jfa journal

AND option

OR option



S. Dupuis-Blanchard1, C. Bigonnesse1, M.K. Andrew2, O. Gould3, D. Maillet1

1. Université de Moncton, Moncton, Canada; 2. Dalhousie University, Halifax, Canada; 3. Mount Allison University, Sackville, Canada.
Corresponding author: Suzanne Dupuis-Blanchard, School of Nursing, Université de Moncton, 18 Antonine Maillet Ave., Moncton, NB E1A 3E9, Canada,
Email : suzanne.dupuis-blanchard@umoncton.ca, Telephone : (506)858-4673, Fax : (506)858-4017

J Frailty Aging 2021;in press
Published online January 18, 2021, http://dx.doi.org/10.14283/jfa.2021.3



Background: The relationship between frailty and variables such as housing are the least included in models of frailty and research on frailty or social frailty and relocation is negligible. The decision to relocate is complex and demanding for older adults with a loss of independence but little is known about what makes older adults relocate to congregated housing designated for older adults, let alone in combination with social frailty, and how they navigate this transition. Objectives: This mixed method descriptive study aims to understand the influence of social frailty for a population of French-speaking semi-independent older adults relocating to a housing continuum community. Design: Semi-structured individual interviews including sociodemographic data and the PRISMA-7 Frailty Scale were conducted with recently relocated older adults. Setting: A newly opened French-speaking housing continuum community in Eastern Canada that offers luxury apartments for independent older adults, two assisted living facilities for semi-independent older adults along with a long-term care facility. Participants: Twenty-nine older adults with a mean age of 85 years, mostly female, married or widowed and highly educated. Measurements: Content analysis of the transcribed recorded interviews and descriptive statistical analyses to examine relationships between the frailty PRISMA-7 scale, answers to additional questions and the sociodemographic data. Results: There was not a significant difference in the scores for socialization before and after relocation nor between prior help and current help; however, there was a significant negative correlation between help and socialization before and after relocation. Three main themes included: imposed influences, push and pull factors and post relocation. Conclusions: The results indicate that several social factors contributed to relocation and that participants were experiencing social frailty. Participants were at the crossover point of being vulnerable to experiencing additional deficits which would potentially have led to higher frailty had they not relocated.

Key words: Social frailty, relocation, social support, community, official language minority.



Although frailty phenotypes have mostly ignored the notion of social frailty (1), the concept is slowly gaining interest in the literature. Defined as the absence of social resources, limited social activity and the inability to accomplish basic social needs (2), social frailty is touted as the precursor to physical frailty (3) or prefrailty (4, 5). Others who have explored the concept of social frailty have identified protective factors such as social support, engagement, living situation, self-esteem, sense of control, relations with others and contextual socio-economic status (6). Additional factors such as not living alone, going out more frequently, visiting friends, feeling helpful and talking with someone every day also had a strong impact on future disability in older adults living in the community (2, 7). Furthermore, despite the relevance of both frailty and social context on decision-making regarding housing and re-location (e.g. moving from rural to more urban areas, down-sizing or moving to supported living settings), the relationship between frailty and variables such as housing are the least included in models of frailty (1) and research on frailty or social frailty and relocation is negligible (8). Therefore, the goal of this mixed method descriptive study was to understand the influence of social frailty on decisions to relocate for a population of mostly well-educated and financially secure French-speaking semi-independent older adults.
The decision to relocate is complex and demanding for older adults with a loss of independence. Factors such as transportation, access to home maintenance services (especially in the language of choice), adequate income and level of education, attitude and resolve, self-perceived health, and choice of home/community have been determined to influence older adults’ ability to stay in their home (9) but little is known about what makes older adults relocate to congregated housing designated for older adults, whether social frailty plays a role in this decision, and how this transition is experienced (8). Although relocation is a common transition, it affects older adults much differently than younger adults. Recognizing components of social frailty in relocating older adults is important to prevent or delay a frailty diagnosis, prevent or lessen disability (10), reduce mortality (3) and improve the lives of families and caregivers (11).
This mixed method descriptive study was conducted at a newly opened Francophone housing continuum community in Eastern Canada that offers luxury apartments for independent older adults, two assisted living facilities for semi-independent older adults along with a long-term care facility. Given the influence of higher levels of education (12) and good socioeconomic status (13, 14, 15) on reducing the risk of frailty and the impact of identifying as a French-speaking older adult living in an official linguistic minority community (OLMC) on health inequalities (16), this study provides insight into the role of language as well as favourable social positioning on social frailty and relocation.



After receiving ethics approval, semi-independent older adults speaking and understanding French, aged 65 years or older, and living in a luxury housing continuum community were recruited. Recruitment strategies included brief presentations to older adults, advertisements in the community’s newsletter and electronic billboards, staff endorsing the study and the assistance of the Citizen Advisory Committee (CAC) which consisted of older adults and employees from the housing community.
The purpose of the CAC was to include stakeholders in the study to provide different perspectives and understanding to the research project to maximize the relevance of the results. CAC members informed and advised the research team and members included five older adults living at the study site, a community representative of an older adult organization, three researchers, one student research assistant and two employees from the housing continuum community. A total of three meetings were held with CAC members throughout the 12 months’ study.
A purposive sample of 29 older adults participated in semi-structured individual interviews of an average 40 minutes in duration. Interviews were conducted at a date and time convenient to the participant and most participants chose to have the interview in their apartment unit (in their new home). Sociodemographic data were recorded at the beginning of the interview and the PRISMA-7 Frailty Scale (17) was administered at the end of the interview process. The PRISMA-7 Frailty Scale is meant for early detection and management of frailty and is composed of seven yes and no questions addressing risk factors for frailty. Three or more “yes” answers are used as the cut-off for being at risk. In addition, corresponding questions related to the scale were administered with the goal of better understanding frailty of study participants. These Likert scale questions explored such components as help needed prior to relocation and after relocation, social activities prior and post relocation as well as asking for help from family and friends. For these items, a 7-point scale was used, with higher scores indicating higher levels of vulnerability.
Qualitative data analysis consisted of conventional content analysis (18) of the transcribed recorded interviews using NVivo 11 software to develop initial codes derived from the data, categories and defining themes. Descriptive statistical analyses for small sample sizes (19, 20) were performed to examine relationships between the frailty PRISMA-7 scale, answers to the additional questions and the sociodemographic data. Study results were discussed with the CAC for context and clarification as well as with the research team.



Participant Characteristics

Most of the 29 participants were female (62.1%) with an average age of 85 years old. Most were either married (38%) or widowed (41%) and 35% had no children living in proximity (20 km radius). Participants had relocated to the study site from a single dwelling (52%), an apartment or condo (28%) or directly from the hospital (14%). At the time of interview, most had relocated within 1-12 months. Participants were highly educated with 62% having a university degree and 17% a college education. Participants self rated their health as very good (35%) and good (41%) although 48% reported health problems that limited their activities.

Frailty Scale

Of the 29 study participants, 17 participants scored 3 or below (58.6 %) on the PRISMA-7 Frailty Scale with a group average score of 3.1 out of 7. Table 1 presents participants scores.

Table 1
PRISMA-7 Scores


Given the PRISMA-7 Frailty Scale scores for questions 3 (related to activities) and 5 (related to health), two corresponding Likert scale questions were analyzed: finding someone to help prior and post relocation as well as socializing before and after relocation. A paired-samples t-test was conducted to compare socialization prior to relocation and post relocation. There was not a significant difference in the scores for socialization before relocation (M = 5.14, SD = 1.38) and after relocation (M = 4.48, SD = 1.70), t(28) = 2.03, p = .052; d = .43 although a larger sample may have yielded statistical significance. Moreover, there was not a significant difference in the scores of prior help (M = 1.79, SD = 1.29) and current help (M = 1.36, SD = .78), t(27) = 1.80, p = .08; d = .40. A Pearson correlation indicated a significant negative correlation between help obtained and socialization both before relocation (r(28) = -.41, p = .03) and after relocation (r(27) = -.42, p = .03).

Imposed Influences

Results from the qualitative analysis indicated that two main life events seem to have compelled participants to move: health deterioration and capacity to source reliable support.

Health deterioration

Declining health happened over time but when a chronic health problem became overly challenging or that a new health issue arose, either for the study participant or their spouse, this was often a trigger factor that made participants decide to relocate. One participant explained: “My concern was mainly falling and finding myself alone.” Another participant shared: “I could see my health failing in terms of mobility, so sooner or later, it was better for me to initiate the move myself.”

Formal social support

Most participants described challenges in receiving formal home support services but also questioned the quality of the services once these were received. Many inconsistencies were identified such as arriving late, employee not staying for the contracted time, and tasks not completed. Even for those participants using private services, it remained challenging to receive the appropriate assistance. One participant shared, “I could have paid someone, but there’s no one reliable. I don’t mind paying $30 an hour, but they have to do the work.”

Informal social support

The majority of participants voiced strong opinions of not wanting to ask for help from family members, especially their children, but also from friends and neighbours: “It’s always trouble because you have to find someone to do your housework and other things, there’s too many things.” One reality shared by many participants was the impact of the loss of a spouse or primary caregiver as a trigger to relocation: “I have no one, I have no one anymore, they’re all deceased.”

Push and Pull Factors

Participants also identified other factors contributing to social frailty and pushing them out of their home: transportation and feelings of insecurity.
The loss of one’s ability to drive had an important impact on aging in place. Being able to drive was deemed an aspect of independence that is irreplaceable by public transit. One participant explained:
I didn’t feel vulnerable, but in a condo, without a car…you need milk, well you have to call a taxi or a friend who has…In that sense, it didn’t make sense anymore. [If] I had been able to keep my car, I would still be there you know.

Feeling of insecurity

Feelings of insecurity were mentioned by many participants and was described as: “I wasn’t feeling well, I didn’t feel safe where I was.” Others explained that they were aware that they were aging and that they needed to make changes to facilitate life: “We knew that sooner or later, we would have to move. We wouldn’t be able to keep up with our activities. Especially since it was a lot more work for me to, to maintain the house.”

Pull factors

There were also reasons for wanting to relocate to the study location that facilitated participants’ decision to relocate. Some of these factors include the location of the housing continuum, near the university and cultural centre, as well as the language spoken in the study location. Many shared: “We wanted to be somewhere Francophone; my English is not too good.” Despite wanting to stay where French was spoken, some expressed difficulties with the different accents and words used by other residents. Other pull factors included the quality of services, the continuity of care options and the ability to be close to family members. Additionally, participants could financially afford to relocate to this relatively expensive housing complex.

Post Relocation

Participants explained that once relocated, they had to adapt to their new home and that this process was different for everyone. In fact, adapting to the new home seemed more difficult for those who had relocated without having made the decision to relocate or were forced by circumstances (or triggers), and while most had made their own decision, attitude towards their relocation seemed to impact their ability to adapt. Like this participant, many shared:
I meet people who ask me how I like my new apartment. It’s too small, but I tell myself, I can’t change it, I have to adapt. There you go. My head speaks to me a lot, I need to have patience. It’s not tomorrow when everything will fall into place. It will take time.
Establishing a routine also seemed like an important step in developing feelings of belonging. This included socializing with others but not overstepping boundaries. One participant explained: “I noticed here that people don’t go from apartment to apartment, and I like that.” Another explained how easy it was to be with others: “At 7PM, if no one calls, I go downstairs, and there’s someone there to play cards. You know, I think we’re a group, it seems like we can talk to each other if we need to talk and we play cards.” For those with hearing or sight impairments, socialization can be challenging and still for others, they feel like they don’t belong. One important element is that the move was not just a move but much more. One participant explained: “I knew it would be a major upheaval. It’s not a move, it’s a life change. It’s not really a move, I can’t count how many times I’ve moved in my life, but this is a major upheaval. I know when I leave here it’s probably going to be feet first.” This followed with a discussion about adapting to an aging self and the realities of aging as part of their relocation. Table 2 presents additional illustrative quotes from study participants.

Table 2
Additional Participant Quotes



Even though 79% of participants had a post-secondary education and all had an adequate income to access private supports before relocating, results indicate that social frailty may have been present before relocation and may have played a role in deciding to relocate. Even for this relatively advantaged group, access to services for aging in place remained challenging and inadequate which resulted in relocation. The post relocation administration of the PRISMA-7 Frailty Scale with a mean group score result of 3.1 suggests that prior to relocation, participants were at the crossover point of being vulnerable to experiencing additional deficits which would potentially have led to higher frailty had they not relocated, as a score of 3 or higher indicates a need for further assessment (17). By addressing social frailty through relocation, participants potentially alleviated multiple factors leading to the social frailty experienced prior to relocation (21). Two important factors that participants identified as forcing them to relocate were loss of social support networks, described as difficulties accessing services and death of a partner/primary caregiver, as well as a sudden change in health status in the self or partner. Other factors mentioned included transportation issues (loss of a driver’s licence) and feelings of insecurity, both previously recognized as components of aging in place (9).
Most of the participants in this study made the choice to relocate, and their new home offered services in their preferred language, a location close to friends and family, and the availability of a continuum of housing and care. Although the transition to this more supportive environment seemed to alleviate social frailty it remained that participants were required to adapt to a new environment, and establish new routines, new relationships and new patterns of socialization. Participants’ ongoing appraisal of their own resiliency, such as strong communication skills, affiliative personalities and favourable health status, combined with the unpredictability of the residential environment could influence their coping mechanism (22). This could potentially explain why participants failed to socialize more since relocation despite previous findings stating that one pull factor in relocation is increased socialization (23). Of particular interest is the finding that both before and after relocating to a supportive environment, those who require more help with daily activities tend to socialize less. If replicated, this finding may suggest that even in supportive environments, enhanced social support and opportunities for socializing may need to be provided when care needs increase, even with relatively independent older adults.
The results of this study provide a better understanding of the concept of social frailty. Specifically, social support networks, formal support services, transportation, and feeling safe were identified as determining factors of social frailty leading to relocation to a housing continuum community. Moreover, the use of a highly educated and financially comfortable sample of older adults allowed us to explore how decisions to relocate are made when options are relatively unconstrained by socioeconomic concerns. Study participants would have had the financial resources to pay for increased supports in their prior home as well as the education and social privilege to advocate for themselves. While recognizing these contributions, limitations of the study include a non-representative sample, limited statistical power due to the small sample size, and the use of limited and self-reported measures. Further research on social frailty is needed to better understand the relationship between social frailty and physical pre-frailty/frailty. In addition, a longitudinal study of older adults with data collections beginning before a relocation transition and continuing to a few years post relocation in congregated housing would provide additional understanding of both social frailty and the transition process, and how these two constructs interact. Clearly, the role that social frailty plays in older adults’ ability to age in place and the decision to relocate is worthy of future study.


Ethical standards: REB approval from Université de Moncton #1920-011.
Conflcit of interest: No conflict of interest.
Funding: This research was supported by funding from the Canadian Frailty Network (CAT2018-42) and the New Brunswick Health Research Foundation (2018-CFN-1775).



1. Bessa B, Ribeiro O, Coelho T. Assessing the social dimension of frailty in old age: a systemic review. Archives of Gerontology and Geriatrics 2018; 78:101-113. doi: 10.1016/j.archger.2018.06.005.
2. Bunt S, Steverink N, Olthof J, van der Schans P, Hobbelen J. Social frailty in older adults: a scoping review. Eur J of Ageing 2017; 14(13): 323-334. doi: 10.1007/s10433-017-0414-7.
3. Freer K, Wallington SL. Social frailty: the importance of social and environmental factors in predicting frailty in older adults. British J of Community Nurs 2019; 24(10): 486-492.
4. Rasiah J, Cummings G, Gruneir A, Oelke ND, Estabrooks C, Holroyd-Leduc J. Prefraitly in older adults: a concept analysis. International J of Nurs Studies 2020; 108: 1-9. doi: https://doi.org/10.1016/j.ijnurstu.2020.103618.
5. Makizako H, Shimada H, Doi T, et al. Social frailty leads to the development of physical frailty among physically non-frail adults: a four-year follow-up longitudinal cohort study. International J of Env Res and Public Health 2018; 15: 1-9. doi: 10.3390/ijerph15030490
6. Andrew M, Keefe J. Social vulnerability from a social ecology perspective: a cohort study of older adults from the national population health survey of Canada. BMC Geriatrics 2014; 14(90): 1-14. doi: 10.1186/1471-2318-14-90.
7. Makizako H, Shimada H, Tsutsumimoto K, et al. Social frailty in community-dwelling older adults as a risk factor for disability. JAMDA 2015; 16, 1003.e7-1003.e11. doi: http://dx.doi.org/10.1016/j.jamda.2015.08.023.
8. Scheibl F, Farquhar M, Buck J, Barclay S, Brayne C, Fleming J. When frail older people relocate in very old age, who makes the decision? Innovation in Aging 2019; 3(4): 1-9. doi: 10.1093/geroni/igz030.
9. Dupuis-Blanchard S, Gould O, Gibbons C, Simard M, Villalon L. Strategies for aging in place: the experience of language-minority seniors with loss of independence. Global Qualitative Nursing Research 2015; Dec: 1-9. doi: 0.1177/2333393614565187.
10. Park H, Jang I-Y, Lee, H, et al. Screening value of social frailty and its association with physical frailty and disability in community-dwelling older Koreans: aging study of PyeongChang rural area. International J of Env Res and Public Health 2019; 16: 1-11. doi: 10.3390/ijerph16162809
11. Andrew M, Dupuis-Blanchard S, Maxwell C, et al. Social and societal implications of frailty, including impact on Canadian healthcare systems. J Frailty Aging 2018; 7(4): 217-223. doi: 10.14283/jfa.2018.30.
12. Gonzalez-Bautista E, Manrique-Espinoza B, Avila-Funes, J, et al. Social determinants of health and frailty are associated with all-cause mortality in older adults. Salud Publica Mex 2019; 61:582-590. doi: https://doi.org/10.21149/10062
13. Marshall A, Nazroo J, Tampubolon G, Vanhoutte B. Cohort differences in the levels and trajectories of frailty among older people in England. J Epidemiol Community Health 2015; 69(4): 316-321. doi:10.1136/jech-2014-204655
14. St John PD, Montgomery PR, Tyas SL. Social position and frailty. Can J Aging 2013; 32(3): 250-259. doi:10.1017/S0714980813000329
15. Szanton SL, Seplaki CL, Thorpe RJ, Allen JK, Fried LP. (2010). Socioeconomic status is associated with frailty: the Women’s Health and Aging Studies. J Epidemiol Community Health 2010; 64(1): 63-67. doi:10.1136/jech.2008.078428
16. Bouchard L, Desmeules M. Les minorités linguistiques du Canada et la santé. Healthcare Policy 2013; 9: 38-47.
17. Raîche M, Hébert R, Dubois MF. Guide d’utilisation du questionnaire PRISMA-7 pour le repérage des personnes âgées en perte d’autonomie modérée à grave. In : Hébert R, Tourigny A, Gagnon M. Intégrer les services pour le maintien de l’autonomie des personnes. 2004. Québec, Edisem, pp 153-175.
18. Hsieh H-F, Shannon SE. Three Approaches to Qualitative Content Analysis. Qualitative Health Research 2005; 15(9):1277-1288. doi: 10.1177/1049732305276687.
19. British Medical Journal. The t tests. The BMJ Statistics. 2020. https://www.bmj.com/about-bmj/resources-readers/publications/statistics-square-one/7-t-tests. Accessed 1 December 2020
20. de Winter JCF. Using the Student’s t-test with extremely small sample sizes. Practical Assessment Research and Evaluation 2013; 18(10): 1-13.
21. Yunhwan L, Doukyoung C, Kim J, Ki S, Yun J. The predictive value of social frailty on adverse outcomes in older adults living in the community. JAMDA 2020; in press: 1-6. doi: https://doi.org/10.1016/j.jamda.2020.03.010
22. Golant S. Residential normalcy and the enriched coping repertoires of successfully aging older adults. The Gerontologist 2015; 55(1): 70-82. doi: 10.1093/geront/gnu036
23. Ewen H, Hahn SJ, Erickson MA, Krout J. Aging in place or relocation? Plans of community-dwelling older adults. J of Housing for the Elderly 2014; 28:288-309. doi: 10.1080/02763893.2014.930366



U. Suthutvoravut1,2, T. Tanaka1,2, K. Takahashi1, M. Akishita2, K. Iijima1


1. Institute of Gerontology, The University of Tokyo, Tokyo, Japan; 2. Department of Geriatric Medicine, The University of Tokyo, Tokyo, Japan.
Corresponding author: Katsuya Iijima, 8th Building 613, Department of Engineering, Hongo 7-3-1, Bunkyo-ku, Tokyo, Tel: 03-5841-1662, Fax: 03-5841-1662, E-mail: iijima@iog.u-tokyo.ac.jp

J Frailty Aging 2019;8(4)198-204
Published online June 28, 2019, http://dx.doi.org/10.14283/jfa.2019.22



Objectives: Eating alone is related to depression, nutritional risk, and mortality. These effects are also influenced by living status. However, little is known about the relationship between eating alone despite living with family and frailty. This study explores the relationship of eating alone and living status with frailty in community-dwelling older adults. Design: Cross-sectional study. Setting and Participants: Kashiwa city, Chiba prefecture, Japan; randomly selected community-dwelling older adults (aged 65 years and over). Measurements: Eating status was assessed by the question, “Do you eat meals with anyone, at least once a day: yes or no?” Frailty was defined by Kihon Checklist (KCL) score 8 or over. Domains of frailty were divided into instrumental activities of daily living (IADL), physical strength, nutrition, eating, socialization, memory, and mood, based on KCL categories. Binary logistic regression analysis was used, adjusting for age, years of education, chronic diseases, number of teeth and cognitive function. Results: Among the total of 1,914 participants, 49.8% were male, and the overall mean age was 72.9 ± 5.5 years. Of all participants, 56 (5.9%) of men and 112 (11.7%) of women were frail. Older adults who ate alone despite living with others were more likely to be frail (OR 2.49, 95%CI 1.1–5.5 for men and OR 2.16, 95%CI 1.0–4.5 for women). Of particular note, eating and living status were associated with lower physical strength and mood in men, whereas in women these statuses were associated with lower scores for IADL, socialization, memory, and mood. Conclusions: Eating alone despite living with others was associated with high frailty in both genders; however, the pathways were different between genders. These results might help yield a simple, fundamental intervention approach to multifaceted frailty, reflecting gender and associated high-risk domains.

Key words: Frailty, eating alone, community, Kihon Checklist.



Frailty is known as an important geriatric syndrome. It increases risk of negative health outcomes such as falls, hospitalization, institutionalization, and mortality (1, 2). Prevalence of frailty increases with age, and it affects approximately a quarter to half of people over 85 years(3).
Frailty is strongly associated with diet. A systemic review found that frailty has significant relationships with malnutrition and risk of malnutrition (4). Evidence also shows that low intake of protein and of specific micronutrients are risk factors for frailty (5, 6). Protein is essential for producing muscle mass, linking it to the prevention of sarcopenia, which is the major component in the development of frailty. Nevertheless, dietary behavior and nutrition are also influenced by various other factors, such as motivation, abilities, and environmental opportunities (7).
Eating alone in older adults, which is a dietary behavior related to both physiologic and social factors, has become a social concern recently. The presence of others while eating increases the caloric intake of food and is related to healthier food habits (8, 9). Moreover, eating with others maintains the motivation of older adults to eat and cook, and provides them with opportunities for social interaction and connectedness (10). Cross-sectional analyses and a longitudinal study found that eating alone interacts with living status in its relation to depression and that eating with others acts as a specific type of social activity with extra benefits additional to those of social participation in general (11–13). One study also found gender differences in the association of eating alone and living status with low diet quality and unhealthy weight (obesity or underweight) (14). Men who eat and live alone were reported to have higher risk for mortality than men who do not (15).
However, the relationship of eating alone and living status with frailty has rarely been explored despite the potential for eating alone and living status to affect many domains of frailty, such as nutrition, socialization, and mood. The results might pave the way for future studies to yield the new practical way for prevention and treatment of frailty. Therefore, we aimed to examine the relationship of eating alone behavior and living status with frailty in community-dwelling older adults.



Study design

This was a cross-sectional study.


We used baseline data from a cohort study which started in 2012 in the city of Kashiwa, in Chiba prefecture, Japan. A total of 12000 community-dwelling older adults aged 65 years and over who were non-eligible for long-term care were randomly selected from resident register. They were asked by mail to participate in the study and 2044 older adults agreed to participate. The baseline examinations were done from September to November 2012 at welfare centers and community centers. We excluded participants who had missing items of data or impaired cognitive function [Mini-Mental State Examination (MMSE) score ≤ 18].


Eating and living status

Eating and living status at present condition were assessed by self-reported questionnaire with the following questions: “Do you eat your meals with anyone else, at least once a day: yes or no?” and “Do you live with your family: yes or no?” Eating and living status were then crossed to make 4 categories: “eating and living with others” (reference), “eating with others yet living alone,” “eating alone yet living with others,” and “eating and living alone.”


Frailty was assessed using the Kihon Checklist (KCL), a Japanese frailty index, which constitutes a self-reported comprehensive health questionnaire. The KCL includes 25 items regarding these 7 domains: instrumental activities of daily living (IADL), physical strength, nutrition, eating, socialization, memory, and mood. This checklist was found to be closely correlated with frailty as defined by the Cardiovascular Health Study criteria; scores of ≥8 were defined as frail (16). Cut-off points for each domain were adopted from a previous systematic review (17), and scores below the cut-off point suggested low or at risk status in that domain (see Appendix 1).

Sociodemographic variables and social engagement

We obtained participants’ age and years of education with a standardized self-report questionnaire. We added the data to the analysis as continuous variables. The Lubben Social Network Scale-6 was used to measure social ties with friends and family (18).

Medical histories

Number of chronic diseases and history of cerebrovascular disease, hypertension, diabetes, osteoporosis, chronic kidney disease, heart disease, and/or cancer were assessed during interviews by nurses.

Function and mental health

Trained staff evaluated cognitive function using the MMSE. The 15-item Geriatric Depression Scale was used to evaluate depressive symptoms. Having trouble with shopping was evaluated by self-report, with the question “Do you have trouble with shopping: yes or no?”

Nutritional, dietary and oral health status

We measured weight and height in order to calculate body mass index (BMI). Mini Nutritional Assessment-Short From (MNA-SF) assessed nutritional status, using self-report questionnaire, BMI, and MMSE data (19). Food quality was evaluated by number of meals per day and 10-item food diversity questionnaire for frequency of meat or fish and vegetable or fruit intake (20). Food enjoyment and food preparation were assessed by self-report questionnaire. The number of functional teeth were checked by dental hygienists. All the assessments including anthropometric, nutritional status, and eating and living status assessments were performed in 2012.

Statistical analysis

Analyses were stratified by gender because previous literature showed different relationship of eating and living status with health outcomes between genders (13–15). Unpaired student’s t-test, Mann Whitney test and Pearson’s chi-squared test were used to compare baseline characteristics between participants with and without frailty. Binary logistic regression analysis was performed with frailty status as the dependent variable. Model 1 was a non-adjusted model; in model 2, we adjusted for age, years of education, chronic diseases, MMSE, and number of functional teeth. To determine further the causes of the relationship of eating and living status with frailty, binary logistic regression analysis was also performed, using each domain from the Kihon Checklist (IADL, physical strength, nutrition, eating, socialization, memory, and mood domain). The characteristics of each eating/living status group were also compared, by chi-squared test for categorical variables, and ANOVA test and Kruskall-Wallis test for continuous variables, with multiple comparisons. IBM SPSS statistics v 22 for Windows (IBM Japan, Tokyo) was used to perform statistical analysis; P value of <.05 was considered statistically significant.

Ethical considerations

The “Kashiwa study” was approved by the Ethics Committee of the university (#12-8). All participants provided written informed consent.



Characteristics of participants

From the baseline of 2,044 participants, we excluded 130 based on missing data or low MMSE score, as described above, resulting in a final number of 1,914 participants, among whom 49.8% were male and whose overall mean age was 72.9 ± 5.5 years. Among men, the “eating and living alone” group accounted for 4.5%, the “eating alone despite living with others” group for 6.7%, “eating with others yet living alone” for 1.3%, and “eating and living with others” for 87.5%. Among women, the respective percentages were 13.1%, 5.3%, 2.8%, and 78.8%. Of all the participants, 56 (5.9%) of men and 112 (11.7%) of women were frail. In both genders, compared to participants without frailty, participants who were frail ate alone more, had more chronic diseases, had higher depressive score, and ate less meat/fish (Table1). In women, frail participants were also older and had fewer years of education and lower cognitive function and oral status.

Table 1 Characteristics of Participants*

Table 1
Characteristics of Participants*

IADL, Instrumental Activities of Daily Living; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; BMI, body mass index; MNA-SF, Mini Nutritional Assessment- Short Form; SD, standard deviation. *Chi squared test was used for categorical variables and nonpaired t-test and Mann-Whitney test were used for continuous variables. † Data is shown as median (interquartile range)


Association between eating and living status and frailty and its domains

Binary logistic regression models were used to analyze associations of eating and living status with frailty (Table 2). Men who ate alone despite living with others were more likely to be frail in model 2. For women, participants who ate alone yet lived with others or who ate and lived alone were more likely to be frail in unadjusted model, but after adjustment, the association only remained for women who ate alone yet lived with others.

Table 2 Association Between Frailty and Each Variable by Binary Logistic Regression

Table 2
Association Between Frailty and Each Variable by Binary Logistic Regression

CI, confidence interval; MMSE, Mini-Mental State Examination; OR, odds ratio


Table 3 shows the associations of eating and living status with frailty domains after adjusting for age, years of education, and number of chronic diseases. In men, eating and living status were significantly associated with physical and mood domains. Men who ate and lived alone were more likely to have low physical strength. Men who ate alone yet lived with others or who ate and lived alone showed higher frequency of depressive risk. On the other hand, women who ate alone despite living with others were more likely to be impaired in IADL, socialization, memory, and mood domains.

Table 3 Association Between Each Frailty Domain and Eating Alone Combined with Living Status by Binary Logistic Regression*

Table 3
Association Between Each Frailty Domain and Eating Alone Combined with Living Status by Binary Logistic Regression*

CI, confidence interval; OR, odds ratio. (-) could not calculate due to small number; *Adjusted variables: Age, years of education, number of chronic diseases


Gender differences in characteristics based on eating and living status

To further examine the mechanism of the association of eating and living status with frailty, the characteristics of each eating and living status group were compared (Appendix 2). In both genders, the “eating alone yet living with others” group was older, had fewer years of education, was more likely to live with their children and not their spouse, and had low food enjoyment compared to older adults who ate and lived with others. Furthermore, men in this group ate less meat/fish and vegetables/fruits than men who ate and lived with others. In women, the “eating alone yet living with others” group ate less meat/fish, reported having trouble with shopping more often, and had more family members than the “eating and living with others” group. We also compared the overall score on questions of 1-20 of KCL to rule out the collinearity effect of depression (question 21-25 in KCL) and frailty. We found that “eating alone yet living with others” group still had the highest score in both genders.



To our knowledge, this is the first study that examined the relationship of eating alone behavior and living status with frailty in community-dwelling older adults. We found that eating alone despite living with others was associated with higher prevalence of frailty both in men and in women. Regarding relationships with each domain of frailty, we also newly found different associations between men and women. In men, eating and living status were associated with impaired physical strength and mood domain, whereas in women these statuses were associated with impairment in IADL, socialization, memory, and mood domain.
Our results were supported by previous literature in which older people “eating alone yet living with others” were vulnerable, reflected in associations with low nutritional status, depression, and mortality (12, 14, 15). Family meal time provides a sense of belonging and mutual aid for older adults with extra benefits additional to general social participation. (12, 21). The “eating alone yet living with others” group missed these opportunities. In addition, eating alone while living with family could be the consequence of many situations: lack of good relationships among family members especially in different generations, different kind of meal or life style, or living in the same house but separate unit.
From our study, it appears that depression domain could be a major cause of frailty in older adults of both genders “eating alone yet living with others.” Later-life depression and frailty share several pathophysiologic mechanisms: subclinical cerebrovascular disease, chronic inflammation, and dysregulation of hormones (22). Longitudinal studies have found increased risk of frailty in older adults with depressive symptomatology (23). Apart from depression, loneliness—the subjective experience of a shortfall in one’s social resources—might also play a role in developing frailty (24).
The sex difference in the associations with frailty components might be the effect of gender roles and psychosocial factors. In men, our results of low physical strength domain might be from low consumption of food and energy as seen in low frequency of meal and vegetables/fruits consumption in men who ate and lived alone since they tend to lack cooking skill and follow poor dietary behavior. Living alone also affects mood in men more because their previous roles such as management and decision-making authority are lost when living alone (25). In women, the mechanisms involved in IADL, memory, and socialization domains might be due to “lifespace constriction” which is more likely to happen in women. Women’s social role involves taking care of their family, especially providing high-quality meals (26). As a result, women who eat alone yet live with family may experience the loss of this social role, and feel less inspiration to cook or to go out and shop for food. Furthermore, our results showed that this group reported having more trouble shopping than the other groups of women. Therefore, we suspect that they do not get enough support (in this realm or in general) from their family and community. A longitudinal study reported that women who leave their neighborhood less frequently have higher risk of frailty (27).
We did not find a significant relationship of the nutrition domain with eating and living, status because the nutrition domain in the KCL focuses on questions about malnutrition, of which frequency was low in the older adults in our study. However, we found that the “eating alone yet living with others” group consumed meat/fish and vegetables/fruits less frequently than the other groups; protein and specific vitamins have been found to be nutrition components related to mechanisms of frailty (6). Positive social feedback from peers increases expected liking and positive attitudes towards a food (28). Older adults who ate alone despite living with others lacked positive emotional experience with food, and thus did not try to meet social norms around eating (29).
It is important to note that we did not find any association with frailty in the “eating and living alone” group or the “eating together yet living alone” group. These two groups of older adults might be able to cope with stress by adaptation over time, providing a sense of control which reduces the effect of stress and is associated with better health outcomes and desired behavioral changes (30).
A few limitations in our study need to be addressed. First, we used cross-sectional data; therefore, we could not make causal inferences. Frailty might restrain older adults from having meals with others; however, previous studies have found that eating alone behavior leads to depression and underweight, which are strongly related to frailty. Thus, we view it as likely that eating alone behavior could have a causal effect on frailty, as well as vice versa. Second, we used only a single item on eating alone, and thus could not estimate the effect of frequency of eating alone, who the eating partner was, or interaction during mealtime. Third, living with family but eating alone could be the consequence of many situations which might confound the relationship between eating behavior and frailty. Fourth, we found only a low frequency of older adults who ate with others but lived alone, leaving us unable to calculate some relationships.



The results of this study suggest that “eating alone yet living with others” is associated with frailty and its domains in community-dwelling older adults, and further that there are pathways of this association among men and women. This might inspire a simple, fundamental intervention approach to multifaceted frailty, depending on older adults’ gender and high-risk domains. Longitudinal observation and intervention studies should be conducted in the future.


Funding: This work was supported by the Health and Labor Sciences Research Grant (H24-Choju-Ippan-002) from the Ministry of Health, Labor, and Welfare of Japan. The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.
Acknowledgement: We thank all the staff and participants in the Kashiwa study.
Conflicts of Interest: None.




1.    Bergman H, Ferrucci L, Guralnik J, et al. Frailty: an emerging research and clinical paradigm–issues and controversies. J Gerontol A Biol Sci Med Sci 2007; 62(7):731–7.
2.     Walston J, Hadley EC, Ferrucci L, et al. Research agenda for frailty in older adults: toward a better understanding of physiology and etiology: summary from the American Geriatrics Society/National Institute on Aging Research Conference on Frailty in Older Adults. J Am Geriatr Soc 2006; 54(6):991–1001.
3.     Song X, Mitnitski A, Rockwood K. Prevalence and 10-year outcomes of frailty in older adults in relation to deficit accumulation. J Am Geriatr Soc 2010; 58(4):681–7.
4.     Lorenzo-López L, Maseda A, de Labra C, Regueiro-Folgueira L, Rodríguez-Villamil JL, Millán-Calenti JC. Nutritional determinants of frailty in older adults: A systematic review. BMC Geriatr 2017; 15;17(1):108.
5.     Bartali B, Frongillo EA, Bandinelli S, et al. Low nutrient intake is an essential component of frailty in older persons. J Gerontol A Biol Sci Med Sci 2006; 61(6):589–93.
6.     Beasley JM, LaCroix AZ, Neuhouser ML, et al. Protein intake and incident frailty in the Women’s Health Initiative observational study. J Am Geriatr Soc 2010; 58(6):1063–71.
7.     Brug J. Determinants of healthy eating: motivation, abilities and environmental opportunities. Fam Pract 2008; 25 Suppl 1:i50-55.
8.     Hammons AJ, Fiese BH. Is frequency of shared family meals related to the nutritional health of children and adolescents? Pediatrics 2011; 127(6):e1565-1574.
9.     Locher JL, Robinson CO, Roth DL, Ritchie CS, Burgio KL. The effect of the presence of others on caloric intake in homebound older adults. J Gerontol A Biol Sci Med Sci 2005; 60(11):1475–8.
10.     Bofill S. Aging and loneliness in Catalonia: The social dimension of food behavior. Ageing Int 2004; 29(4):385–98.
11.     Kimura Y, Wada T, Okumiya K, et al. Eating alone among community-dwelling Japanese elderly: association with depression and food diversity. J Nutr Health Aging 2012; 16(8):728–31.
12.     Kuroda A, Tanaka T, Hirano H, et al. Eating Alone as Social Disengagement is Strongly Associated With Depressive Symptoms in Japanese Community-Dwelling Older Adults. J Am Med Dir Assoc 2015; Jul 1;16(7):578–85.
13.     Tani Y, Sasaki Y, Haseda M, Kondo K, Kondo N. Eating alone and depression in older men and women by cohabitation status: The JAGES longitudinal survey. Age Ageing 2015; 44(6):1019–26.
14.     Tani Y, Kondo N, Takagi D, et al. Combined effects of eating alone and living alone on unhealthy dietary behaviors, obesity and underweight in older Japanese adults: Results of the JAGES. Appetite 2015; 95:1–8.
15.     Tani Y, Kondo N, Noma H, Miyaguni Y, Saito M, Kondo K. Eating Alone Yet Living With Others Is Associated With Mortality in Older Men: The JAGES Cohort Survey. J Gerontol B Psychol Sci Soc Sci 2017; Jan 15; pii: gbw211.
16.     Satake S, Senda K, Hong Y-J, et al. Validity of the Kihon Checklist for assessing frailty status. Geriatr Gerontol Int 2016; 16(6):709–15.
17.     Sewo Sampaio PY, Sampaio RAC, Yamada M, Arai H. Systematic review of the Kihon Checklist: Is it a reliable assessment of frailty? Geriatr Gerontol Int 2016; 16(8):893–902.
18.     Lubben J, Blozik E, Gillmann G, et al. Performance of an abbreviated version of the Lubben Social Network Scale among three European community-dwelling older adult populations. The Gerontologist 2006; 46(4):503–13.
19.     Kaiser MJ, Bauer JM, Ramsch C, et al. Validation of the Mini Nutritional Assessment short-form (MNA-SF): a practical tool for identification of nutritional status. J Nutr Health Aging 2009; 13(9):782–8.
20.     Kumagai S, Watanabe S, Shibata H, et al. [Effects of dietary variety on declines in high-level functional capacity in elderly people living in a community]. Nihon Koshu Eisei Zasshi Jpn J Public Health 2003; 50(12):1117–24.
21.     Vesnaver E, Keller HH. Social influences and eating behavior in later life: a review. J Nutr Gerontol Geriatr 2011; 30(1):2–23.
22.     Vaughan L, Corbin AL, Goveas JS. Depression and frailty in later life: a systematic review. Clin Interv Aging 2015; 10:1947–58.
23.     Lakey SL, LaCroix AZ, Gray SL, et al. Antidepressant use, depressive symptoms, and incident frailty in women aged 65 and older from the Women’s Health Initiative Observational Study. J Am Geriatr Soc  2012; 60(5):854–61.
24.     van Baarsen B, Snijders TAB, Smit JH, van Duijn MAJ. Lonely but Not Alone: Emotional Isolation and Social Isolation as Two Distinct Dimensions of Loneliness in Older People. Educ Psychol Meas 2001; 61(1):119–35.
25.     Mouodi S, Bijani A, Hosseini SR, Hajian-Tilaki K. Gender differences in the health status of elderly living alone compared to those who are not alone: Evidence of the AHAP study, North of Iran. Caspian Journal of Internal Medicine. 2016;7(2):126-132.
26.     Fürst EL. Cooking and femininity. Womens Stud Int Forum 1997; 20(3):441–9.
27.     Xue Q-L, Fried LP, Glass TA, Laffan A, Chaves PHM. Life-space constriction, development of frailty, and the competing risk of mortality: the Women’s Health And Aging Study I. Am J Epidemiol 2008; 167(2):240–8.
28.     Robinson E, Higgs S. Liking food less: the impact of social influence on food liking evaluations in female students. PloS One 2012; 7(11):e48858.
29.     Higgs S, Thomas J. Social influences on eating. Curr Opin Behav Sci 2016; 9:1–6.
30.     Thompson SC, Spacapan S. Perceptions of Control in Vulnerable Populations. J Soc Issues 1991; 47(4):1–21.