Mutual aid among older residents in two socially different districts of a rural municipality in Japan: a comparative qualitative study

Abstract

Background

Mutual aid among older residents is increasingly important in ageing communities in Japan. However, little is known about how mutual aid is actually formed and reconfigured across socially different districts within the same municipality.

Methods

This comparative qualitative study analysed semi-structured interviews with 11 community-dwelling older residents aged 77–90 years in one rural Japanese municipality (fishing district, n = 6; in-migrant “I-turn” district, n = 5). Seven interview sessions were conducted across the two districts, totalling 597 min. The primary analysis used Kuckartz-based qualitative text analysis. The Trajectory Equifinality Approach / Trajectory Equifinality Modeling (TEA/TEM) was then used as a secondary interpretive framework to reinterpret how mutual aid changed over irreversible time.

Results

Four synthesized comparative themes were identified. First, in the fishing district, mutual aid was supported by long-term co-residence, neighbours’ routine awareness of doors, curtains, and other daily signs, neighbourly access to homes when needed, and an expectation that help would be reciprocated over time. Second, in the I-turn district, mutual aid was more often built intentionally through participation in local activities, transport arrangements, and selectively cultivated friendships after migration. Third, across both districts, ageing and health events shifted residents between providing help and receiving help. Fourth, mutual aid had clear relational boundaries: not all residents were equally involved in neighbourhood activities or ties, and helping coexisted with restraint, privacy/non-intrusion, and family-first norms. Mutual aid was experienced not as a fixed role but as a shifting position between providing and receiving help over time.

Conclusions

Mutual aid among older residents was neither uniform across districts nor fixed over time. Rather, it was shaped by district-specific relational infrastructures and reconfigured through age-related changes in health, mobility, and everyday routines. Community interventions should therefore be tailored to the relational infrastructure already present in each district and should support modest, repeatable opportunities for reciprocal contact and gentle monitoring.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12877-026-07792-1.

Background

Japan is one of the most rapidly ageing societies in the world, and the number of older adults living alone continues to increase [1, 2]. Within Japan’s community-based integrated care policy framework, “mutual aid” refers to voluntary, informal support among residents that is not institutionally guaranteed [3]. Previous concept analyses in Japan have treated mutual aid more broadly as residents’ interaction to address daily life needs, together with empathic awareness of those needs and a spontaneous intention to support one another [4]. In the present study, however, mutual aid was operationally defined more narrowly as resident-to-resident informal support, excluding family-based support, formal care, and commercial services.

Previous Japanese research has shown that social relationships, social support, neighbourhood cohesion, and social participation are closely associated with mortality, depressive symptoms, life satisfaction, disability, and well-being among older adults [5–15]. These studies collectively suggest that local relational environments matter for ageing in place. At the same time, most of this evidence is quantitative, and the term mutual aid is often treated as a broad attitude, a correlate of social capital, or a proxy for support availability rather than as an evolving lived practice between residents [7–9, 11–15].

Relevant qualitative evidence remains limited. Tsuji and Khan described how support and life satisfaction were intertwined among rural older adults in Japan [10], while Kuroda and colleagues showed that island-based support for older adults depended on dense local ties and flexible information sharing [16]. However, few studies have compared how older residents themselves experience and enact mutual aid in districts with clearly different social histories within the same municipality. Such a comparison is important because it allows broader municipal conditions to be held relatively constant while examining variation in relational infrastructure, everyday interaction, and norms of helping.

In the present study, we focused on two districts within one rural municipality in Japan: a traditional fishing district and an in-migrant residential district locally described as an “I-turn” area, a Japanese term commonly used for migration from an urban area to a non-native rural locality. The aim was to clarify how older residents in these two socially different districts experienced, practised, and reconfigured mutual aid over time.

Although this study is grounded in the Japanese policy and social context, the contrast between a long-established coastal community and an ageing in-migrant residential district may also inform other regions outside Japan where population ageing, household shrinkage, and selective neighbourhood ties are advancing.

Across this paper, “relational infrastructure” refers to the district-specific social and spatial conditions that make noticing, asking, accepting, and reciprocating help more or less possible in everyday life. We use “gentle monitoring” to refer to low-intensity checking, such as noticing routine signs or making a brief call, that remains socially acceptable because it stops short of intrusive surveillance. “Mutual aid boundaries” refers to the relational limits of what neighbours feel able, willing, or entitled to do for one another.

Aim

This study aimed to explore how engagement in mutual aid among older residents was formed, practised, and reconfigured over time in two socially different districts within the same rural municipality in Japan.

Methods

Study design

This was a comparative qualitative study of community-dwelling older residents in two districts within one rural municipality. The manuscript was prepared with reference to the Consolidated Criteria for Reporting Qualitative Research (COREQ) [17] and broader qualitative reporting guidance [18]. A completed COREQ checklist is provided in Additional file 6.

Study setting

The study was conducted in Onjuku Town, a rural coastal municipality in Chiba Prefecture, Japan. Municipal planning documents during the study period described the town as a compact territory of 24.86 km² with a warm climate, proximity to the sea and satoyama landscapes, and a rapidly ageing population [19]. Town materials also noted continuing out-migration of younger generations but in-migration around retirement age, making later-life relocation a visible demographic feature within the same municipality [19].

Two districts within the municipality were selected because they differed in social history, housing form, and migration pattern while sharing the same administrative context. The fishing district (Iwawada) was part of the long-established coastal settlement associated with fishing livelihoods, closely spaced homes, and dense everyday visibility. Local newsletters and historical materials in this area also commemorate collective helping narratives, suggesting that helping is culturally legible within local history [20]. The I-turn district (Onjukudai) was a planned residential development of about 93 ha located roughly 1.4 km west of Onjuku Station. Planning documents described it as a quiet detached-housing area with roads, parks, and a regulated residential landscape intended to preserve a good living environment [21]. Town materials also indicated that this district had become a major destination for later-life in-migration and had a particularly high ageing rate [19, 21]. In this study, these contextual differences were treated not as stereotypes but as conditions likely to shape how mutual aid became visible, acceptable, and actionable in later life.

Participants

Purposive sampling was used to recruit older residents who were able to reflect on everyday life, neighbourhood relations, and experiences of helping and being helped. Potential participants were identified in consultation with municipal public health nurses and local gatekeepers after discussion of district selection and feasibility with municipal staff. Welfare commissioners were excluded from the present paper and reserved for separate reporting. The final sample comprised 11 residents aged 77–90 years: six from the fishing district and five from the I-turn district. No approached participant refused participation after receiving the study explanation, and no enrolled participant withdrew. For the comparative aims of this study, the sample of 11 residents and seven interview sessions was considered to provide sufficient information power because the study question was relatively focused, the interviews were information-rich, and the dataset yielded both convergence and contrast across the two districts rather than seeking population prevalence [22].

Data collection

Semi-structured interviews focused on everyday life, neighbourhood relations, experiences of practical and emotional mutual aid, and changes in mutual aid with ageing. In the fishing district, first-round group interviews were conducted with residents A–C (80 min) and G–I (84 min), followed by a second-round group interview with four residents (A, B, C, and G; 105 min). In the I-turn district, the first round consisted of one group interview with residents J–L (107 min) and two individual interviews (P: 60 min; Q: 75 min), followed by a second-round group interview with J–L (86 min). In total, seven interview sessions were conducted, totalling 597 min (fishing district: 269 min; I-turn district: 328 min). No non-participants were present during the resident interviews reported in this paper.

A semi-structured interview guide was developed specifically for this study and used in interviews with older residents (Additional file 7). The guide began with questions about participants’ age group and residential background (newcomer to the area or local resident). It then explored participants’ current health status and any medical treatment or diagnoses, previous experiences of sudden emergencies or acute changes in condition (e.g., loss of consciousness, falls, or serious injuries), experiences of giving and receiving help within the community, and views on mutual aid among older residents when someone experiences an emergency or other difficulties. All interviews were audio-recorded, transcribed verbatim in Japanese, anonymized, and managed manually using Excel-based worksheets rather than dedicated qualitative software. Handwritten field notes and analytic memos were kept during and after data collection. Full verbatim transcripts were not returned to participants; however, draft TEM diagrams and summaries were returned twice for factual checking and minor correction. Representative quotations in the present paper were translated into English by the authors for manuscript preparation.

All resident interviews were conducted by Yoshiko Ohno. During the individual interviews with participants P and Q, Masako Kaneko and Yukari Ishida were also present as attending researchers. Interviews were conducted in the two study districts, either in participants’ homes or in nearby shared community facilities according to participant preference. Before each interview, the researchers introduced themselves and explained the study purpose, procedures, voluntary participation, confidentiality, and recording; written informed consent was obtained.

Research team and reflexivity

At the time of manuscript preparation, the seven-author team comprised one Professor, two Assistant Professors, one Lecturer, and three municipal public health nurses in the Health and Welfare Division of Onjuku Town Hall. All authors were female licensed public health nurses; one held a PhD, three held master’s degrees, and three held bachelor’s degrees in nursing.

Yoshiko Ohno, who conducted all resident interviews, and Masako Kaneko, who attended two interviews as a non-author researcher, each had more than 10 years of qualitative research experience and formal training in qualitative methods. Three municipal public health nurse collaborators had more than 3 years of practice experience in the study area, and some participants were already known to them through routine community health work before study commencement. This positioning facilitated contextual understanding of local health practices, community assessment, and interpretation of everyday mutual aid, but it also required ongoing reflexive discussion about prior assumptions, professional role duality, and the risk of over-reading familiar local narratives. Interpretations were therefore checked against verbatim transcripts, code tables, and case trajectories rather than accepted solely on the basis of field familiarity.

Primary qualitative analysis

In this study, mutual aid was defined as resident-to-resident informal support, excluding family-based support, formal care, and commercial services. The transcripts were analysed primarily using Kuckartz-based qualitative text analysis [23]. First, the transcripts were read repeatedly to gain an overall understanding of each case. Second, meaning units relevant to mutual aid, neighbourhood interaction, and changes over time were identified. Third, initial codes were generated and grouped into subcategories through within-case and cross-case comparison. Fourth, coding decisions were compared across cases and districts by Yoshiko Ohno, Eriko Tane, Masako Kaneko, Yukari Ishida, and Keiko Maeda, and the codebook was refined through discussion. Fifth, the analysis was compressed into cross-district thematic patterns that could explain both commonalities and contrasts between the two districts. Before finalizing the present manuscript, the analytic claims retained in the paper were rechecked by the remaining authors against the code tables, TEM diagrams, and cross-case summaries, and the remaining authors take responsibility for the reported interpretations.

Use of the Trajectory Equifinality Approach / Trajectory Equifinality Modeling (TEA/TEM) as a temporal interpretive framework

The Trajectory Equifinality Approach / Trajectory Equifinality Modeling (TEA/TEM) was then used as a secondary interpretive framework to reinterpret these findings in relation to change over irreversible time, without guiding code generation or theme construction [24–27]. More specifically, the Kuckartz-based coding and thematic analysis produced the primary categories, subcategories, and four synthesized comparative themes, whereas TEA/TEM was applied afterwards to revisit the same material in temporal terms and to model branching and converging pathways across later life. During a series of team discussions, the Kuckartz-based categories and the TEA/TEM-informed temporal interpretations were compared and refined; however, TEA/TEM did not function as a separate co-primary coding method. Consistent with methodological discussions in TEA/TEM, this step was treated as a theoretically informed modelling process rather than a rigid standardized procedure [24–27].

In this study, TEA/TEM concepts were operationalized as follows: irreversible time referred to the later-life course through which mutual aid relationships changed; bifurcation points (BFPs) referred to events that redirected mutual-aid trajectories, such as surgery, falls, widowhood, or mobility decline; equifinality points (EFPs) referred to concrete acts of resident-to-resident mutual aid in everyday life; polarized EFPs (P-EFPs) referred to occasions in which mutual aid was not realized, was refused, or was effectively replaced by external services; obligatory passage points referred to conditions or encounters that made mutual aid possible; social guidance referred to forces that facilitated mutual aid; and social direction referred to forces that constrained or discouraged it.

These concepts were applied after coding to reread the material at the case and cross-case levels so that static categories could be reconsidered as temporal pathways. Detailed decision rules used to determine whether a segment qualified as a resident-to-resident mutual-aid EFP are provided in Additional file 1: Figure S1. A full category-subcategory-code table is presented in Additional file 2: Table S1, while Table 3 shows the four synthesized comparative themes. Two district-specific cross-case TEM diagrams are provided in Additional file 3: Figure S2 and Additional file 4: Figure S3. Study-specific operational definitions of the TEA/TEM concepts are summarized in Additional file 5: Table S2.

Table 3.

Analytic overview of the four synthesized comparative themes

Synthesized theme	Fishing district	I-turn district	Temporal / TEA-TEM cue
1. Different relational bases of mutual aid	Long-term co-residence, shared histories, routine awareness of doors, curtains, and other daily signs, and neighbourly access to homes made mutual-aid acts recognizable and legitimate.	Ties usually had to be intentionally built through participation in local activities, transport arrangements, and selectively cultivated friendships after migration.	District context as enabling/constraining relational infrastructure (visibility, housing form, migration history, participation opportunities; SG/SD).
2. Mutual aid was practical and everyday	Mutual aid centered on concrete acts such as meal sharing, grass cutting, rides, dog eye-drops, and noticing closed doors.	Mutual aid centered on practical acts such as food sharing or delivery, rides, information exchange, phone check-ins, and volunteer-linked garbage help.	Concrete everyday providing-help / receiving-help acts as observable EFPs.
3. Ageing and health events reconfigured roles	Surgery, falls, widowhood, and declining mobility shifted residents from providing help to receiving help.	Accidents, bereavement, declining driving ability, and chronic illness similarly moved residents toward receiving help.	BFP → later-life EFP; provider ↔ recipient reconfiguration in irreversible time.
4. Mutual aid had clear relational boundaries	Family availability, reduced communal participation, generational change, and limits on entering others’ lives restricted what neighbours would do.	Reluctance to burden others, privacy / non-intrusion, uneven involvement in local activities, and preference for family, physicians, volunteers, or formal services shaped the boundary of mutual aid.	P-EFP / SD; restrained help-seeking, family or service substitution, and selective non-intervention.

Trustworthiness

Several strategies were used to strengthen trustworthiness [28, 29]. First, the use of two interview rounds in part of the sample enabled temporal clarification and comparison. Second, coding and category judgments were repeatedly discussed in research meetings. Third, comparing two districts within the same municipality allowed analytical contrast without changing the broader administrative context. Fourth, participants checked draft TEM diagrams and summaries on multiple occasions, and factual corrections were incorporated. Fifth, the team deliberately retained boundary cases and contrasting narratives rather than looking only for consensus or positive examples of mutual aid. Data saturation was not used as a formal stopping criterion. Instead, analytic sufficiency was judged in terms of information power, interview depth, analytic focus, and whether the two-district dataset had generated clear convergence and contrast relevant to the comparative aim.

Supplementary AI-assisted analytic audit

After completion of the human-led analysis, ChatGPT (OpenAI; GPT-5.4 Pro) was used only as a supplementary analytic audit tool to flag possible inconsistencies in code application and potentially ambiguous category boundaries for re-checking by the research team. It was not used to generate data, codes, themes, analytic categories, or interpretations. All outputs were critically checked against the original transcripts, coding records, and team discussions. All final analytic decisions and manuscript wording remained the responsibility of the human authors. No other AI tools were used.

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and relevant guidelines and regulations. Ethical approval was originally obtained from the Dokkyo Medical University Nursing Research Ethics Committee on 5 November 2021 and was subsequently approved by the Research Ethics Review Committee of Gunma University of Health and Welfare (approval no. RS23-21; approved 21 March 2024). All participants received oral and written explanations of the study, and written informed consent was obtained prior to participation.

Results

The analysis was organized into four themes that captured district-specific relational infrastructures as well as shared processes of reconfiguration across later life. Tables 1 and 2 summarize sparticipant characteristics and interview participation. Table 3 summarizes the four synthesized comparative themes derived from the primary Kuckartz-based analysis. The district-specific cross-case TEM diagrams in Additional file 3: Figure S2 and Additional file 4: Figure S3 then show how these empirically grounded findings were reinterpreted through TEA/TEM in terms of temporal branching, diversity, and movement between providing help, receiving help, and refraining from help-seeking. Finally, Fig. 1 integrates these district-specific pathways into a single cross-case model linking district-specific relational infrastructures, everyday mutual aid before bifurcation points, later-life reconfiguration triggers, and later-life EFP/P-EFP pathways. This staged presentation is intended to make the relationship between the primary thematic analysis and the later temporal reinterpretation transparent.

Table 1.

Characteristics of the 11 resident participants and interview participation. Panel A. Fishing district residents (n = 6)

ID	Age / sex	Life-course and residential background	Current context relevant to mutual aid	Interview participation
A	83 / F	Born and raised in the district; worked in Tokyo, then returned through marriage.	Caring for husband with dementia; own cardiac surgery and knee/back pain; attends health class; described practical help to neighbors (e.g., grass cutting, mobility aid, shopping support after discharge).	GI-1 (A/B/C, 80 min); GI-2 (A/B/C/G, 105 min)
B	85 / F	Married into the district about 60 years ago; former barber; now lives alone after widowhood.	Long-standing reciprocal neighborhood ties; gave nightly meal support to an older neighbor for 13 years; also described grass cutting and food sharing.	GI-1 (A/B/C, 80 min); GI-2 (A/B/C/G, 105 min)
C	77 / F	Married into the district from another city in the prefecture; worked until age 71.	Active in the health class; encourages others to join; monitors neighbors through doors/household signs; has family nearby and therefore does not usually seek neighbor help herself.	GI-1 (A/B/C, 80 min); GI-2 (A/B/C/G, 105 min)
G	82 / F	Long-term resident; lives alone.	Mobility limitation after femur surgery; daughter assists transport; neighbors helped quickly after a bicycle fall; often keeps the door unlocked and receives small gifts/visits.	GI-1 (G/H/I, 84 min); GI-2 (A/B/C/G, 105 min)
H	85 / M	Born and raised in the district; widowed former fisher; lives alone.	Son assists with transport; perceives weak greeting/help from neighbors; limited participation in clubs and communal activities.	GI-1 (G/H/I, 84 min)
I	87 / F	Born and raised in the district; widowed; lives alone.	Daughter brings daily supplies; distrusts current neighborhood help; relies on phone/ALSOK and recalls older mutual-aid customs as having diminished.	GI-1 (G/H/I, 84 min)

GI group interview. Only participant G joined the second group interview among the G/H/I first-interview set

Table 2.

Characteristics of the 11 resident participants and interview participation. Panel B. I-turn district residents (n = 5)

ID	Age / sex	Life-course and residential background	Current context relevant to mutual aid	Interview participation
J	89 / F	Remarried and moved to the district at age 65; husband died 4 years ago.	Active in walking/exercise groups; says help is available if she asks; partly relies on physician and ALSOK rather than intimate neighborhood ties.	GI-1 (J/K/L, 107 min); GI-2 (J/K/L, 86 min)
K	79 / F	Migrated from outside the prefecture in her 50s.	Active participant in local activities; responded to three neighbor emergencies, including hospital accompaniment and end-of-life presence; notes difficulty when people avoid contact.	GI-1 (J/K/L, 107 min); GI-2 (J/K/L, 86 min)
L	77 / F	Migrated from outside the prefecture in 1993; lives alone.	Nearby friends are more reliable than distant family; wants to reciprocate help received; uses self-arranged alert and boundary-setting systems.	GI-1 (J/K/L, 107 min); GI-2 (J/K/L, 86 min)
P	85 / F	From another area within the same town; has lived in the district for 33 years; lives alone after widowhood.	Vision and mobility decline; uses garbage-help volunteers, mobile shops, and transport support; says close neighborhood mutual aid is unlikely because ties are thin.	Individual interview (60 min)
Q	90 / F	Migrated from outside the prefecture; widow.	Long-term participation in walking, dance, and weaving groups; actively cultivates ties; uses garbage-help volunteers; values heart-to-heart ties and younger neighbors’ support.	Individual interview (75 min)

J/K/L participated in both first- and second-round group interviews; P and Q participated in individual first interviews only

Fig. 1.

Cross-case conceptual integration derived from the results. Building on the Kuckartz-based thematic analysis and the subsequent TEA/TEM-informed temporal reinterpretation, the figure integrates the district-level patterns shown in Additional file 3: Figure S2 and Additional file 4: Figure S3. The upper panels summarize district-specific relational infrastructures, the middle panel summarizes shared pre-BFP mutual-aid practices, the lower panel shows how later-life triggers reconfigured these practices into later-life EFP and P-EFP pathways, and the bottom note highlights the cross-district relational boundaries that continued to constrain mutual aid. The model is grounded in the empirical findings but represents an analytical abstraction rather than a direct visualization of raw data or a full TEM diagram

Theme 1. In the fishing district, mutual aid rested on long-term co-residence, routine visibility of daily signs, neighbourly access to homes, and tacit reciprocity

Residents in the fishing district described mutual aid as emerging from longstanding everyday familiarity. Neighbours routinely noticed signs such as doors opening later than usual, curtains not moving, or laundry left outside in the rain. This routine visibility of daily signs functioned as an informal monitoring system, especially for older adults living alone.

Practical help was enabled by spatial closeness and neighbourly access to homes when necessary. Some residents described houses being left unlocked, or entry through a back door being possible, so that neighbours could bring in washing when rain started or check on a resident who had not appeared. One resident stated, “If the door doesn’t open, everyone worries and asks, ‘What happened today?’” Another described how neighbours would “come in and bring in the laundry because the door is unlocked.”

These practices were not framed as exceptional altruism but as ordinary reciprocity. Help received after illness or injury was narrated as something that should later be returned. In this sense, tacit reciprocity referred to an expectation that help, concern, and practical assistance would be reciprocated over time rather than repaid immediately or formally.

Theme 2. In the I-turn district, mutual aid was built through participation, transport arrangements, and selectively cultivated friendships

In the I-turn district, mutual aid was less often described as inherited from long shared history and more often as something actively built through participation in circles, exercise groups, singing activities, and friendships formed after migration. These were not universal ties with all neighbours, but selectively cultivated relationships with activity companions and nearby non-kin others.

Mutual-aid acts in this district included information exchange, rides to the station, shopping assistance, and gentle monitoring by phone or through routine visual signs. Here, gentle monitoring refers to low-intensity checking that remained socially acceptable because it stopped short of intrusive surveillance. One resident explained that if her lace curtains remained closed, nearby residents would notice and call. Another described neighbours bringing food after back pain or giving rides to community activities.

Mutual aid in this district therefore depended on ongoing participation and the maintenance of selectively cultivated relationships. Rather than being taken for granted, mutual aid was built through repeated contact and through residents’ willingness to approach particular others with whom trust had developed.

Theme 3. Across both districts, ageing and health events shifted residents between providing help and receiving help

Across both districts, residents described mutual aid as something that changed over time rather than remaining fixed. Health events such as heart surgery, fractures, falls, pain, reduced mobility, and changes in strength or stamina altered what residents could do for others and what they needed from others.

Several participants explicitly contrasted an earlier period in which they had mainly helped others with their current position as recipients of help. One fishing-district resident said, “I’m the one being helped now.” Another explained that after heart surgery, neighbours became concerned if her door was not opened at the usual time. After a fall and clavicle fracture, food and other practical support became especially meaningful because standing in the kitchen was difficult.

These narratives suggest that mutual aid in later life is not a one-way attribute of either independence or dependence. Rather, residents moved along a changing continuum between providing help and receiving help, and these shifts were iterative and context-dependent rather than linear. At times, this reconfiguration also included refraining from help-seeking.

Theme 4. Mutual aid boundaries were clear and coexisted with restraint, family responsibility, and uneven involvement in local activities and ties

Participants in both districts emphasized that mutual aid was not unlimited. They valued mutual aid, but they also valued restraint. In the I-turn district, residents sometimes described deliberately holding back from asking neighbours for help, even when support might have been available. In the fishing district, dense familiarity did not erase concerns about burdening others or interfering excessively. In this sense, mutual aid boundaries referred not to district boundaries but to the relational limits of what neighbours felt able, willing, or entitled to do.

Residents also distinguished neighbourhood help from family support. Some needs were managed within the family and therefore not brought into neighbourhood relationships. In addition, not all residents were equally involved in local activities or neighbourhood ties. Some participants noted that a few neighbours avoided local activities or kept distance from others, making mutual aid selective rather than universal.

In the fishing district, some narratives also suggested generational change and a weakening of older collective routines. Although dense reciprocity remained visible among some long-term residents, others pointed to situations in which younger people or less connected residents were less involved. Mutual aid was therefore sustained through local norms, but it was also fragile, uneven, and bounded.

Together, these four themes informed the comparative model shown in Fig. 1. The model is grounded in the empirical findings but represents an analytical abstraction rather than a direct visualization of raw data.

Discussion

This study showed that mutual aid among older residents was neither uniform across districts nor fixed over time. Instead, it was anchored in district-specific relational infrastructures. In the fishing district, mutual aid relied on long-term co-residence, routine visibility of daily signs, neighbourly access to homes, and a moral logic of reciprocity. In the I-turn district, it relied more on participation, transport, and selectively cultivated friendships created after migration.

This contrast extends previous Japanese findings on social relationships in later life. Quantitative studies have shown that social support, neighbourhood cohesion, and social participation are associated with mortality, depressive symptoms, and disability-related outcomes among older adults in Japan [5, 6, 11–15]. Our findings help explain how these associations may be lived in practice. In the fishing district, social cohesion was embedded in everyday routine and tacit observation. In the I-turn district, social participation appeared to operate as a relational gateway through which support became possible.

The study also suggests that mutual aid should be understood as reconfigured over time. Mutual aid was experienced not as a fixed role but as a shifting position between providing and receiving help over time. These shifts were iterative and context-dependent rather than linear, and at times included refraining from help-seeking. This challenges static or deficit-based interpretations of ageing and care. Ageing, illness, injury, and household change redirected residents from one position to another, sometimes repeatedly. This resonates with rural qualitative work showing that support and life satisfaction are intertwined with local relational worlds [10], but our study adds a comparative temporal perspective by showing that the same municipality may contain distinct routes into support and distinct thresholds for receiving it.

At the same time, mutual aid was not inherently benign or limitless. Residents valued help while also carefully policing the relational boundaries of involvement. This restraint appeared in concerns about intruding into another person’s life, in the tendency to distinguish neighbourhood support from family responsibility, and in recognition that not everyone joins local activities or neighbourhood ties to the same degree. Accordingly, the practical challenge for community policy is not simply to increase mutual aid in the abstract. Rather, it is to support the relational infrastructures that make low-burden, acceptable, and reciprocal forms of helping possible.

For practice and ageing-in-place policy, the implications differ by district type. In long-established districts, preserving light-touch routines of visibility and neighbourly access may matter as much as creating new formal programs. In in-migrant districts, by contrast, sustained opportunities for repeated contact—exercise groups, accessible meeting places, low-threshold transport, and small collaborative activities—may be more important for gradually building the relational basis of mutual aid. In both settings, neighbourhood-based strategies should strengthen everyday, low-burden contact while recognizing that mutual aid complements rather than replaces formal care and family support.

Taken together, these findings suggest that mutual aid in ageing communities is better understood as a temporally reconfigured relational process shaped by district-specific infrastructures than as a fixed attribute of a district or its residents.

Methodologically, the combination of Kuckartz and TEA/TEM was productive for this study. Kuckartz supported systematic within-case and cross-case coding, whereas TEA/TEM helped the team reinterpret the same material in terms of irreversible time, branching, and converging pathways. The district-specific cross-case TEM diagrams were especially useful for clarifying how later-life events redirected residents toward providing help, receiving help, refraining from help-seeking, or externally substituted support. For reporting clarity, however, the main text presents a synthesized comparative figure rather than full case-level diagrams.

Recent methodological studies suggest that large language models may identify descriptive qualitative patterns efficiently but remain weaker at contextual thick description, theoretical derivation, and culturally nuanced Japanese data [30–32]. This is why AI was positioned only as a supplementary audit in the present study. The codes, themes, interpretations, and final wording reported here were generated and finalized by the human research team.

These findings may be useful beyond Japan in ageing localities where long-established communities and retirement- or amenity-led in-migrant districts coexist, because they show that mutual aid depends not only on need but also on the local social and spatial conditions that make noticing, asking, and accepting help possible. Future research should examine whether similar district-specific relational infrastructures and mutual aid boundaries operate in other municipalities and how local interventions can strengthen acceptable low-burden reciprocal ties without displacing formal services.

Strengths and limitations

A major strength of this study is the direct comparison of two socially different districts within the same municipality. This design reduced broad administrative variation while allowing district-specific contrasts to emerge.

A second strength is the inclusion of follow-up interviews for part of the sample, which helped clarify temporal change and the reconfiguration of mutual aid after health events and later-life transitions.

The study also has limitations. First, the sample was small and purposive, and the findings are intended for analytical rather than statistical transferability. Second, follow-up data were not available for all participants. Third, because the present paper intentionally focused only on residents, it does not include welfare commissioners’ perspectives. Fourth, English quotations were translated from Japanese and may not capture every cultural nuance of the original speech.

Conclusions

Mutual aid among older residents in this municipality was shaped by district-specific relational infrastructures and reconfigured over time through ageing, illness, mobility change, and the social boundaries of everyday relationships. A comparative qualitative approach showed that mutual aid in later life was neither uniform across districts nor static over time. Community and ageing-in-place interventions should therefore be tailored to the relational infrastructure already present in each district and should support modest, repeatable opportunities for reciprocal contact and gentle monitoring, while recognizing the limits of neighbour support. Future research should test how these relational processes operate in other ageing localities and how formal services can most effectively work alongside locally grounded mutual aid.

Abbreviations

BFP

Bifurcation point

COREQ

Consolidated Criteria for Reporting Qualitative Research

EFP

Equifinality point

OPP

Obligatory passage point

P-EFP

Polarized equifinality point

SG

Social guidance

SD

Social direction

TEA/TEM

Trajectory Equifinality Approach / Trajectory Equifinality Modeling

Authors’ contributions

YO: Conceptualization, Methodology, Investigation, Formal analysis, Validation, Supervision, Project administration, Writing – original draft, Writing – review & editing. ET: Investigation, Formal analysis, Validation, Writing – review & editing. YI: Investigation, Formal analysis, Validation, Writing – review & editing. KM: Formal analysis, Validation, Writing – review & editing. ETA: Investigation, Resources, Validation, Writing – review & editing. MY: Investigation, Resources, Validation, Writing – review & editing. YS: Investigation, Resources, Validation, Writing – review & editing. All authors rechecked the analyses retained in the final manuscript, approved the final version, and take responsibility for its content.

Funding

This study received no specific external funding. The research activities were supported by approved individual research funds from the authors’ respective institutions. The article processing charge for this manuscript is planned to be supported by JSPS KAKENHI Grant Number JP24K14011. The funder had no role in the design of the study; data collection, analysis, or interpretation; writing of the manuscript; or the decision to submit the article.

Data availability

De-identified analytic materials supporting the conclusions of this article are included within the article and its additional files. Additional de-identified transcript extracts from the first- and second-round letter-coded resident interviews are available from the corresponding author on reasonable request, subject to ethics approval and participant confidentiality. Full transcripts are not publicly available because the study was conducted in a small municipality and complete contextual records could permit deductive identification.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and relevant guidelines and regulations. Ethical approval was originally obtained from the Dokkyo Medical University Nursing Research Ethics Committee on 5 November 2021 and was subsequently approved by the Research Ethics Review Committee of Gunma University of Health and Welfare (approval no. RS23-21; approved 21 March 2024). All participants received oral and written explanations of the study, and written informed consent was obtained prior to participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.