Aging in Place in a Depopulated, Mountainous Area: The Role of Hometown-Visiting Family Members in Shimogo, Japan

Abstract

In this study, we examined the factors that contribute to successful aging in place in the town of Shimogo, a depopulated mountainous area in Japan, focusing on the interplay between household support types, the performance of daily activities, and the involvement of geographically dispersed family members. A mixed-methods approach was used, combining survey data with multiple correspondence analysis and cluster analysis. The resident household findings reveal significant differences in community engagement and reliance on community support across three household support types: self-help, mutual aid, and combined support. The self-help households demonstrated high self-sufficiency and community participation, while the mutual-aid and combined-support households exhibited greater reliance on family and community resources. An analysis of the activities of out-migrated relatives uncovered diverse engagement profiles, motivated by both providing practical support and strengthening family bonds. The visit frequencies and activity types of the relatives were significantly influenced by both their own characteristics and the levels of self-reliance and community support needs of the resident households. Four distinct relative engagement clusters emerged, reflecting varying levels of support, community involvement, and visit frequency. Our findings underscore the crucial role of both informal (family) and formal (community) support networks in facilitating successful aging in place in rural depopulated areas. Policy implications include strategies supporting both self-reliance and the diverse mechanisms of family- and community-based support to ensure the sustainability of life in these challenging environments.

1. Introduction

1.1. The Global Aging Trend and Japan’s Unique Demographic Profile

The global increase in life expectancy and decline in fertility rates [1] have created a demographic imbalance in developed nations, placing considerable strain on social support systems and hindering economic growth. This imbalance is particularly pronounced in Japan, where a high proportion (29.1%) [2] of the population is aged 65 years and older, including 13.0% aged 65–74 years and 16.1% aged 75 years and older [3]. This aging population, coupled with a shrinking workforce, resulted in a critical support ratio of 2.0 working-age individuals per elderly person in 2023, which is projected to fall to 1.3 by 2070 [2]. Japan’s record-low total fertility rate of 1.20 in 2023 [4] further compounds these challenges, especially in remote mountainous areas that are experiencing rapid population aging. In this study, we investigated how individual coping strategies and support from out-migrated family members contribute to the sustainability of aging residents’ lives and the maintenance of community functions in the town of Shimogo, a rural community in Fukushima Prefecture, Japan, where these pressures are especially acute.

1.2. Rural Depopulation Challenges

Rural depopulation in Japan, particularly in mountainous areas, poses significant challenges to the sustainability of both community functions and the livelihoods of residents. Driven by the out-migration of younger generations seeking opportunities in urban areas since the 1960s, this trend has led to shrinking workforces and an aging population [5]. This “regional shrinkage” [6], a self-reinforcing cycle of depopulation, economic decline, and social disruption, creates significant obstacles for these communities. These regions, often designated as “depopulated areas” (Kaso regions) by the Japanese government [5], frequently face insufficient local government funding, limited transportation, and a lack of essential services, highlighting the urgent need for targeted support strategies to ensure their long-term sustainability.

1.3. “Aging in Place”: A Framework for Rural Sustainability

“Aging in place”, a concept that emphasizes the ability of older adults to remain in their homes and communities [7], is crucial for understanding the sustainability of rural communities, particularly those facing depopulation and limited resources. This concept centers on maintaining independent living through access to necessary support networks, both formal, such as medical institutions, and informal, such as family and community ties [8]. While older adults, particularly those aged 80 years and above, strongly desire to age in place [9], which stems from a deep connection to home and community [10], they often face significant obstacles in rural Japan. Limited access to vital services such as healthcare, social services, and transportation, exacerbated by economic decline and a shrinking workforce, creates hurdles to successful aging in place [11]. In this study, we explored how the support from out-migrated families, a crucial informal support, impacts aging in place in Shimogo, considering individual needs and the available community resources.

2. Theoretical Framework

The understanding of rural depopulation in Japan has evolved from an initial focus on the disappearance of communities to a nuanced recognition of their adaptive resilience, with particular emphasis on the critical role of family support. This theoretical framework builds on extensive prior research, reflecting a progression in perspectives that underscores the urgency and importance of examining how communities can sustain themselves amidst demographic decline. Figure 1 illustrates the theoretical framework guiding this research.

Figure 1. Theoretical framework for aging in rural communities. This diagram illustrates the relationship between demographic transition challenges, community resilience, and the role of out-migrated family support in ensuring aging in place for elderly residents in rural communities. The arrows indicate the mechanisms under investigation in this study.

2.1. Reframing “Marginal Communities”: Towards Resilience

The understanding of rural depopulation in Japan has shifted from early predictions of widespread community disappearance to the recognition of adaptive resilience and the essential role of family support, reflecting a paradigm shift in how researchers view rural sustainability.

In early studies, researchers examined the processes leading to the functional decline and eventual disappearance of communities struggling to maintain sustainability. Before the publication of the “Masuda Report (2014)” [12], which reverberated across Japanese society with its stark prediction that nearly one-half of Japan’s municipalities will disappear naturally based on projected demographic trends, scholars had already begun to highlight the resilience of rural communities and the importance of family support. Ohno (2005) defined “marginal communities” (Genkai Shuraku) [13] as settlements with over 50% of their populations aged 65 years and older that are facing significant out-migration and struggling to maintain social functions, a characterization further developed by Sakuno (2008) through a categorization based on population decline, household size, and aging rates [14]. Hatamoto (2010) challenged the purely negative perspective, advocating for policies that prioritize community life over large-scale revitalization efforts [15]. Kudo and Yarime (2013) further expanded this debate, emphasizing the interplay between demographic shifts, self-sufficient farming, and the continued support of out-migrated family members in efforts to sustain rural communities [16]. These works collectively foreshadowed a paradigm shift, reframing depopulation not as inevitable decline but as an opportunity to understand community resilience.

Subsequent research, especially since the publication of the “Masuda Report,” challenged this deterministic view by underscoring the capacity for adaptation and resilience in rural areas. Scholars such as Yamamoto (2014) [17] critiqued the reliance on aging rates as a sole criterion, while others, including Odagiri (2014) [18], Yamashita (2014) [19], and Okada (2015–2016) [20], emphasized that focusing solely on population loss overlooks positive transformations and the importance of sustaining existing lifestyles. Despite depopulation, the cyclical movement of people—often to second homes—persists in rural areas [21]. Tokuno and Kashio (2014) further demonstrated that even “marginal” settlements can maintain fulfilling lifestyles through connections with nearby areas and out-migrated family support [22]. Altogether, these findings underscore the resilience and adaptability of rural communities and the indispensable role of family ties in mitigating the threats of depopulation and disappearance.

2.2. The Significance of Family Support in Rural Communities

Several researchers have explored the resilience of rural communities in Japan, focusing on the continued contributions of out-migrated family members to key community functions, such as agricultural production [23], the preservation of traditional cultural events, and community governance [24,25]. However, driven by the large number of suburban housing developments constructed during Japan’s period of rapid economic growth, the research on intergenerational support from geographically dispersed family members has primarily focused on suburban areas [26,27,28], leaving rural contexts significantly under-researched. Moreover, rural elders face difficulties with succession, infrequent family contact, and service access [29]. While small-scale surveys have highlighted the role of out-migrated family members in maintaining village functions [30], the study did not specifically focus on the support provided to aging households in depopulated rural areas. This study fills a critical gap by investigating how geographically dispersed family members contribute to the resilience and sustainability of rural communities facing demographic decline.

2.3. Research Questions and Framework Overview

While much research on family caregiving focuses on co-resident family members, the role of geographically dispersed relatives in supporting elderly individuals in rural areas remains underexplored.

Building on the theoretical framework in Figure 1, this study investigates how visiting relatives contribute to the daily life sustainability and overall community resilience of aging residents in Shimogo.

The research objectives were as follows: (1) to describe the demographic characteristics, household compositions, health conditions, and support networks of elderly residents in Shimogo; (2) to analyze the frequencies, durations, and motivations of homecoming visits by relatives; (3) to examine the types and frequency of support provided by visiting relatives, including their contributions to daily tasks, social activities, and community events; (4) to assess the life sustainability of elderly residents and explore the relationships between individual capabilities, family support, and community resources; and (5) to identify potential gaps in the support and suggest recommendations for policies and community initiatives that could enhance the sustainability of life for elderly residents in Shimogo and similar rural communities.

3. Materials and Methods

3.1. Data Collection

In this study, we employed a mixed-methods approach to investigate the living conditions of elderly residents in Shimogo, a depopulated, mountainous rural area, uniquely focusing on the support from out-migrated family members—a crucial aspect under-researched in previous studies. Initial semi-structured interviews with Shimogo officials provided critical contextual information. These interviews highlighted the significant challenges of directly contacting out-migrated individuals due to inconsistencies in the local registration system: some had completely deregistered, while others, despite remaining registered, no longer resided in Shimogo. This crucial insight shaped this study’s methodology. A comprehensive, anonymous survey, distributed and collected via the Town Office to ensure confidentiality and minimize the respondent burden, targeted all households with at least one resident aged 65 years or older, as identified by Town Office staff. The survey focused on the daily behaviors of the residents and the support received from their out-migrated family members during hometown visits, information that was gathered indirectly through resident households—a practical and privacy-sensitive approach given the difficulties of contacting those removed from the local registration system. Participation was voluntary, and the respondents were informed that the data would be used solely for research purposes. In total, we collected 378 completed surveys.

3.2. Questionnaire Design

Despite the shared reliance on family support, the challenges of aging in place differ significantly between urban and rural Japan. In urban areas, institutional care typically becomes necessary when health deteriorates and the existing home-based and family support proves insufficient [31]. However, in rural areas, the strong reliance on family care [32] often delays this transition until the physical limitations of the elderly necessitate alternative care arrangements, a situation that is exacerbated by out-migration and results in increased instances of caregiving among older adults [33]. This contrast necessitates research on the specific support provided by out-migrated family members in rural settings. In this study, we employed a questionnaire focusing on objective behavioral indicators rather than a subjective satisfaction scale to achieve two key aims: first, the capture of the daily activities of elderly residents and the specific actions of their out-migrated family members during visits; second, the minimization of the respondent burden and the avoidance of biases inherent in subjective satisfaction measures by focusing on easily reportable behaviors rather than on subjective feelings, recognizing that identical visit frequencies could yield vastly different satisfaction levels depending on individual circumstances and expectations [34]. Thus, the questionnaire prioritized the collection of data that reflected the actual support provided rather than potentially inconsistent subjective assessments.

The existing research on urban and suburban lifestyles among retired elderly individuals offers extensive insights into their various life behaviors within their local spheres, including routine shopping and regular healthcare appointments, and further explores individual recreation and other personally driven activities [35]. However, to investigate the unique aspects of rural Japanese life, where farming practices are often centered on older adults [36] and community activities encourage broad participation [37], we developed a questionnaire drawing upon this existing research while incorporating categories relevant to the potential contributions of out-migrated family members. These categories, encompassing key aspects of daily life in rural communities, are detailed in Table 1, which presents a classification of life behaviors, including essential activities (daily living), productive activities (farming and yard work), social and leisure pursuits, and community engagement. The main contents of the questionnaire are presented in Table 2.

Table 1. The rural life behaviors of the elderly residents in this survey.

Life Behaviors in Rural Communities	Items
Daily Living Activities (maintaining basic life functions)	Medical care, shopping (daily necessities)
Productive Activities (contributing to self-sufficiency and economic activity)	Farm work, yard work
Social and Leisure Activities (fulfilling personal interests and social needs)	Spending time with friends (e.g., teatime), engaging in hobbies, etc.
Community Engagement (maintaining community functions)	Community meetings, clean-up, festivals, etc.

Based on existing research [35], this study categorized four types of life behaviors that potentially involve participation by both residents and out-migrated family members, considering the unique characteristics of rural communities.

Table 2. Main contents of the survey questionnaire distributed in Shimogo.

Questionnaire Summary	Resident Households 1	Hometown Visit Relatives 2
Basic Information 3	1. Residential area (name of settlement)	5. Location of Visit Relatives
	2. Age and Gender of all family members	6. Age 7. Frequency of Hometown Visit
Healthcare and Shopping 4	3. Health conditions: Select the health conditions that apply to your family (multiple choice). 3.1. All family members are in good health 3.2. Need to go to the hospital regularly 3.3. Uses a home visit helper 3.4. Uses a day service (bathing, rehabilitation, etc.). 3.5. Uses home visit medical services 3.6. Need caregiving.
	4. Shopping (medical care) methods: Select the shopping (medical care) conditions that apply to your family (multiple choice) 4.1. Completed by family without others’ help. 4.2. Requires relatives to go shopping (go to hospital) with family members during their hometown visit. 4.3. Requires nearby neighbors to go shopping (go to hospital) with family members. 4.4. Relatives bring the items to our family. 4.5. Neighbors bring the items to our family. 4.6. Use mobile vending vehicles in communities. 4.7. Use online services from co-ops, Amazon, and others.
Activities During Hometown Visits		8. Household Chore Support: Frequency of cleaning the house, doing laundry, and cooking during hometown visiting.
		9. Outdoor Leisure Activities: Frequency of engaging in Outdoor Leisure Activities during Hometown Visits, including: Barbecue (BBQ) Foraging for Wild Plants (Mountain Vegetables), Fishing, Hiking, Walking/Strolling, Carpentry Work, Agricultural Work
Frequency of Daily Shopping	10A. Daily Essentials Shopping	10B. Providing Daily Essentials Shopping
Frequency of Agricultural Work	11A. Yard Work around house 12A. Farmland Cultivation, including: Planting in Pots, Gardening Home Vegetable Gardening	11B. Providing Yard Work 12B. Providing Farmland Cultivation, including Planting in Pots, Gardening, Home Vegetable Gardening
Frequency of Personal Social Activities	13A. Social Interactions and Leisure Activities in the Community, Including: Spending Time with Friends (e.g., Teatime) Engaging in Hobbies Going Out	13B. Social Interactions and Leisure Activities in the Community during hometown visits, Including: Spending Time with Friends (e.g., Teatime) Engaging in Hobbies Going Out
Frequency of Community Activities	14A. Community Activities, including: Community Lawn Mowing Community Clean-Up (Trash Collection) Local Festivals	14B. Participating on Behalf of Residents with Community Activities, including: Community Lawn Mowing Community Clean-Up (Trash Collection) Local Festival

To ensure data consistency and minimize the respondent burden, several survey design choices were implemented. ¹ Responses were collected from a household perspective, ² focusing on the most frequent visitor when multiple relatives visited. ³ Open-ended questions allowed for the accurate recording of the settlement names and locations of visiting relatives not included in the provided options. To protect privacy, only age and gender, not specific relationships, were collected for household members. ⁴ Household health status was based on the most severe health condition reported among members. All Shopping (medical care) methods employed by household members were recorded.

3.3. Questionnaire Analysis

In this study, we employed a two-stage multiple correspondence analysis (MCA) to investigate the interplay between the resident household characteristics and the activities of visiting relatives in Shimogo [38,39]. This approach was deemed necessary due to the likely divergence in the factors that influence the actions of residents and relatives. To minimize the subjective bias and respondent burden, all the variables were categorical, measuring the frequencies of specific behaviors rather than relying on subjective scales [40].

Following the descriptive analyses of the household characteristics (Section 4.1), healthcare access (Section 4.2), and daily life activities (Section 4.3), two separate MCAs were conducted. The first MCA, which focused on resident capabilities, examined the relationships between the household composition (Q2), health status (Q3), and various daily activities (Q10A, Q11A, Q12A, Q13A, Q14A) (Section 3.4). The second MCA, which investigated the factors driving the activities of relatives, analyzed the relationships between the demographic factors (age (Q5), location (Q6)), support provided (Q8, Q9, Q10B, Q11B, Q12B), and participation in leisure and community activities (Q13B, Q14B) (Section 4.5). A subsequent cluster analysis (Section 4.6) was then performed on the object scores from the second MCA to identify distinct groups of relatives based on their activity patterns.

3.4. Study Area

We focused on Shimogo, a town located at the eastern edge of the Minami–Aizu District in the southwestern part of Fukushima Prefecture (Figure 2). Shimogo borders Tochigi Prefecture to the south and the city of Aizu–Wakamatsu to the north. According to data from the Shimogo Town Office for the 2023 fiscal year, the municipality has a total population of 5071 individuals residing in 2165 households [41]. According to the 2020 national census data, the population of individuals aged 65 years and over in Shimogo constitutes 44.8% of the total population [42], indicating a significantly higher aging rate compared to Fukushima Prefecture as a whole, which has a 33.8% aging rate [43]. Therefore, Shimogo is a region within Fukushima Prefecture that is experiencing an accelerated aging trend. Shimogo is accessible by private railway, with connections from Tokyo via the Tobu Railway and from Aizu–Wakamatsu via the Aizu Railway, and it is about 37 km from the Shirakawa IC and 35 km from Aizu–Wakamatsu by car.

Figure 2. Location and transportation to Shimogo in Fukushima Prefecture (created by authors).

Fukushima Prefecture exhibits a significant disparity in its aging rates, with the western Aizu and Minami–Aizu districts showing considerably higher rates (over 40% in Minami–Aizu; Figure 3a) than the more densely populated central region, where major cities are located (under 35%; Figure 3b). Shimogo, situated in Minami–Aizu, is particularly relevant to this study, exhibiting an exceptionally high aging rate of 44.8% in 2020, exceeding both the prefectural (33.3%) and national (28.4%) averages. Further emphasizing its suitability as a case study, Shimogo’s population experienced a decline of over 23% between 1975 and 2015, in addition to having an aging rate exceeding 35% and meeting the Ministry of Internal Affairs and Communications’ criteria for depopulated areas [5], which include a high aging rate, a low population density, economic decline, and limited public services. This is further illustrated in Figure 4a,b, which show that while residents are distributed along national highways and in more remote mountainous areas, key infrastructure is concentrated in central Shimogo, resulting in inadequate service coverage for many. These factors—the high aging rate, location in a high-aging region, proximity to designated depopulated areas, and demonstrable infrastructure limitations—make Shimogo a compelling case study.

Figure 3. Population trends in Shimogo and Fukushima Prefecture (created by authors; data sources: Shimogo Town Office, Fukushima Prefecture Statistics Department, and 2020 National Census): (a) proportion of population aged 65 years and over in Fukushima Prefecture; (b) population density in Fukushima Prefecture. Areas affected by radiation from the atomic bomb were excluded from the analysis.

Figure 4. Spatial analysis of Shimogo: population and infrastructure (created by authors, data sources: https://mieruka.city; Google Earth, accessed on 1 September 2024): (a) nighttime population density map of Shimogo; (b) distribution map of key infrastructure in Shimogo.

4. Results

4.1. Characteristics of Elderly Resident Households in Shimogo Town

We employed a comprehensive survey to investigate the daily lives of elderly residents in Shimogo and the contributions of their homecoming relatives. Questionnaires were distributed to all households with at least one member aged 65 years or older, with responses received from all areas within Shimogo (Figure 5). Based on data from the Ministry of Internal Affairs and Communications (MIC, 2020, accessed on 1 September 2024) and the 2023 Basic Resident Register Survey, the estimated response rate for the target population was approximately 41.4% (378 completed surveys/913 estimated eligible households) [42,43]. The distribution of the responses across settlements is detailed in Table 3.

Figure 5. Shimogo survey response map (created by authors; data sources: 2020 National Census data, agricultural settlement boundary data from the Ministry of Agriculture, Forestry and Fisheries’ 2020 Agricultural Census).

Table 3. Basic information on resident households.

Survey Questions	Survey Items		Details and Proportion
1. Location of Resident Households by Settlements N = 378	Settlement Name		Quantity	Proportion
	Toyonari		72	19.1%
	Eitomi		40	10.6%
	Yunokami		33	8.7%
	Shiono		31	8.2%
	Nakaduma		24	6.3%
	Ohmatsukawa		22	5.8%
	Yagoshima		20	5.3%
	Takashima		19	5.0%
	Otokane		17	4.5%
	Aikawa		15	4.0%
	Ochiai		15	4.0%
	Shiraiwa		13	3.4%
	Sawada		10	2.6%
	Ikuda		8	2.1%
	Kuwatobi		7	1.9%
	Ouchi		7	1.9%
	Toaka		7	1.9%
	Nagurasawa		6	1.6%
	Onumazaki		4	1.1%
	Nakayama		3	0.8%
	Ohzawa		3	0.8%
	Edamatsu		2	0.5%
2. Household Composition N = 376	Categorizations of Household Composition		Details of Demographic Characteristics	Quantity (Proportion)
	Single Senior household N = 146 (38.8%)		Gender
			Male	66 (45.2%)
			Female	75 (51.4%)
			Gender Not Provided	5 (3.4%)
			Age
			Early Senior (65–79)	89 (61.0%)
			Later Senior (≥80)	55 (37.6)
			Age Not Provided	2 (1.4%)
	Multi-Senior Households		Same-Generation Senior Household	172 (85.6%)
	N = 201 (53.5%)		Intergeneration Senior Household	29 (14.4%)
	Intergenerational Households		Two-Generation Household	13 (44.8%)
	N = 29 (7.7%)		Three-Generation Household	16 (55.2%)
3. Health Conditions 2 N = 357	Categorizations of Health Conditions 2		Options (Multiple-Choice Questions) 1	Selection Rate 1
	Healthy Households N = 102 (28.5%)		3.1. All Family Members are In Good Health	102 (28.5%)
	Self-Reliant Medical Care Households N = 207 (58.0%)		3.2. Require Regular Medical Attention	237 (66.4%)
	Support-Reliant Households N = 48 (13.5%)	Support-Needing Households N = 16 (4.5%)	3.3. Utilize In-Home Care Services (for basic assistance)	9 (2.5%)
			3.4. Attend Adult Day Care Programs	29 (8.1%)
		Care-Needing Households N = 32 (9.0%)	3.5. Receive Home Visit Medical Services	3 (0.8%)
			3.6. Require Long-Term Care Assistance	29 (8.1%)
4. Shopping Medical-care Methods 3 N = 364	Categorizations of Shopping Medical-care Methods 3		Options (Multiple-Choice Questions) 1	Selection Rate 1
	Self-help N = 233 (64.0%)		4.1. Shop (Seek Medical Care) Independently with Cohabiting Family	305 (83.8%)
	Mutual aid N = 38 (10.4%)		4.2. Shop (Seek Medical Care) with Hometown Visiting Relatives	50 (13.7%)
			4.3. Shop (Seek Medical Care) with Community Residents	43 (11.8%)
			4.4. Have Shopping Done on Their Behalf by Hometown Visiting Relatives	18 (4.9%)
			4.5. Have Shopping Done on Their Behalf by Community Residents	11 (3.0%)
	Combined support N = 93 (25.6%)		4.6. Utilize Community Mobile Vending for Shopping	27 (7.4%)
			4.7. Utilize Home Delivery Services (e.g., Co-op) for Shopping	36 (9.9%)

¹ Selection rates for options in multiple-choice questions 3 and 4 are shown in Table 3. ² Household Health Conditions are categorized as follows: Healthy Households (all family members in good health, no formal long-term care; selection of option 3.1 only); Self-Reliant Medical Care Households (one or more family members require regular medical attention but no formal long-term care; selection of option 3.2 only); and Support-Reliant Households (utilizing formal long-term care services; selection of any combination of options 3.3–3.6). Support-Reliant Households are further categorized as Support-Needed (options 3.3 or 3.4) and Care-Needed (options 3.5 or 3.6) Households. ³ This categorization focuses on the primary method of obtaining daily necessities and healthcare: self-help (option 4.1 only); mutual aid (options 4.2–4.5 alone or in combination, highlighting the reliance on relatives and neighbors); and combined support (all other combinations of options 4.1–4.7), reflecting the complex support networks observed in Shimogo.

Elderly residents in Shimogo have adapted their lifestyles to the challenges of this mountainous area, developing diverse strategies for accessing healthcare and essential goods and maintaining their daily lives (Table 3, Figure 5).

The household composition data (Q2) reveal a community significantly impacted by population aging and out-migration, as evidenced by the high proportion of single-senior households (39.4%, with 37.7% headed by those aged 80 years or older) and multi-senior households consisting primarily of same-generation seniors (85.6% of the 54.2% of multi-senior households). This pattern strongly suggests that younger generations have largely migrated out, leaving many older adults living alone or with other older adults. The smaller percentage (14.4%) of multi-senior households with generational age gaps may reflect instances of retired adult children returning to care for their aging parents, indicating that out-migrated family members themselves are aging. The low proportion of inter-generational households (6.5%) further emphasizes the significant impact of this out-migration on household structures.

Access to healthcare also reflects the adaptation to the geographic context (Q3). While 28.5% of households required no formal long-term care (“Healthy Households”), the majority (66.4%) needed regular medical attention but managed without formal support (“Self-Reliant Medical Care Households”). Only 13.5% relied on formal long-term care (“Support-Reliant Households”).

An examination of how households access daily necessities and healthcare (Q4) further reveals these adaptive strategies, particularly concerning the significant contributions of out-migrated family members. Three distinct patterns emerged: self-help households (64.0%), relying solely on household members; mutual-aid households (10.4%), utilizing relatives or neighbors for assistance; and combined-support households (25.6%), employing combinations of independent efforts, relative assistance, and com-munity services. An analysis of the mutual-aid group reveals a substantial reliance on relatives for support, with those receiving assistance (13.7% with relatives; 11.8% with neighbors) exceeding those who have tasks performed on their behalf (4.9% with relatives; 3.0% with neighbors). This highlights the significance of mutual aid, particularly in instances in which the independent management of daily necessities and healthcare is not feasible; receiving assistance directly (going together) was a far more frequent response than having tasks performed on one’s behalf.

4.2. Characteristics of Visiting Relatives and Their Hometown-Visiting Patterns

Out-migrated family members maintain consistent connections with Shimogo, providing support to elderly residents and engaging in personal leisure activities (Table 4). Their geographic distribution (Q5; Figure 6 and Figure 7) reveals strong ties within Fukushima Prefecture, particularly the Aizu region (50% of relatives), with significant representation in Shimogo (14.6%, indicating semi-proximity living), the city of Aizuwakamatsu (26%), and the Kantō region (primarily Tokyo) (29.2%). This dispersed geographic distribution, coupled with a diverse age profile (Q6; nearly one-half under 50 years, one-third between 50 and 64 years, and a smaller but significant portion 65+ years) suggests that factors beyond the local aging population, such as family obligations and proximity, significantly influence the support provision.

Table 4. Characteristics of visiting relatives and their hometown-visiting patterns.

Survey Questions	Survey Items	Details and Proportion
5. Location of Visit Relatives N = 342	Wide range division	Prefecture/City/Town	Proportion
	Aizu area 1 N = 171 (50.0%)	Aizuwakamatsu City Shimogo Town Minamiaizu Town Others	89 (52.1%) 50 (29.2%) 20 (11.7%) 12 (7.0%)
	Other areas in Fukushima Prefecture N = 52 (15.2%)	Fukushima City Koriyama City Shirakawa City Others	13 (25.0%) 11 (21.2%) 11 (21.2%) 17 (32.6%)
	Kanto Region 2 N = 100 (29.2%)	Tokyo Kanagawa Chiba Saitama Tochigi Gunma Ibaraki	46 (46.0%) 15 (15.0%) 11 (11.0%) 11 (11.0%) 10 (10.0%) 4 (4.0%) 3 (3.0%)
	Other Regions N = 19 (5.6%)	Miyagi Others	9 (47.4%) 10 (52.6%)
6. Age N = 335	Under 50 50~64 65~74 Over 75		163 (48.7%) 109 (32.5%) 48 (14.3%) 15 (4.5%)
7. Frequency of Hometown Visit N = 349	More than once a week More than once a month More than once a year Less than once a year		66 (18.9%) 101 (28.9%) 146 (41.8%) 36 (10.3%)
8. Frequency of Chores Support N = 305	More than once a week More than once a month More than once a year Almost never		23 (7.5%) 41 (13.5%) 87 (28.5%) 154 (50.5%)
9. Outdoor Leisure Activities N = 312	More than once a month Several times a year Almost never		31 (9.9%) 168 (37.8%) 113 (52.2%)

¹ The Aizu area includes both the Aizu District and Minami Aizu District (Figure 3a and Figure 6). ² The Kanto Region does not have a legally defined boundary. In this study, the Kanto Region refers to the prefectures of Tokyo, Ibaraki, Tochigi, Gunma, Saitama, Chiba, and Kanagawa (Figure 5).

Figure 6. Geographic distribution of visiting relatives across Japan.

Figure 7. Geographic distribution of visiting relatives in Fukushima Prefecture.

The visit frequency (Q7) demonstrates a high level of engagement, with almost 90% visiting more than once annually, reflecting both cultural obligations (New Year, Obon) and life events. However, a substantial portion (47.8%) visit more than once monthly, suggesting a strong ongoing commitment, with almost 20% visiting more than once weekly. Chore assistance (Q8) is less frequent, with over one-half reporting that they almost never assist, yet significant numbers provide help more than annually (28.5%) or monthly (20.7%). Finally, while over one-half (52.2%) of visiting relatives rarely participate in outdoor leisure activities (Q9), a considerable number engage in such activities several times a year (37.8%) or more than monthly (9.9%), suggesting a personal connection to the rural landscape.

4.3. Life Behaviors, Family Support, and Community Engagement in Rural Shimogo

This section details how elderly residents in Shimogo maintain their daily lives and social engagement, drawing upon individual strategies and support from out-migrated family members to navigate the challenges of the town’s unique geographic and demographic context. The data in Table 5 reveal varying degrees of self-reliance and reliance on support networks across various activities:

Table 5. Comparison of daily activities of resident households and support from hometown-visiting relatives.

Resident Households			Hometown-Visiting Relatives
Survey Questions on Questionnaire	Details and Proportion		Survey Questions on Questionnaire	Details and Proportion
10A. Daily Essentials Shopping N = 282	More than twice a week More than once a month More than once a year Rarely go out	142(50.4%) 107 (37.9%) 13 (4.6%) 20 (7.1%)	10B. Providing Daily Essentials Shopping N = 252	More than once a week More than once a month More than once a year Almost never	20 (7.9%) 42 (16.7%) 75 (29.8%) 115 (45.6%)
11A. Yard Work around house N = 289	More than once a month Several times a year Almost never	165 (57.1%) 90 (31.1%) 34 (11.8%)	11B. Providing Yard Work N = 263	More than once a month Several times a year Almost never	30 (11.4%) 48 (18.3%) 185 (70.3%)
12A. Farm Work N = 278	More than once a month Several times a year Almost never	194 (69.8%) 20 (7.2%) 64 (23.0%)	12B. Providing Farm Work N = 249	More than once a month Several times a year Almost never	35 (14.1%) 30 (12.0%) 184 (73.9%)
13A. Social Interactions and Leisure Activities N = 283	More than once a week More than once a month Several times a year Almost never	122(43.1%) 69 (24.2%) 55 (19.4%) 37 (13.1%)	13B. Social Interactions and Leisure Activities during hometown visits N = 253	More than once a week More than once a month Several times a year Almost never	10(4.0%) 19 (7.5%) 85 (33.6%) 139 (54.9%)
14A. Community Activities N = 290	Participate All Participate Part Almost never	228 (78.6%) 25 (8.6%) 37 (12.8%)	14B. Participating on Behalf of Residents with Community Activities N = 260	Participate All Participate Part Almost never	23 (8.8%) 38 (14.6%) 199 (76.5%)

The analysis of visiting relatives’ activities (questions 10B–14B) is based on respondents who reported that relatives visit at least once a year (Table 3, Q7); “Less than once a year” samples were excluded.

Essential activities: Regarding shopping for daily necessities (Q10A), 50.4% of residents shopped more than twice weekly, 37.9% more than once a month, and 11.7% less frequently. Support from visiting relatives (at least annually) varied considerably (Q10B), with 45.6% providing almost no assistance, 29.8% assisting more than once yearly, and 24.6% assisting more than monthly.

Productive activities: Similar patterns of self-reliance and family support were observed for agricultural activities (farmland cultivation, Q12A and B), with high resident engagement (nearly 80% at least several times yearly, and 69.8% more than once a month) and assistance from 30% of visiting relatives (14.1% more than once a month, and 12% several times a year). Yard maintenance (Q11A), reflecting both individual and community responsibility, also showed high resident participation (nearly 90% at least several times yearly; 57.1% more than monthly), with approximately 30% receiving assistance from visiting relatives. Given the national average age of 68.7 years for primary agricultural workers in 2020 (Ministry of Agriculture, Forestry and Fisheries, 2020), the high level of agricultural activity among older residents in Shimogo underscores the need for support strategies that consider the unique challenges and contributions of this demographic group [36].

Social engagement: Community engagement (Q14A) was high among residents (nearly 90% participating: 78.6% in all activities, and 8.6% in some), with visiting relatives also demonstrating involvement (8.8% in all activities; 14.6% in some). High levels of personal social engagement (Q13A) were also observed, with over 60% of residents participating regularly (43.1% more than weekly; 24.2% more than monthly).

4.4. Life Sustainability in Shimogo: Examining Daily Activity Performances Across Household Support Types

This section analyzes how three household support types—self-help, mutual aid, and combined support (categorized by respondents’ shopping and healthcare access methods, Q4, Table 3)—influence daily activity performance (shopping, yard work, farm work, social activities, community participation) in Shimogo. A multiple correspondence analysis (MCA, Table 6) examined household composition, health status, and activity participation to uncover underlying factors influencing self-reliance and support needs.

Table 6. Model summary for MCA of relatives’ activities.

Dimension	Cronbach’s Alpha	Variance Accounted For
		Total (Eigenvalue)	Inertia	Variance Explained (%)
1	0.746	2.480	0.496	59.1%
2	0.517	1.705	0.341	40.9%
Total		4.185	0.837	100.0%
Mean	0.653	2.093	0.419

Eigenvalues represent the amount of variance explained by each dimension. Inertia is proportional to the eigenvalue and represents the total variance explained by each dimension. Cronbach’s alpha is a measure of the internal consistency reliability; values above 0.6 indicate acceptable reliability. The mean Cronbach’s alpha is based on the mean eigenvalue [44,45].

The MCA yielded two dimensions shaping life sustainability in Shimogo: Dimension 1 (Community Engagement), reflecting participation in productive and social activities, and Dimension 2 (Community Support Needs), indicating reliance on external resources. Although Dimension 2’s Cronbach’s alpha is below the typical threshold, its acceptability is justified [46,47]. Notably, even highly self-sufficient self-help households (primarily multi-senior households) demonstrated a reliance on supplementary family support (40.7%, Figure 8), highlighting the pervasive influence of family networks across all support types. Their high community engagement and minimal external reliance (Figure 9) further underscore this point. Conversely, mutual-aid households (mostly single-senior) displayed lower community engagement and moderate external reliance, highlighting family support’s role in reducing formal service dependence. Combined-support households showed a balanced reliance on family and formal support.

Figure 8. Relatives’ daily essentials shopping assistance by support type. Relatives’ shopping assistance frequency (N = 243) by shopping support type (self-help, mutual aid, combined support; p < 0.001; Cramer’s V = 0.70) [44]. Data exclude households with almost no relative visits (Table 3, Q7).

Figure 9. MCA dimension scores by shopping and medical care support methods. This bar graph displays the mean scores (±1 standard error) for Dimension 1 (community engagement) and Dimension 2 (community support needs) for self-help, combined support, and mutual aid. Error bars represent the standard error of the mean. Higher scores for Dimension 1 indicate greater engagement in productive and social activities, while higher scores for Dimension 2 reflect greater reliance on external support for managing daily life.

Multinomial logistic regression (Table 7) revealed that higher community engagement significantly predicts self-help, indicating greater self-reliance. Conversely, higher community support needs strongly predict combined support, underscoring the importance of formal services when family support proves insufficient. Mutual-aid households showed no significant increase in community resource utilization, highlighting their primary reliance on family.

Table 7. Multinomial logistic regression predicting different household support types based on MCA dimensions.

Household Support Types	B	Std. Error	Wald	df	p	Exp(B)	95% CI for Exp(B)
Self-help (Reference) Combined support
Intercept	−0.924	0.126	53.697	1	<0.001
Dimension 1	−0.341	0.151	5.114	1	0.024	0.711	0.529, 0.956
Dimension 2	0.283	0.137	4.230	1	0.040	1.326	1.013, 1.736
Mutual aid
Intercept	−1.970	0.197	100.260	1	<0.001
Dimension 1	−0.818	0.166	24.348	1	<0.001	0.441	0.319, 0.611
Dimension 2	0.023	0.174	0.018	1	0.894	1.024	0.728, 1.438

Model fit: Cox and Snell R² = 0.089; Nagelkerke R² = 0.108; McFadden R² = 0.053; −2 Log Likelihood: 318.024; chi-square: 33.838; df = 4; p < 0.001. These metrics confirm the model’s overall significance while reflecting moderate explanatory power. Dimension 1 (community engagement) and Dimension 2 (community support needs) are based on Table 6.

4.5. Exploring the Motivations for Relatives’ Activities During Homecoming Visits

This section investigates factors driving relatives’ actions during their homecoming visits to Shimogo Town using a two-stage analysis. In the first stage, chi-square tests (Table 8) examined the frequency of visits, finding that age and location were significant predictors, but household composition and resident health status were not. Significantly, frequent visits occurred even among Kantō region relatives and those aged 65 or older, suggesting that strong motivations overcome distance and age barriers.

Table 8. Summary of chi-square tests: associations between frequency of hometown visits and various factors related to daily life sustainability in Shimogo Town.

Independent Variables	χ2	df	p	Cramer’s V	Effect Direction
2. Household Composition	-	-	>0.05	-	-
3. Health Condition	-	-	>0.05	-	-
4. Shopping Medical-care Method Type	44.079	4	<0.001	0.44	Mutual aid↑ Weekly visits Self-help↑ Annually visits
5. Location	63.795	6	<0.001	0.45	Aizu area↑ Weekly visits, Kanto region↓ Kanto region↑ Annual visits
6. Age	18.690	6	0.005	0.25	Under 50 ↑ Weekly visits, 20% over 65 50–64 ↑ Monthly visits Over 75 ↓ Annual visits
8. Chores Support	172.119	6	<0.001	0.76	Weekly visits↑ More frequent chore support Annual visits↑ Almost Never chore support
9. Outdoor Leisure	40.964	4	<0.001	0.37	Weekly visits and Monthly visits↑ Several times yearly; Annual visits↑ Almost Never
10B. Daily Essentials Shopping Support	135.516	6	<0.001	0.72	Weekly visits↑ More frequent shopping support Annual visits↑ Almost Never shopping support
11B. Yard Work Support	45.214	4	<0.001	0.38	Weekly visits↑ More than once monthly Annual visits↑ Almost Never
12B. Farm Work Support	42.243	4	<0.001	0.37	Weekly visits↑ More than once monthly Annual visits↑ Almost Never.
13B. Social Interactions and Leisure Activities	62.331	6	0.025	0.23	Self-help More than once a week↑ Mutual aid Several times a year and Almost Never↑.
14B. Community Activities Substitute participation	29.809	4	<0.001	0.32	Weekly visits ↑ Participate All and Part Monthly visits like Weekly visits Annual visits ↑ Almost Never.

Effect directions highlight the most prominent characteristic for each visit frequency. ↑ indicates a higher proportion; ↓ indicates a lower proportion. This analysis is based on respondents who reported at least one annual visit from relatives (Question 7, Table 5), Weekly = more than once a week, monthly = More than once a month, Annual = more than once a year, excluding those reporting “Less than once a year” visits. Cramer’s V effect sizes are categorized as: 0.0–0.2 (weak), 0.2–0.4 (moderate), 0.4–0.6 (strong), >0.6 (very strong) [44,45]. The numbered items in independent variables column correspond to those in Table 2.

The second stage employed MCA (Figure 10) to identify the underlying dimensions shaping visiting relatives’ activities across different household support types. This MCA, combined with multinomial logistic regression analysis (Table 9), revealed two primary dimensions significantly predicting visit frequency:

Figure 10. MCA of relatives’ activities in Shimogo. (a) Discrimination measures (contributions) for each variable for Dimensions 1 and 2. Age and geographic location are included as supplementary variables. (b) Joint category plot of variables. Dimension 1, representing family support for productive activities, is displayed on the horizontal axis; Dimension 2, representing the community engagement of residents, is displayed on the vertical axis. The proximities of the points reflect the strengths of their associations.

Table 9. Multinomial logistic regression predicting the hometown visit frequency based on MCA dimensions.

Hometown Visit Frequency	B	Std. Error	Wald	df	p	Exp(B)	95% CI for Exp(B)
More than once a month
Intercept	0.717	0.196	13.431	1
Dimension 1	0.402	0.145	7.683	1	0.006	1.495	1.124, 1.986
Dimension 2	0.700	0.180	15.056	1	<0.001	2.014	1.414, 2.867
More than once a year
Intercept	−0.188	0.305	0.381	1
Dimension 1	4.007	0.495	65.522	1	<0.001	55.001	20.843, 145.134
Dimension 2	0.944	0.204	21.401	1	<0.001	2.571	1.723, 3.835

Model fit: Cox and Snell R² = 0.764; Nagelkerke R² = 0.871; McFadden R² = 0.689; −2Log Likelihood: 338.687; chi-square: 193.676; df = 4; p < 0.005. This analysis is based on respondents who reported at least one annual visit from relatives (Question 7, Table 4).

Dimension 1 (Resident Self-Reliance) captures residents’ daily task independence. A significant negative correlation exists between higher self-reliance scores and visit frequency: as resident self-sufficiency increases, the need for frequent relative visits diminishes, as evidenced by self-help households and confirmed by multinomial logistic regression (Table 9).

Dimension 2 (Community Support Needs) measures reliance on external support. Higher community support needs unexpectedly correlate with fewer relative visits. This suggests that when relatives provide substantial support, the need for external community resources decreases. Conversely, when family support is lacking (as in households with high community support needs), community services become more crucial. This pattern aligns with mutual-aid and combined-support household dynamics, confirmed by the statistically significant negative correlation (Table 9).

4.6. Cluster Analysis of Relatives’ Activities: Emergence of Four Distinct Profiles

To further characterize the groups of relatives identified in the MCA, a k-means cluster analysis was performed on the MCA object sores of Dimensions 1 and 2. The four-group solution was chosen over solutions with two, three, or five clusters due to its superior fit (78% vs. 36% explained variance for the three-cluster solution; Table 10), even distribution of cases, and readily interpretable results. Figure 11 displays the mean scores (±standard deviations) for these two dimensions for each of the four clusters (N = 313). Table 10 and Figure 11 present descriptive statistics for each cluster, revealing distinct profiles in terms of geographic location, the visit frequency, support activities, and social engagement.

Table 10. Explained variance and case distribution in MCA dimension scores by number of clusters (N = 313).

Number of Clusters	Dimension		Distribution of Cases (%)
	1	2	Cluster 1	Cluster 2	Cluster 3	Cluster 4	Cluster 5
2	0.776	0.024	263 (84.0%)	50 (16.0%)
3	06.22	0.498	186 (60.0%)	46 (14.7%)	81 (25.3%)
4	0.780	0.776	107 (34.2%)	47 (15.0%)	46 (14.7%)	113 (36.1%)
5	0.844	0.776	32 (10.2%)	112 (35.8%)	42 (13.4%)	19 (6.1%)	108 (34.5%)

This table presents the explained variance in the MCA dimension scores for different numbers of clusters. The analysis is based on respondents reporting at least one annual visit (Table 4, Q7); those reporting less than one annual visit were excluded. Explained variance was calculated as 1 (within-group sum of squares/total sum of squares). Significant differences (p < 0.001 for both dimensions) between clusters were observed in all solutions.

Figure 11. Mean (± SD) scores for Dimension 1 (residents’ self-reliance) and Dimension 2 (residents’ community support needs) for four clusters (N = 313) identified via a k-means cluster analysis of the MCA object scores. One-way ANOVAs: p < 0.001 for both dimensions (Dimension 1: F (3, 309) = 361.071; Dimension 2: F (3, 309) = 414.875).

The four-cluster solution was selected as the optimal solution for the k-means cluster analysis based on a consideration of the explained variance, the evenness of the cluster size, and the interpretability of the resulting clusters (Table 10). The four-cluster solution showed a substantial increase in the explained variance for both Dimension 1 (relative support for resident needs; 78% vs. 64% for three clusters) and Dimension 2 (relative and resident community engagement; 78% vs. 50% for three clusters), indicating that it captured a significantly larger proportion of the variation in the data compared to the three-cluster solution. Furthermore, the four-cluster solution resulted in a relatively even distribution of cases across the clusters (34.2%, 15.0%, 14.7%, 36.1%), minimizing the potential bias associated with disproportionately sized clusters. While the five-cluster solution explained slightly more variance, this marginal improvement did not outweigh the less even case distribution and reduced interpretability of the resulting clusters. The two- and three-cluster solutions demonstrated considerably lower explained variance and less even distributions. Thus, the four-cluster solution provided the best balance between maximizing the explained variance, ensuring even cluster sizes, and maintaining the interpretability of the distinct clusters.

The four clusters represent distinct profiles of relative engagement (Table 11, Figure 11):

Table 11. Descriptive statistics of characteristics of visiting-relative clusters in Shimogo.

	Percentages 1 in Clusters				Characteristic
	Cluster 1	Cluster 2	Cluster 3	Cluster 4
Relatives over 50 years	37 (38.6%)	30 (65.2%)	24 (54.4%)	52 (47.7%)	Cluster 1: younger; Cluster 2: older
Relatives from Aizu area	49 (47.1%)	33 (71.7%)	24 (55.8%)	56 (50.0%)	Cluster 2: predominantly Aizu-based
Relatives from Kanto region	37 (35.6%)	4 (8.7%)	6 (14.0%)	36 (32.1%)	Clusters 1 and 4: significant Kanto representation
High-frequency hometown visits (>monthly)	44 (41.1%)	47 (100%)	24 (52.2%)	52 (46.0%)	Cluster 2: high visit frequency
Annual-frequency hometown visits (yearly only)	63 (58.9%)	0 (100%)	22 (47.8%)	61 (54.0%)	Cluster 2: no annual visits; Clusters 1 and 4: high annual visits
High-frequency chore support (>monthly)	8 (8.1%)	30 (75.0%)	8(20.0%)	18 (18.9%)	Cluster 2: high chore support; Clusters 1, 3, and 4: low chore support
Outdoor leisure activities 2	10 (10.1%)	39 (90.7%)	43 (97.7%)	57 (60.6%)	Cluster 2: high frequency; Cluster 3: annual frequency
High-frequency shopping support (>monthly)	7 (7.4%)	33 (90.7%)	3 (11.1%)	19 (21.3%)	Cluster 2: high shopping support; Clusters 3 and 4: low shopping support
Annual-frequency shopping support (yearly only)	9 (9.5%)	1 (2.4%)	21 (77.8%)	44 (49.4%)	Cluster 3: high annual shopping support
Yard work support 2	4 (4.0%)	35 (85.4%)	23 (74.2%)	16 (17.4%)	Cluster 2: high yard work support; Cluster 3: yearly support
Farm work support 2	1 (1.0%)	28 (77.8%)	23 (79.3%)	13 (14.8%)	Cluster 2: high yard work support; Cluster 3: yearly support
High-frequency social activities (>monthly)	6 (6.3%)	20 (54.1%)	2 (6.9%)	1 (1.1%)	Cluster 2: high social activity
Social activities 2	19 (19.8%)	28 (75.7%)	22 (75.9%)	45 (49.5%)	Clusters 2 and 3: high participation rate; Clusters 1 and 4: low participation rate
Community activities 2	9 (9.2%)	27 (67.5%)	9 (32.1%)	16 (17.0%)	Cluster 2: high participation rate; Cluster 3: low participation rate
Shopping and medical care support needed by residents 3	20 (19.8%)	32 (68.1%)	17 (37.8%)	44 (41.1%)	Cluster 2 families: high support needed; Cluster 4 families: high unmet support needed

¹ Percentages are within-cluster proportions based on valid responses; sample sizes vary across variables. ² Outdoor leisure activities, yard work support, farm work support, social activities, and community activities indicate participation, not frequency. ³ Support needed by residents refers to the proportion of households within each cluster requiring external support for shopping or healthcare (Table 3, Q4; the numbers of mutual-aid and combined-support households).

Cluster 1: obligation-driven visits (N = 107): This cluster (high scores for Dimension 1, low scores for Dimension 2) is characterized by predominantly young, Kantō-based relatives who make primarily annual visits, with minimal involvement in support or community activities. This suggests that their visits may primarily be driven by a sense of obligation, maintaining a tenuous connection with Shimogo despite limited active engagement in either support or community life. The resident households in this cluster exhibit high self-reliance, reducing the need for frequent relative support, as evidenced by their high scores for Dimension 1;

Cluster 2: high support and engagement (N = 47): This cluster (low scores for both dimensions) is characterized by predominantly older, Aizu-based relatives with high visit frequencies, extensive involvement in various support activities (chores, shopping, yard work, farm work), and a high degree of community engagement. These relatives provide significant ongoing support, effectively meeting the substantial needs of their resident households (low scores for Dimension 1). Their high community engagement (low scores for Dimension 2) further indicates that family support may largely substitute for the need for community-based support;

Cluster 3: community-maintenance-focused visits (N = 46): This cluster (low scores for Dimension 1 and high scores for Dimension 2) is characterized by primarily Aizu-based relatives with moderate, primarily annual visit frequencies and significant involvement in support activities, especially farm work. These households show high needs for support (low scores for Dimension 1), but these needs are largely met through community resources (high scores for Dimension 2). Relatives may provide support primarily related to productive activities (farm work), which are not readily accessible through community services. This group is characterized by moderate participation in both support and community activities, with the potential for increased engagement upon retirement;

Cluster 4: leisure-oriented visits (N = 113): This cluster (high scores for both dimensions) is characterized by a mix of Aizu- and Kantō-based relatives with moderate visit frequencies, prioritizing social and recreational activities within the town of Shimogo, with relatively little direct support provided for elderly residents. These households demonstrate high self-reliance (high scores for Dimension 1) and a low need for community-based support (high scores for Dimension 2).

These distinct clusters highlight the diverse motivations and engagement levels among out-migrated family members that support elderly residents in Shimogo.

5. Discussion

The challenge of ensuring life sustainability in aging rural communities is increasingly pressing worldwide. This study contributes to this critical issue via its examination of the interplay between the household structures, support networks, and daily activity performances among elderly residents in Shimogo, Japan. Our findings go beyond simple models of family support to reveal the complex interplay between self-reliance, family-based support (both from co-residents and geographically dispersed relatives), and community resources. We identified significant variations across the household support types in community engagement and the reliance on external support, highlighting both the resilience and vulnerabilities of the different household structures. Our investigation into the motivations of out-migrated family members further sheds light on the multifaceted nature of family support in this context.

5.1. Resilience and Vulnerability in Different Household Structures: Geographically Dispersed Family Support in Depopulating Rural Areas

In this section, we explore how the different household structures in the depopulating town of Shimogo cope with aging, highlighting the crucial role of geographically dispersed family support alongside self-reliance and community resources. Rapid aging and depopulation [48,49], which have resulted in few intergenerational households (6.5% in our sample, Table 3), challenge the assumption that co-resident family members are the primary source of eldercare [50], necessitating an examination of alternative support mechanisms, especially for the highly vulnerable single-senior households (39.4%, Table 3).

Our findings reveal contrasting patterns of resilience and vulnerability (Section 4.1 Table 3). The resilience of elderly residents is strikingly demonstrated by the high proportion (54.2%) of multi-senior household residents who successfully manage their daily lives independently, effectively leveraging peer support networks to maintain high levels of functioning. This unexpected finding challenges the assumption that intergenerational co-residence is essential for successful aging and underscores the adaptability of older adults in navigating the challenges of limited resources and changing family structures. This self-reliance is further evidenced by the high proportion (66.4%) of household residents who manage their healthcare and daily needs without formal support (Section 4.1), demonstrating their capacity to overcome potential health challenges and limited access to formal support systems. Their active participation in social and community activities (Section 4.3) highlights the maintenance of strong social networks and a continued sense of purpose and belonging, despite shrinking communities and out-migration.

However, this resilience contrasts sharply with the vulnerabilities faced by single-senior households, particularly those headed by individuals aged 80 years and older. These households demonstrate a significantly greater reliance on external support (Table 3), struggling to access essential services, a challenge that is exacerbated by complex community care systems [51].

This context underscores the often-overlooked contribution of geographically dispersed family members. Despite significant out-migration, these relatives maintain strong ties with Shimogo (Section 4.2), providing crucial support through frequent visits (almost 90% annually; many monthly or weekly). This sustained engagement, driven by factors beyond practical assistance (cultural obligations, life events, personal connections to the landscape—Section 4.2), offers vital emotional and practical support. While direct chore assistance is less frequent, this consistent contact is a critical resource, especially given the decreasing likelihood of co-residing with children in today’s rural context, necessitating a more comprehensive understanding of support networks that extend beyond the traditional focus on co-resident family members.

5.2. The Interplay of Self-Reliance, Family Support, and Community Resources: Optimizing Support for Successful Aging in Place in a Depopulating Rural Context

Section 4.4 reveals the complex interplay between the household support type and daily activity performance in Shimogo, highlighting the varying contributions of self-reliance, family support, and community resources in shaping both community engagement (Dimension 1) and community support needs (Dimension 2) (Figure 9). In this section, we explore these interrelationships, considering the limitations of relying solely on family- or community-based informal care within Shimogo’s aging and depopulating context.

The significant self-reliance observed, particularly among the residents of multi-senior households, underscores the importance of individual agency and the quality of community life [52]. The high proportion (54.2%) of multi-senior household residents managing independently demonstrates resilience, challenging assumptions about the necessity of intergenerational co-residence for successful aging. However, this self-reliance is often complemented by family support, acting as a flexible safety net; even self-help households frequently receive some relative assistance (Figure 8).

Family support, which is especially crucial for mutual-aid households, significantly mitigates external support needs (Dimension 2) through frequent shopping and chore support (Section 4.3). These findings, consistent with broader research on aging adults in communities [53,54,55], underscore the importance of informal care, especially family support from spouses or children [56], in supporting older adults. However, this reliance on family support may limit community engagement (Dimension 1), as our findings suggest a potential trade-off between these two aspects of well-being. While family support is essential, particularly for mutual-aid households, the strong association between the high scores for Dimension 2 (community support needs) and less frequent visits (Section 4.5) indicates that high levels of family support may not necessarily translate into high levels of community engagement (Figure 9). Indeed, some relatives’ visits appear to be primarily leisure-oriented (Section 4.6, Cluster 4), suggesting that their activities may be driven more by personal enjoyment than by addressing the practical support needs of elderly residents. In these cases, the residents appear to rely more on community resources to address their daily needs (high scores for Dimension 2 and low scores for Dimension 1 in Section 4.4). This suggests a potential trade-off: high levels of family support may be a substitute for community engagement, potentially hindering the development of broader social networks. Further research is needed to more comprehensively explore this potential trade-off between family support and community participation among elderly individuals.

Community resources are particularly critical for combined-support households, reflected in their higher scores for Dimension 2 (Section 4.4). While family and self-reliance are crucial, accessible community services—assistance with daily tasks and social engagement opportunities [56]—are vital for successful aging in place and enhanced quality of life [52]. This integrated approach leverages both formal and informal support, recognizing the value of strong community ties and social participation in rural contexts [56].

In conclusion, successful aging in Shimogo hinges on the complex interplay between self-reliance, family support, and community resources. Policies must support this interplay, acknowledging the limitations of informal care and the need for accessible community services [54], particularly given the aging caregiver population and out-migration trends. In future studies, researchers should investigate how to best integrate formal and informal support systems to maximize effectiveness and sustainability, paying particular attention to the potential trade-offs between different support mechanisms.

5.3. Motivations Behind Relatives’ Homecoming Visits: Diverse Forms of Family Support Under the Influence of Filial Piety in Rural Japan

This section analyzes the diverse motivations behind out-migrated family members’ visits to Shimogo, focusing on practical support, emotional connection, community engagement, and the enduring influence of filial piety (oyakokō) within Japan’s rural revitalization efforts. While Confucian traditions underpin filial piety [57], its expression varies across cultures. Financial support is often prioritized in China and Korea [58,59], contrasting with the greater financial independence of the elderly in Western societies like Europe and North America [60], though familial obligations persist in the United States [61]. Although some argue that traditional filial piety is less significant in contemporary Japan [62], this study in Shimogo explores how contemporary expressions of filial piety manifest in a rapidly changing, depopulating rural context. Our four relative engagement clusters (Section 4.6) reveal diverse expressions of filial piety, extending beyond the traditional emphasis on financial support and co-residence [50] to encompass practical assistance, emotional connection, and community engagement.

Clusters 2 (High Support and Engagement) and 3 (Community-Maintenance-Focused Visits) illustrate how filial piety contributes directly to the sustainability of the community. Cluster 2 relatives provide extensive practical support crucial for single-senior households, reducing the need for costly external care and supporting the economic health of the community. Cluster 3 relatives, through their involvement in farm work, directly contribute to both the economic and environmental sustainability of the community [63], maintaining local food production and potentially sustainable agricultural practices. These intensive forms of ongoing practical care thus support both individual well-being and the broader sustainability of the Shimogo community.

Clusters 1 (Obligation-Driven Visits) and 4 (Leisure-Oriented Visits) represent distinct yet equally valid expressions of filial piety, both characterized by regular visits. While Cluster 1 emphasizes maintaining family connections through annual visits, often with limited practical support [64], Cluster 4 prioritizes strengthening emotional bonds and combating isolation through shared leisure activities [65]. Both cluster types, particularly given the younger age of relatives and higher resident self-reliance, could potentially transition towards the more practically focused support patterns of Clusters 2 and 3 over time, with future research needed to explore this possibility.

The cultural practice of ancestral grave maintenance [59] further underscores the enduring influence of filial piety, even as its expression evolves [66,67,68]. This deep-rooted cultural expectation shapes family interactions and significantly contributes to the well-being of elderly residents. The interplay of practical and emotional support, combined with the contribution to community sustainability, demonstrate a complex, multifaceted expression of filial piety [66]. Ensuring Shimogo’s long-term sustainability requires leveraging both the strength of these family ties and robust community services [69].

5.4. Policy Recommendations and Future Research Directions: Ensuring Life Sustainability in Aging Rural Communities

Shimogo’s complex interplay between self-reliance, family support, and community resources necessitates a multifaceted eldercare approach. To optimize successful aging in place and vibrant rural communities, we recommend:

Strengthening Family Support: Support diverse family support models (intensive to culturally driven visits) through flexible programs offering caregiver respite, financial aid, and initiatives fostering regular contact between elderly residents and dispersed relatives. Address the aging caregiver population and youth out-migration.

Enhancing Community Services: Invest in accessible community services (affordable transport, home maintenance, age-friendly healthcare) and initiatives combating isolation (community centers, social clubs, volunteerism) to enhance quality of life and belonging [56].

Leveraging U-Turn Migration: Encourage U-turn migration from those with strong community ties, particularly those providing significant support, through housing subsidies, job creation, and integration programs. Foster intergenerational connections to counter future declines in family support.

Future research should assess the long-term sustainability of diverse support models and the impact of policy changes on support networks, incorporating qualitative research on the experiences of both elderly residents and their families. A comprehensive strategy leveraging both informal and formal support is crucial for thriving rural communities.

5.5. Strengths and Limitations

This study offers valuable insights into the diverse nature of family support networks in rural Japan. However, the findings should be interpreted within the context of certain limitations. This study relies on residents’ descriptions of their relatives’ visit frequencies and activities, given the difficulty of directly contacting out-migrated family members. This approach effectively mitigates potential bias from subjective evaluations, revealing residents’ real-life experiences and complex strategies for adapting to mountainous challenges. Moreover, this approach illuminates the motivations behind relatives’ return visits, highlighting the intricate interplay between residents’ lives, community resources, and family support.

However, Shimogo’s abundance of natural resources and tourism infrastructure, combined with the strong cultural connections within the Aizu region, may contribute to a higher frequency of visits by relatives, potentially influencing the findings. Furthermore, the relatively small sample size (378 samples, 41.4% response rate) limits the generalizability of the findings to the wider population of elderly residents in other areas.

Future research could benefit from exploring the perspectives of out-migrated family members, expanding the sample size to encompass a wider range of rural settings, and considering the influence of specific regional characteristics on family support patterns.