr 60, 305311.CrossRefGoogle Scholar
Fukuda, Y, Ishikawa, M, Yokoyama, T et al. (2017) Physical and social determinants of dietary variety among older adults living alone in Japan. Geriatr Gerontol Int 17, 22322238.CrossRefGoogle ScholarPubMed
Harada, K, Masumoto, K & Okada, S (2020) Distance to supermarkets and dietary variety among Japanese older adults: examining the moderating role of grocery delivery services. Public Health Nutr 24, 18.Google ScholarPubMed
Zheng, Z & Chen, H (2020) Age sequences of the elderly’ social network and its efficacies on well-being: an urban-rural comparison in China. BMC Geriatr 20, 372.CrossRefGoogle ScholarPubMed
Schwartz, E & Litwin, H (2018) Social network changes among older Europeans: the role of gender. Eur J Ageing 15, 359367.CrossRefGoogle ScholarPubMed
Otaki, N, Yokoro, M, Yano, M et al. (2022) Social contact impacts physical activity and sedentary behavior among older adults in Japan due to COVID-19. BMC Geriatr 22, 491.CrossRefGoogle ScholarPubMed
Lubben, J, Blozik, E, Gillmann, G et al. (2006) Performance of an abbreviated version of the Lubben social network scale among three European community-dwelling older adult populations. Gerontologist 46, 503513.CrossRefGoogle ScholarPubMed
Otaki, N, Tanino, N, Yokoro, M et al. (2018) Relationship between economic security and self-rated health in elderly Japanese residents living alone. J Nutr Heal Aging 22, 695699.CrossRefGoogle ScholarPubMed
Nomura, Y, Ishii, Y, Chiba, Y et al. (2021) Structure and validity of questionnaire for oral frail screening. Healthcare 9, 45.CrossRefGoogle ScholarPubMed
Casson, RJ, Franzco, D, Dm, L et al. (2014) Understanding and checking the assumptions of linear regression: a primer for medical researchers. Clin Experiment Ophthalmol 42, 590596.CrossRefGoogle ScholarPubMed
Conklin, AI, Forouhi, NG, Surtees, P et al. (2014) Social relationships and healthful dietary behaviour: evidence from over-50s in the EPIC cohort, UK. Soc Sci Med 100, 167175.CrossRefGoogle ScholarPubMed
Bukman, AJ, Ronteltap, A & Lebrun, M (2020) Interpersonal determinants of eating behaviours in Dutch older adults living independently: a qualitative study. BMC Nutr 6, 55.CrossRefGoogle ScholarPubMed
Thoits, PA (2011) Mechanisms linking social ties and support to physical and mental health. J Health Soc Behav 52, 145161.CrossRefGoogle ScholarPubMed
Narita, M, Kitamura, A, Taniguchi, Y et al. (2021) Trajectories of Dietary Variety Score among community-dwelling older Japanese and their related factors. Nihon Ronen Igakkai Zasshi 58, 8190.CrossRefGoogle ScholarPubMed
Bloom, I, Edwards, M, Jameson, KA et al. (2017) Influences on diet quality in older age: the importance of social factors. Age Ageing 46, 283.Google ScholarPubMed
Rugel, EJ & Carpiano, RM (2015) Gender differences in the roles for social support in ensuring adequate fruit and vegetable consumption among older adult Canadians. Appetite 92, 102109.CrossRefGoogle ScholarPubMed
Bloom, I, Lawrence, W, Barker, M et al. (2017) What influences diet quality in older people? A qualitative study among community-dwelling older adults from the Hertfordshire Cohort Study, UK. Public Health Nutr 20, 26852693.CrossRefGoogle ScholarPubMed
Kimura, Y, Ogawa, H, Yoshihara, A et al. (2013) Evaluation of chewing ability and its relationship with activities of daily living, depression, cognitive status and food intake in the community-dwelling elderly. Geriatr Gerontol Int 13, 718725.CrossRefGoogle ScholarPubMed
Figure 0

Table 1 Overall participant characteristics and those according to DVS groups (n 1229)

Figure 1

Table 2 Liner regression analysis between the DVS and LSNS-6 (n 1148)

Figure 2

Table 3 Logistic regression analysis between low DVS (< 4) and social isolation (n 1148)

Figure 3

Table 4 Stratified analysis of the association between the LSNS-6 and DVS using liner regression analysis (n 1148)