Introduction
Spousal loss is one of the most profoundly disruptive lifecourse events in later life (Holmes and Rahe Reference Holmes and Rahe1967; Infurna and Luthar Reference Infurna and Luthar2017; Yu et al. Reference Yu, Kahana, Kahana and Han2021; Fang and Carr Reference Fang and Carr2024; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a). It is associated with suffering, health decline and increased mortality (Lee and Carr Reference Lee and Carr2007; Manzoli et al. Reference Manzoli, Villari, Pirone and Boccia2007; Jones et al. Reference Jones, Bartrop, Forcier and Penny2010; van den Berg et al. Reference van den berg, Lindeboom and Portrait2011; Buckley et al. Reference Buckley, Sunari, Marshall, Bartrop, McKinley and Tofler2012; Utz et al. Reference Utz, Caserta and Lund2012; Holland et al. Reference Holland, Futterman, Thompson, Moran and Gallagher-Thompson2013; Shear et al. Reference Shear, Ghesquiere and Glickman2013; Ang Reference Ang2023). Additionally, given how intimate partnerships are central to daily life and self-understanding, spousal loss often entails redefining identity and existence (Naef et al. Reference Naef, Ward, Mahrer-Imhof and Grande2013; Richardson Reference Richardson2014; Larsson et al. Reference Larsson, Rämgård and Bolmsjö2017; Carr Reference Carr, Wister and Cosco2021; Fang and Carr Reference Fang and Carr2024).
The concept of resilience is widely used to explain positive outcomes among older adults facing diverse adversities (Cosco et al. Reference Cosco, Howse and Brayne2017; Infurna Reference Infurna2020; Merchant et al. Reference Merchant, Aprahamian, Woo, Vellas and Morley2022). Why adverse outcomes persist for some but not others is of great interest in the social sciences and psychology and has implications for targeted prevention and support in ageing societies (Hicks and Conner Reference Hicks and Conner2014; Pruchno et al. Reference Pruchno, Heid and Genderson2015; Cosco et al. Reference Cosco, Howse and Brayne2017; Beeris et al. Reference Beeris, Niemeijer and Machielse2022; Merchant et al. Reference Merchant, Aprahamian, Woo, Vellas and Morley2022; Brinkhof et al. Reference Brinkhof, Ridderinkhof, Bieleke, Murre, Krugers and de Wit2024, Reference Brinkhof, Ridderinkhof, de Wit, Krugers and Murre2025; Trică et al. Reference Trică, Golu, Sava, Licu, Zanfirescu, Adam and David2024). Although resilience is widely described as a dynamic process of adaptation and transformation (Fisher et al. Reference Fisher, Ragsdale and Fisher2019; Angevaare et al. Reference Angevaare, Roberts, van Hout, Joling, Smalbrugge, Schoonmade, Windle and Hertogh2020), it is often operationalized in ways that flatten everyday complexity and ambiguity (Cristancho Reference Cristancho2016). For example, even though resilience does concern individuals, its framing is often criticized for over-individualizing responsibility, obscuring structural and contextual constraints (Bottrell Reference Bottrell2013; Wild et al. Reference Wild, Wiles and Allen2013; Evans and Reid Reference Evans and Reid2014; Stephens et al. Reference Stephens, Breheny and Mansvelt2015; Gill and Orgad Reference Gill and Orgad2018; Beeris et al. Reference Beeris, Niemeijer and Machielse2022). This framing risks depicting those in difficulty as personally at fault or ‘unsuccessful’ in ageing, which may discourage help-seeking (Portacolone Reference Portacolone2011; Beeris et al. Reference Beeris, Niemeijer and Machielse2022). Infurna and Luthar (Reference Infurna and Luthar2017) therefore argue that resilience research should move beyond labelling individuals as ‘resilient’ to examine the conditions shaping trajectories after adversity. This shift is supported by multiple studies. For example, Wiles et al. (Reference Wiles, Wild, Kerse and Allen2012) show that, although discussions of resilience start with individual traits, focus group participants highlight its multi-layered, contextual nature. Similar emphases appear in Gibb (Reference Gibb2018) on ageing in remote places, Stephens et al. (Reference Stephens, Breheny and Mansvelt2015) using Sen’s capability approach to foreground context, and Nakashima and Canda (Reference Nakashima and Canda2005) linking resilience in end-of-life experiences to cultural embeddedness.
Although spousal loss in later life has been well studied, including qualitative work that documents older adults’ everyday experiences after the death of a spouse (Almevall et al. Reference Almevall, Juuso, Melander and Zingmark2024; Bennett and Vidal-Hall Reference Bennett and Vidal-Hall2000; DeMichele Reference DeMichele2009; van den Hoonaard et al. Reference van den Hoonaard, Bennett and Evans2014; Vasara Reference Vasara2020), relatively few studies explicitly examine how resilience emerges in this context (e.g. Bonanno et al. Reference Bonanno, Wortman and Nesse2004; Bennett Reference Bennett2010; Bennett et al. Reference Bennett, Morselli, Spahni and Perrig-Chiello2020). Research further indicates that older adults elucidate the complexities of resilience as a situated process when in dialogue, and that first-person accounts of resilience and spousal loss add distinct value to understanding how resilience operates in daily life (Wiles et al. Reference Wiles, Wild, Kerse and Allen2012; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a). Yet within research on resilience after spousal loss, participatory, dialogical studies that foreground older adults’ interpretative work are, to the authors’ knowledge, virtually absent. Taken together, this reveals crucial conceptual and empirical gaps in understanding resilience in the context of spousal loss, despite the prevalence and profound impact of this life event in later life. This article addresses these gaps using a participatory research design that foregrounds ambiguity, ambivalence and complexity in participants’ experiences and examines the conditions under which resilience emerges after spousal loss. The guiding research question is: how do older adults experience spousal loss and resilience in their daily lives?
Spousal loss and resilience
Spousal loss is often discussed in terms of bereavement, the period immediately after a partner’s death, and widowhood, the longer-term state of living without the partner (Bennett and Soulsby Reference Bennett and Soulsby2012). Grief, central to both phases, has been theorized extensively, from Freud’s mourning and melancholia (Freud Reference Freud and Strachey1957) to contemporary frameworks such as Worden’s task-based model, the dual process model (Stroebe and Schut Reference Stroebe and Schut1999, Reference Stroebe and Schut2010), theories of continuing bonds (Klass Reference Klass2006; Klass et al. Reference Klass, Silverman and Nickman2014) and the integrated process model of loss and grief (Guldin and Leget Reference Guldin and Leget2024). Together, these frameworks and theories emphasize the idiosyncratic, fluctuating nature of grief, showing that bonds with the deceased may endure and that experiences of loss shift across the lifecourse. This is relevant to experiences of resilience in bereavement and widowhood as it influences how people navigate spousal loss over time (Guldin and Leget Reference Guldin and Leget2024; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a).
These continuities and changes in grief align with contemporary views of resilience as a dynamic process of adaptation and transformation in the face of adversities (Infurna and Luthar Reference Infurna and Luthar2017, Reference Infurna and Luthar2018; Bourbeau Reference Bourbeau2018; Fisher et al. Reference Fisher, Ragsdale and Fisher2019; Angevaare et al. Reference Angevaare, Roberts, van Hout, Joling, Smalbrugge, Schoonmade, Windle and Hertogh2020; IJntema et al. Reference IJntema, Schaufeli and Burger2023; Voie et al. Reference Voie, Wiles, Blix, Kristiansen, Helgesen and Mæhre2025; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a). Adversities, often described as stressors, challenges or risks (Angevaare et al. Reference Angevaare, Roberts, van Hout, Joling, Smalbrugge, Schoonmade, Windle and Hertogh2020), can be understood as embodied, emplaced experiences of ‘pain, disruption, exhaustion, disorientation, loneliness, or grief’ in both exceptional and commonplace circumstances (Wright-St Clair et al. Reference Wright-St Clair, Kerse and Smythe2011; Power et al. Reference Power, Bell, Kyle and Andrews2019). Navigating adversities depends on mobilizing resources such as physical health, cognitive functioning, adequate housing, supportive relationships and professional support (MacLeod et al. Reference MacLeod, Musich, Hawkins, Alsgaard and Wicker2016; Angevaare et al. Reference Angevaare, Roberts, van Hout, Joling, Smalbrugge, Schoonmade, Windle and Hertogh2020). Accordingly, resilience emerges not only from individual characteristics but from interactions among individuals, relationships and contexts that shape how people deal with adversities over time (Fisher et al. Reference Fisher, Ragsdale and Fisher2019; Wilson et al. Reference Wilson, Walker and Saklofske2021).
Alongside adversities and resources mobilized, a third key element in resilience is outcomes: observable adjustments and trajectories across life domains as people navigate adversities (Bonanno et al. Reference Bonanno, Wortman and Nesse2004; MacLeod et al. Reference MacLeod, Musich, Hawkins, Alsgaard and Wicker2016; Infurna and Luthar Reference Infurna and Luthar2017, Reference Infurna and Luthar2018). Across studies, outcomes are defined in various ways but typically span a continuum from less desirable (e.g. vulnerability, rigidity) to more desirable (e.g. transformation, recovery, sustainability, thriving) (Angevaare et al. Reference Angevaare, Roberts, van Hout, Joling, Smalbrugge, Schoonmade, Windle and Hertogh2020; Beeris et al. Reference Beeris, Niemeijer and Machielse2022; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a; IJntema et al. Reference IJntema, Schaufeli and Burger2023). Outcomes are rarely uniform within a single life: after spousal loss, for instance, a person may maintain social functioning yet experience mood decline or recover psychological wellbeing while financial strain persists. Negative and positive outcomes can therefore coexist (Lenette et al. Reference Lenette, Brough and Cox2013; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a). Outcomes also feed back into resilience processes by reshaping perceptions, opportunities and resources (Angevaare et al. Reference Angevaare, Roberts, van Hout, Joling, Smalbrugge, Schoonmade, Windle and Hertogh2020; IJntema et al. Reference IJntema, Schaufeli and Burger2023).
Adversities, the resources people mobilize and the outcomes they experience are widely understood as dynamic, shifting over time and across contexts and continuously shaping experiences of resilience (Beeris et al. Reference Beeris, Niemeijer and Machielse2022; Glas et al. Reference Glas, Rhebergen, Widdershoven, Huisman and Kok2025; Voie et al. Reference Voie, Wiles, Blix, Kristiansen, Helgesen and Mæhre2025). Several authors have therefore shifted from understanding resilience as ‘bouncing back’ to a pre-adversity state towards ‘bouncing forward’ (Manyena et al. Reference Manyena, O’Brien, O’Keefe and Rose2011; Bourbeau Reference Bourbeau2018; Hynes et al. Reference Hynes, Trump, Love and Linkov2020; Métais et al. Reference Métais, Burel, Gillham, Tarquinio and Martin-Krumm2022). This shift is often attributed in social sciences and psychology to Walsh’s (Reference Walsh2002) work on family resilience after the 9/11 attacks in the United States. It emphasizes that there is no simple return to ‘how things were’ after adversity, but rather a potentially transformative reorientation to a changed reality. Despite the popularity of using the metaphor of ‘bouncing forward’ to signal dynamic, transformative responses to adversity, it has also been critiqued as conceptually vague, normatively loaded and inattentive to the mechanisms and lived processes it purports to describe (Sleijpen et al. Reference Sleijpen, June ter Heide, Mooren, Boeije and Kleber2013; Horn Reference Horn2021; Clark Reference Clark2021b; Métais et al. Reference Métais, Burel, Gillham, Tarquinio and Martin-Krumm2022). Such critique underscores the need to ground understandings of resilience in situated, everyday experiences (Infurna and Luthar Reference Infurna and Luthar2018; Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a). This article therefore examines how resilience after spousal loss unfolds in the everyday lives of older adults. It draws on two complementary theoretical perspectives to deepen interpretation of the findings: legal scholar and resilience theorist Janine Natalya Clark’s (Reference Clark2021b) holonic framework, and political philosopher Hannah Arendt’s (Reference Arendt2018 [1958]) theory of action, plurality and natality in a shared world.
First, Clark (Reference Clark2021a, Reference Clark2021b) urges caution in describing resilience as ‘bouncing forward’, as this can carry positive connotations that obscure the severity of adversity, imply linearity in a non-linear process and suggest uniform improvement across life domains, even though gains and declines often coexist. She instead proposes a holonic framework that conceptualizes resilience as ‘a dialectical process of expansion and contraction’ (Clark Reference Clark2021b, p. 557). In this view, people are nested holons (wholes that are also parts) and resilience unfolds both within the individual holon and in relation to wider social holons. ‘Expansion’ denotes periods when opportunities and capacities for favourable outcomes widen through supportive systemic interactions, whereas ‘contraction’ refers to narrowing or withdrawal. Because these levels are nested, change at one level can ripple through others, expanding or contracting a person’s room for action (Clark Reference Clark2021b). Viewed through this lens, resilience after spousal loss can be understood as ongoing experiences of expansion and contraction across individual, relational, network and service domains.
Second, Arendt’s (Reference Arendt2018) account of action reinforces Clark’s (Reference Clark2021a, Reference Clark2021b) emphasis on interdependence between individuals and their social and systemic contexts, while keeping the acting individual clearly in view. In Arendt’s account, action is inherently relational. It becomes possible when a person is received and recognized by others within a situated network of relationships, and it unfolds in a shared world of institutions, rules, practices and artefacts that render words and deeds intelligible (Arendt Reference Arendt2018, pp. 50–56, 181–188, 198–199). Action therefore depends not only on what individuals do but also on how they are recognized and positioned in relation to others. Also central to Arendt’s work is natality, the capacity to begin anew through words and deeds within a shared world structured by relationships, material things and institutions (Arendt Reference Arendt2018, pp. 8–9, 176–177). Natality draws attention to how new beginnings may, or may not, become possible within these shifting conditions. In this article, natality provides an interpretive lens for examining how older adults navigate a new everyday reality after spousal loss.
Methods
Research design
This study employs participatory action research (PAR) to foreground older adults’ perspectives on spousal loss and resilience. The PAR design is an approach through which knowledge is co-produced via collaborative inquiry between researchers and participants. It prioritizes research with rather than about those whom it concerns, aiming to empower participants to take action and create locally meaningful change (Abma et al. Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019c). This makes PAR well-suited to the study’s relational view of resilience as a process developing through interaction and shared meaning-making. Through structured dialogue, it enables co-researchers to directly experience how resilience takes shape within the often-challenging relational circumstances of everyday life. This strengthens data interpretation (Kralik et al. Reference Kralik, van Loon and Visentin2006; Beeris, Niemeijer and Machielse Reference Beeris, Niemeijer and Machielse2025b) and supports knowledge generation and social change relevant to the target group (Abma et al. Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019a). The first author acted as facilitating researcher, overseeing implementation.
Co-researchers
Co-researchers were purposively recruited across one rural and two urban small- to medium-sized Dutch municipalities through social care and health-care networks, flyers, social media and local newspapers. Eligible applicants were aged 50 or older, interested in participatory research on resilience after spousal loss, and able to conduct peer interviews in a manner that safeguarded themselves and participants. Each applicant met the facilitating researcher individually for eligibility screening and the final group comprised 21 co-researchers, seven per municipality. A few applicants were not selected as expectations did not match or they were considered unsuitable for the peer-interviewer role. All selected co-researchers provided a Certificate of Good Conduct (Verklaring Omtrent het Gedrag, VOG) issued by Dutch authorities, confirming no objections for peer-interviews. Table 1 presents their ages and genders.
Age and gender co-researchers
Table 1 Long description
Co-researcher demographics are summarized by age and gender. Ages span from 48 to 78, and the mean age is 70. Gender counts show 15 female co-researchers and 6 male co-researchers, indicating a predominantly female group. The age distribution centers on older adulthood, with one participant notably younger than the rest. No additional breakdown by specific ages or other demographic characteristics is provided.
Note: One 48-year-old co-researcher was included for their relevant life experience on resilience in later life.
To prepare for peer-interviews, the facilitating researcher organized three meetings with co-researchers in each municipality. The first established ground rules for open and respectful dialogue and explored resilience through interactive exercises. The second focused on interview techniques and co-designing an interview guide. This guide provided practical tips for planning, conducting and concluding interviews. The third meeting finalized roles and safeguards for co-researchers and participants, and a semi-structured topic list of eight guiding questions across three themes: past (experiences surrounding the loss), present (its continuing impact) and understandings of resilience. The topic list was originally developed for a prior study that examined everyday experiences of resilience and spousal loss (Beeris, Niemeijer, Bisseling et al. Reference Beeris, Niemeijer, Bisseling and Machielse2025a). It was designed with open questions to uncover such experiences in detail (Kvale and Brinkmann Reference Kvale and Brinkmann2009). In keeping with PAR principles (Abma et al. Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019a, Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019c), co-researchers reviewed the list and collectively agreed that it was suitable for the present study, requiring only minor adjustments. The finalized topic list is available in the online supplementary material. The facilitating researcher circulated both the interview guide and the topic list with co-researchers, ensuring consistent use across all interviews.
Interview participants
After the preparatory meetings, participants were purposively recruited through the same channels as the co-researchers and matched with the trained co-researchers. Individuals interested in an interview contacted a local social care professional for telephone screening to verify eligibility and identify possible urgent care needs. Eligible individuals were referred to the facilitating researcher.
Inclusion criteria were:
• Age: 50 years or older. The 50-plus threshold reflects a life stage in which age-related gains and losses become more salient (Merkt et al. Reference Merkt, Haesen, Meyer, Kressig, Elger and Wangmo2020; TESTAD et al. Reference Testad, Ushakova, Aakre, Sabatini and Gjestsen2022).
• Loss: Spousal loss.
• Timing: At least one year since the partner’s death, to reduce emotional burden and enable reflection on grief and resilience.
• Health: Sufficient for an interview of at least 45 minutes without excessive burden.
Overall, 50 interviews took place with 33 female and 17 male participants, aged between 54 and 94 years (mean age 73). Table 2 shows the age and gender distribution across the three anonymized municipalities, referred to as Municipalities A, B and C.
Age of interview participants
Table 2 Long description
Age distribution of 50 interview participants is broken down by age band, municipality, and sex. The largest group is ages 65–74 with 18 participants, followed by 75–84 with 13 and 85–94 with 10. Ages 55–64 include 7 participants, and ages 50–54 include 2, all female. Women are the majority overall with 33 participants compared with 17 men, and women outnumber men in every age band. By municipality totals, municipality A has 21 participants, municipality B has 19, and municipality C has 10. Municipality A is mostly female, municipality B is closer to balanced, and municipality C has more females than males.
Note: MM = males per municipality (A, B, C); FM = females per municipality (A, B, C).
Data collection
The facilitating researcher matched each eligible participant with a co-researcher via email. The co-researcher then scheduled a two-hour interview and sent an information letter, data collection protocol and consent forms by email or post. These documents were co-designed with co-researchers to ensure accessible language.
Participants chose the interview location, usually their home or a private room in a community centre. Before each interview, participants signed the consent form in duplicate (one copy for themselves and one for the co-researcher). This confirmed receipt and understanding of project and data management information, their right to withdraw at any time and their consent to participate, including preferences for audio recording and receiving project results. Contact between co-researchers and participants was limited to the single interview to support clear boundaries. With permission, additional needs (e.g. questions about health care, wellbeing or local activities) were referred to the facilitating researcher, who contacted a social care professional; this occurred once. If permission was not granted, co-researchers provided information on support services. If interviews became overwhelming, co-researchers were trained to suggest a break or, if needed, reschedule or end the interview. A break was suggested once. Additionally, with 21 co-researchers and 50 interview participants, most co-researchers conducted multiple interviews based on availability and preference. A minimum of two interviews per co-researcher was recommended to support analysis. Only one co-researcher conducted a single interview owing to personal circumstances.
The facilitating researcher’s ongoing availability and support of the co-researchers was an important aspect of data collection. After each interview, she checked in with each co-researcher, offering them additional support via an extra phone call or home visit when interviews were challenging or concerns persisted. A phone call was requested twice. This supported reinforced collaboration and trust within the research team.
After debriefing, co-researchers wrote an anonymized report structured around the topic list. Report styles varied, from narrative accounts and collections of direct quotations to more condensed responses aligned with the topic list questions. For each participant, only the age, gender and residential setting (city, town or rural) were recorded; additional personal details for participants wishing to receive updates (name, address, telephone number and email) were stored separately. Audio files, recorded on devices provided by the facilitating researcher, were used to prepare the reports and deleted after completion.
Data analysis
After the interviews, co-researchers in each municipality met in two rounds of meetings to analyse the reports. First, small groups of three to four co-researchers met for two-hour sessions. They exchanged findings along the topic list themes of past, present and resilience and discussed participants’ experiences in depth. Second, larger groups of six to ten co-researchers met for three-hour sessions to synthesize insights from the smaller groups. Reflections were discussed first in pairs or trios and then summarized in plenary on large sheets organized by topic list theme. This meeting also offered space to reflect on co-researchers’ roles and experiences, enhancing transparency. Together, the meetings promoted shared ownership of results and analytic validity by incorporating multiple perspectives across the three locations (Abma et al. Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019b). In total, 39 hours of meetings took place from interview preparation to the end of analysis. Every meeting was attended and facilitated by the facilitating researcher.
In addition to the collective analysis, the facilitating researcher coded all 50 interview reports in Atlas.ti, using the topic list themes to ensure alignment with the co-researchers’ meetings. Within these themes, codes were developed inductively from the data, keeping findings grounded in participants’ accounts. This coding functioned as a secondary layer of ‘sense-making’ (Abma et al. Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019b), enriching and refining interpretations from the collective sessions. It also made insights traceable across interviews and easier to audit as co-created research data, thereby strengthening dependability (Lincoln and Guba Reference Lincoln and Guba1985). All interview reports, Atlas.ti files and contact information were uploaded to a secure drive at the University of Humanistic Studies in Utrecht, Netherlands. Each report received a unique identification code, with the code list stored securely on the same drive.
To document analytic decisions and support verification, the facilitating researcher created an audit trail in Excel, listing all Atlas.ti codes with their frequency, spread (number of interviews), English translation, study-specific definition and valence (positive, negative or neutral). Codes were not collapsed into broader thematic categories to avoid adding an extra interpretative layer and overly abstracting participants’ lived experiences. This reflects the low-inference qualitative descriptive approach of this study (Sandelowski Reference Sandelowski1995, Reference Sandelowski2000). Results are therefore presented in alignment with the topic list structure, covering past (bereavement experiences), present (widowhood) and resilience (what helped/hindered across these periods).
The entire process, from recruiting co-researchers to completing data analysis, spanned 12 months (2023–2024). While this article focuses on the interview report data, information on the dialogue sessions is published in a separate paper (Beeris, Niemeijer and Machielse Reference Beeris, Niemeijer and Machielse2025b).
Results
This section presents participants’ accounts of loss and resilience from co-researchers’ reports, organized around four themes: (1) the experience of bereavement; (2) what helped or hindered around the time of the partner’s death; (3) the experience of widowhood; and (4) what helps or hinders during widowhood. For each theme, the three most frequent codes are illustrated with quotations; less frequent codes are not elaborated here but remain meaningful. This is revisited in the methodological considerations. Ellipses (…) indicate omitted text; words in square brackets mark clarifications where the original wording may be unclear (e.g. [her partner]). As co-researchers had flexibility in writing reports, some included verbatim quotations while others wrote third-person summaries; this section therefore contains both. Retaining original phrasing reflects this article’s commitment to low-inference description, staying close to the co-researcher’s and participants’ own words and limiting further abstraction (Sandelowski Reference Sandelowski1995, Reference Sandelowski2000). Interview participants are anonymized with letter-number codes (e.g. F1, M1), where F denotes female and M male. Co-researchers are not named.
Experience of bereavement
The most frequent codes here were unexpected death, expected death owing to illness, and care-giving. These reflect how participants experienced their partner’s dying and the immediate aftermath.
Unexpected death
For many participants, their partner’s death came without warning and was hard to comprehend. For example, a male participant recounted how his wife ‘suddenly passed away in her chair … after 55 years of marriage’, leaving him ‘completely devastated’ and unsure ‘how to go on’, angry for months and asking why it had happened to him (M2). Similarly, a female participant described the acute shock of her husband’s abrupt diagnosis: ‘It started with pain while cycling … we found ourselves in the hospital. Three hours later: pancreatic cancer, untreatable … A bolt of lightning from a clear sky’ (F4). For another female participant, the reality of the unexpected loss sank in during the everyday act of driving home through rush hour: ‘The journey home [from work] took forever because of the traffic, and somewhere along the way I realized I was on my way to our house, to my dead husband’ (F5).
Expected death owing to illness
Many participants anticipated their partner’s death because of various illnesses. For some, advance knowledge allowed them to prepare and make the most of their time together. One female participant, for example, emphasized that knowing death was approaching helped her and her husband to ‘enjoy … those last months to the fullest’ (F2), and another described the period as ‘both bewildering and beautiful’, marked by small celebrations and ‘togetherness, love, and trust’ (F3). For others, however, advance knowledge did not soften the experience. Many described the dying phase as traumatic. One female participant, for example, recalled her partner’s final night in hospital as ‘chaotic and frightening’: ‘He became very aggressive … I called for the nurses … It took ten minutes before anyone came … it became a traumatic goodbye’ (F1). Others also stressed that expectation did not eliminate shock. One male participant shared, for instance, how he felt prepared in practical terms but not emotionally, noting that he ‘had no real sense of what the grieving process would actually entail … someone who hasn’t lost a partner cannot imagine what grief does to you’ (M1).
Care-giving
Intensive care-giving often shaped later experiences of bereavement. Although this occurred before death, participants frequently described it when discussing bereavement. Where illness was long term, especially dementia, they depicted a drawn-out ‘goodbye’ that reshaped the moment of death itself. As one female participant put it: ‘The grieving process had already been going on for quite some time … I kept saying goodbye little by little’ (F8). When he passed, she felt prepared, noting that the man she had known over the years was, in many ways, already gone for a long time. Another female participant similarly felt that she had ‘already taken a head start on the grief that was bound to come, by having to face the reality of the impending death’ (F7). For others, immersion in daily care had the opposite effect and obscured how close death was. A female participant explained that being with [her partner] every day made the decline harder to see: ‘[The children] saw that … their father didn’t have long to live. I hadn’t seen it that clearly yet’ (F6).
What had helped or not helped when the partner died
The most frequent codes here were characteristics, network and professional support. These capture what participants found helpful or unhelpful around their partner’s death.
Characteristics
This code captures personality traits that participants identified as both helpful and unhelpful during bereavement. For example, many described characteristics that they believed helped them. A female participant described how some people avoided her in the supermarket because they ‘don’t know what to say’. When an acquaintance switched aisles, however, she followed and told them that she ‘does not want to be avoided’ (F14). Another female participant described a grounded, open stance as helpful and part of who she is: ‘I’m down to earth, easy-going. I’m open to discussion, open to life. I think this fits with who I am’ (F13). Prior loss experiences also reinforced a sense of personal capacity. A female participant, for instance, reflected on the loss of her son: ‘Back when I thought I would never be able to laugh or enjoy anything … I picked myself up and embraced life … So, I told myself: I [now] have to face [my partner’s death]. And I can do this’ (F15). At the same time, participants recognized a double edge to certain traits. This was emphasized, for example, by a female participant in how she was ‘glad that she is someone who can take action and persevere’, while acknowledging ‘that pushing through isn’t always good’ (F2). Others pointed to self-limiting characteristics formed earlier in life. As noted by a co-researcher, for instance, a female participant explained that she had ‘developed fears a long time ago. In the past, her partner handled this for her’. Managing these fears alone after his death proved difficult (F12).
Network
This code concerns wider social networks, excluding family and close friends, in navigating bereavement. Several participants highlighted supportive aspects of these networks. A female participant described neighbours’ care upon returning home, where her dead husband had been laid out on the couch: ‘Two neighbours were there. They had dimmed the lights, lit some candles … It was really beautiful … the warmth and care from people in the neighbourhood’ (F5). Structured support also helped. A male participant, for example, pointed to groups and clubs, including ‘support groups … social club … cooking club … and cycling buddies … They were great listeners’ (M1). Another male participant described how a neighbour linked him to a widowers’ walking group and that this became ‘a very valuable monthly activity’. He ‘always looked forward to it’ and, ‘gradually, this helped [him] regain control of life’ (M2). At the same time, many participants described difficulties with others’ responses to their loss. A female participant emphasized: ‘It’s strange when people avoid you. They simply don’t know how to handle it’; she recalled being told: ‘Losing your husband is hard, but losing a child is worse’, asking, ‘Why would someone say something like that?’ (F9). Others felt a mismatch between how they appeared to others and how they really felt. One female participant recalled being told, ‘You’re so strong’, yet she felt very lonely. She spent days sitting in her garden, watching stars, hoping her husband could see her, and felt utterly miserable (F10). Another female participant similarly emphasized: ‘Everyone keeps telling me I’m “doing so well”, but I’m still far from being my old self. On the inside, I feel completely different’ (F11).
Professional support
Regarding professional support, several participants described encounters that felt supportive. One female participant felt seen by a home health-care professional during the Covid-19 pandemic. This professional regularly stayed for coffee: ‘After caring for my husband, she would stay …, and I could share my story. I truly felt seen and heard’ (F6). Similarly, following a hospital admission after his wife’s death, a male participant who had little desire to continue living, highlighted how conversations with staff ‘helped him a great deal’. As a co-researcher wrote: ‘They helped him see that he still had enough reasons to keep going’ (M4). At the same time, several participants emphasized the effect that negative experiences can have. During Covid-19, for example, staff had to enforce restrictions in health-care facilities. One male participant recalled being barred from visiting his wife in the nursing home: ‘I went to the director in anger … “Rules are rules” and he couldn’t make an exception’ (M3). Furthermore, after a partner’s death, gaps in aftercare were also noted. A female participant reflected on this: ‘I received little support … It would have been nice if there had been some aftercare … you fall into a deep void’ (F16). The quality of communication mattered as well. One female participant shared: ‘We were treated by an unfriendly oncologist … very skilled, but he lacked social skills’ (F11).
Experience of widowhood
The most frequent codes here were ongoing feelings of missing the partner, sadness and regularly thinking about the deceased partner. These depict how participants experience widowhood in everyday life.
Missing
Many participants still missed their partners to varying degrees. Some emphasized that missing could coexist with positive experiences. Seven years after her husband’s death, a female participant reflected, for example, how ‘Being alone at a party without him is confronting. But I notice that I’m open to meeting someone new and ready to embrace new things in life. I’m able to truly enjoy beautiful moments again; they now exist alongside the sorrow and the sense of loss’ (F7). For others, missing weighed heavily in daily life at different points in time. Two years after her husband’s death, one female participant feels his absence acutely: ‘I miss my buddy, my partner, with whom I dared to be completely myself … not even a hundred people can replace my husband’. She asked, ‘Where is the spark in life? You always come home alone. I’m running on reserves; the missing is getting more intense!’ (F19). Another female participant also expressed how she misses her husband in a specific way, nine years after his death: ‘I still miss him, the way he would cycle [towards their home] and wave at me’ (F18). Furthermore, as reported by a co-researcher, one male participant became visibly emotional when speaking about ‘the love of his life’ during the interview, three years after her death, and suggested a coffee break (M7). Additionally, four years after the death of his wife, another male participant talked about how her death still affects him every day and that he no longer enjoys anything on his own; everything at home reminds him of her. A co-researcher noted: ‘Every week, he buys a bouquet of red roses because she loved them so much’. Asked whether he could handle the loss of his wife and child who had also passed away, he answered, ‘No and actually, still not … life no longer has any meaning’.
Sadness
Many participants described sadness that still arises, sometimes unexpectedly, even years after the loss. Two and a half years on, for a male participant, selling his home triggered sadness: ‘Grief sits very deep in your soul’, he said, ‘believing it will always be there’ (M10). He recalled that when the real-estate agent moved his wife’s photo and drinking cup, he apparently reacted with an angry look; the startled agent quickly put the items back in place. Three years on, another male participant noted: ‘The loss no longer has a negative impact on daily functioning. But at unexpected moments, it suddenly comes to the forefront again’ (M9). Additionally, four years after her husband’s death, a female participant was still surprised by sudden waves of emotion as noted by a co-researcher: ‘She now often finds herself suddenly overwhelmed by sadness … She feels alone and simply misses him deeply’ (F21).
Thinking often about the partner
Participants often described thinking about their deceased partner, with effects that ranged from minimal disruption to profoundly positive change. Some, for example, framed these continuing thoughts as life-affirming. As noted by a co-researcher, one female participant reflected upon how ‘the love for life has actually been strengthened by the loss …. By now embracing life, the searing loss and pain have given way to acceptance and understanding. She has a good life, a loving new partner, and a meaningful job. She feels privileged’ (F17). Four years after his wife’s death, another male participant described recurring thoughts as liberating. As noted by a co-researcher, when thinking of the loss, this participant remembers how he ‘has discovered a different, beautiful side of himself’ and how he ‘especially wants to celebrate life’ (M6). His relationship with his children has also improved. He shared how he ‘now dares to tell [his children] that he loves them’. Furthermore, three years on, another male participant noted that, even though his wife is ‘in his thoughts every day’, it ‘hardly affects his daily life anymore’, adding: ‘In the hallway, there’s still a large photo of her, along with the funeral card … “Just put me in the hallway,” she said’ (M5). Furthermore, following a female participant’s description of ‘ongoing turmoil’ 11 years after her partner’s death, a co-researcher noted: ‘She is still thinking about the loss of her partner. It comes back every day, along with the thought: “It isn’t true, he’s still alive”’ (F12).
What helps or does not help now
The most frequent codes here were daily activities, network and characteristics. These describe what participants currently see as helpful or unhelpful in widowhood.
Daily activities
Daily activities were often described as influencing experiences of widowhood in different ways. Some participants framed activity as self-care and a way to turn outward. As a co-researcher wrote, one female participant emphasized how ‘She takes good care to avoid “withering away” … cycling 60 kilometres on her e-bike … walking … and, after some effort, she found a good place to start dancing again’ (F24). Drawing on her own experience, another female participant encouraged early re-engagement with activities outside the house: ‘Do something, take up volunteering, meet new people … Go to the theatre or a film. Try to find someone to go with you’ (F10). At the same time, limited activity could intensify dwelling on loss. A female participant noted: ‘Because I’m no longer able to do much physically … I’m confined to my home … I have a lot of time to dwell on [the loss of her husband and daughter]’ (F22). Disrupted routines also mattered. Another female participant missed, for example, the shared rhythms of going out, gardening and making food together as a co-researcher described: ‘They did everything together … After his passing, she still went out … But since she has moved house, that has also fallen away’ (F23).
Network
This code concerns wider social networks, excluding family and close friends, in navigating widowhood. With age, several participants described how ties thinned and were difficult to rebuild. One female participant noted: ‘As you get older, you lose people around you, but you don’t gain new ones’ (F21). Another female participant felt lonely and missed like-minded people, as a co-researcher recorded: ‘She still has conversations with her [late] husband … she has so much to share and would love to do so with like-minded people’ (F2). Moving house could further disrupt existing networks. After relocating, one female participant found that initial efforts to connect with people in her new neighbourhood ‘did not continue’ and regular contact gradually fell away (F23). Furthermore, a male participant tried an activity at a community centre but, finding ‘only women there’, did not return (M11). At the same time, social networks could provide practical help and pathways back into everyday life. A female participant described steady neighbour support with ‘small technical things like her walker, or groceries and such’ (F22). Another female participant, widowed at 87, gradually rejoined familiar circles: ‘After one or two years I went back to the choir. They kept asking me: when are you coming back? The library is [also] close by; they bring the books to me’ (F18). Online spaces also helped some to recalibrate perspective. One female participant followed a long-Covid chat group and ‘reminds herself that others have it even harder’, drawing strength from shared accounts (F21).
Characteristics
This code concerns personality traits that participants experienced as both helpful and, at times, limiting in later widowhood. One female participant, for example, emphasized her stoic, pragmatic stance: ‘I am quite down to earth … I then think, “I cannot do anything about it”’ (F26). Another female participant framed her self-reliant, forward-looking stance as helpful: ‘My motto is “fight, don’t complain, and keep going” … look at what you do have instead of complaining about what you don’t … Taking control yourself, that is my character’ (F25). She also recognized the downside of this: ‘My pitfall is daring to ask for help; I seem stronger than I am’. Similarly, one female participant drew energy from a positive outlook: ‘I have picked up my life again … I go out among people, commit myself to others … My positive attitude gives me the strength and energy to go on’ (F6). Yet she also emphasized a downside of this: ‘I find it difficult to ask others for help; I always try to solve everything myself first’. Other participants noted similar tensions. One female participant described herself as ‘a positive person … a doer’ (F2), while the co-researcher added: ‘That doesn’t stop her from sometimes wondering whether her life still has meaning … She feels more like a survivor than someone who truly lives’. Another female participant also highlighted perseverance and independence explaining: ‘[She is] a real doer and someone who perseveres … not wanting to be dependent’ (F21).
Discussion
The findings illustrate how the circumstances of a partner’s death (sudden or expected) and experiences of care-giving shaped the grieving process in diverse ways for many participants. Persistent sadness, longing and thoughts of the deceased partner continued throughout widowhood to varying degrees. Individual characteristics, daily activities and social and professional support networks were recognized as both helpful and unhelpful over time. Drawing on Clark’s (Reference Clark2021b) critique of resilience as ‘bouncing forward’, the findings deepen understanding of resilience and spousal loss in everyday life, as follows.
Firstly, Clark (Reference Clark2021b) notes that ‘bouncing forward’ can carry positive connotations that obscure the severity of adversity, especially when survival is central. Participants’ trajectories show uneven movement rather than uniformly positive change, at times centring mainly on survival. Expected deaths were recalled as both ‘bewildering and beautiful’ (F3) and ‘chaotic and scarring’ (F1). Sudden deaths were experienced more as shock than as a clear high or low; one participant, for instance, likened the prognosis of imminent death to a bolt of lightning from a clear sky (F4). Trajectories may therefore pass through both difficult and more favourable phases, at times centring on survival or reorientation. The ‘bounce’ in ‘bouncing forward’ can be read as movement between highs, lows and pauses; the ‘forward’ can be read as movement through these experiences over time, rather than towards uniformly positive change.
Secondly, Clark (Reference Clark2021b) warns that ‘bouncing forward’ can imply linearity, yet the findings show that resilience after spousal loss is experienced as non-linear. Participants described periods of steadiness punctuated by resurgences of grief. One participant, for example, felt that life had settled ‘until, at unexpected moments, it suddenly comes to the forefront again’ (M9). Another recalled being ‘suddenly overwhelmed by sadness’ years later (F21). Specific triggers and growing older also mattered: selling his home reopened grief for one participant (M10), and another noted how reduced mobility gave her more time to dwell on her loss (F22). These patterns complicate linear understandings of post-loss resilience, portraying trajectories as dynamic and context-dependent, in line with contemporary models of resilience and grief (Stroebe and Schut Reference Stroebe and Schut2010; Fisher et al. Reference Fisher, Ragsdale and Fisher2019; Métais et al. Reference Métais, Burel, Gillham, Tarquinio and Martin-Krumm2022; Guldin and Leget Reference Guldin and Leget2024). These findings contribute to discussions of resilience as ‘bouncing forward’ by decoupling ‘forward’ from steady progress, reading it instead as movement through alternating phases of stability, grief and reorientation over time, with recurrent returns to feelings of loss.
Finally, Clark (Reference Clark2021b) highlights that ‘bouncing forward’ can suggest uniform improvement across life domains, even though gains and constraints often coexist. The findings illustrate this coexistence clearly. One participant described herself as a positive ‘doer’, which helped in widowhood, yet she sometimes wonders whether life still has meaning and feels more like a survivor than someone who truly lives (F2). Similar tensions appeared in one participant’s difficulty in asking for help despite a self-reliant stance (F25), and another’s move to a better house meant losing a familiar network (F23). These cases show that improvements in some domains can occur alongside persistent constraints in others, consistent with views of resilience as multi-dimensional and uneven (Lenette et al. Reference Lenette, Brough and Cox2013; Pruchno et al. Reference Pruchno, Heid and Genderson2015; Infurna and Luthar Reference Infurna and Luthar2017). ‘Bouncing forward’ need not therefore describe a single, unified bounce ‘up’ or ‘down’ for the whole person. Different domains reach different ‘highs’ and ‘lows’ at different times, with some becoming more manageable while others remain marked by difficulty or constraint. The metaphor can therefore foreground uneven, domain-specific patterns of change rather than global improvement. The findings do not present these domain-specific fluctuations as new but demonstrate how they play out in later-life spousal loss and how they map onto ‘bouncing forward’.
Taken together, these points show that the metaphor of ‘bouncing forward’ can accommodate ambiguities, continuities and fluctuations in participants’ lives and describe how they reconfigure everyday life after loss. At the same time, as Clark (Reference Clark2021a, Reference Clark2021b) cautions, the metaphor tends to foreground individual change and can understate relational and multi-level processes. Examining the findings further through the theoretical lenses of Clark’s holonic framework and Arendt’s concepts of action and natality (Arendt Reference Arendt2018) deepens understanding of the relational, material and institutional processes at play.
Clark (Reference Clark2021b), for example, conceptualizes resilience as a dialectical process of expansion and contraction in possibilities for action, unfolding within nested holons (the individual, close relationships, wider networks and services). A central element in Clark’s (Reference Clark2021b) framework is the fluidity of movement between holons: people can shift attention, action and sources of support across levels. Under stress, this movement may become less fluid as individuals turn inward or contexts restrict or exclude. Contraction is not inherently positive or negative but context dependent; in some situations, it functions as a protective narrowing that allows adversity to be absorbed and worked through, enabling new outcomes (Clark Reference Clark2021b, p. 573). By contrast, easier movement across holons aligns with expansion and wider possibilities for action.
The findings add to Clark’s (Reference Clark2021b) framework by identifying periods of relative stillness within particular holons, when possibilities for action neither clearly widen nor narrow but appear suspended. One co-researcher described this as ‘no man’s land’, a time of stillness to face emotions and work out the next move. For example, one female participant recalled spending hours in the garden at night staring at the stars, feeling miserable, before later encouraging others to resume activities (F10). Stillness also appeared in social holons. After moving house, a female participant described how her initial efforts to connect ‘did not continue’ and contact with neighbours fell away, leaving this social holon in suspension (F23). These findings suggest that stillness may function as a recognizable mode within resilience processes. ‘Bouncing forward’ can thus be understood as interconnected, multi-level movement in possibilities for action across holons, which over time may expand, contract or pause unevenly within and between holons, echoing the highs, lows and pauses of a bounce.
Clark’s (Reference Clark2021b) framework maps how possibilities for action expand, contract or pause across interconnected domains. Arendt’s (Reference Arendt2018) account of action complements this by clarifying how such shifts occur in practice through recognition by others and through the material and institutional world in which action must appear. For instance, where Clark (Reference Clark2021b) theorizes resilience as multi-level movement in possibilities for action across holons, Arendt shows how these possibilities are situationally and relationally generated. According to Arendt, action is situated because words and deeds must enter a ‘space of appearance’ where others may recognize, ignore or push them back (Arendt Reference Arendt2018, pp. 198–199). This shared world comprises not only other people but also a durable realm of things and institutional arrangements that both relate and separate them, giving action a place to land (Arendt Reference Arendt2018, pp. 52–53). Read alongside the findings, this suggests that shifts in possibilities for action are shaped by how others’ responses are mediated through everyday artefacts, practices and institutions. For example, a male participant’s angry reaction when a real-estate agent moved his wife’s photo and drinking cup illustrates how recognition after loss can be mediated through ordinary objects (M10). Institutional responses also mattered. During Covid-19, one male participant was barred from visiting his wife in the nursing home and was told by the director that ‘rules are rules’, showing how services can foreclose possibilities for action (M3). By contrast, a female participant felt ‘seen and heard’ when a home health-care worker sensed her loneliness and stayed for coffee, illustrating how professional recognition within services can widen possibilities for action (F6).
Finally, Arendt’s notion of natality (the human capacity to begin anew) helps interpret how new beginnings arise within shifting configurations of recognition and possibilities for action (Arendt Reference Arendt2018, pp. 8–9, 176–177). It frames resilience neither as recovery nor linear progress but as the emergence of new forms of action and involvement in everyday relational life after spousal loss. One female participant, for example, began cycling and dancing again, gradually regaining rhythm and confidence in daily routines (F24). A male participant spoke of ‘celebrating life’ and daring to tell his children for the first time that he loves them (M6). These examples show acts of renewal that take shape with others and through encounters with activities, places and objects that open space for fresh starts. The findings emphasize the importance of this by highlighting the influence of social networks, excluding family and close friends.
At the same time, such renewal is fragile and context dependent. It may falter when recognition or support is lacking, as in one participant’s account of falling into ‘a deep void’ after insufficient aftercare (F16). Alternatively, recognition be actively sought despite uncertainty, as in a female participant’s choice to approach an acquaintance herself and reach out even when initially avoided (F14). These examples show that new beginnings depend on sustained, tentative action amid delayed or ambivalent recognition and that individuals can actively reopen room for action. Bringing natality into resilience research highlights resilience as a series of small, situated beginnings that may or may not take hold over time, and directs attention to the situational and relational patterns that enable or hinder them.
Methodological considerations
Firstly, regarding data collection, each interview report was co-produced by a participant and a co-researcher. Dialogue shaped the narrative, which was recorded by the co-researcher in a report. This added an interpretative layer beyond a first-person account. For PAR, this fostered shared ownership and empowerment but also created analytic challenges. A shared topic list supported structure and consistency, and collective analysis sessions helped distinguish co-researcher interpretations from participants’ views, strengthening reflexivity and transparency. Conducting the study across three municipalities enhanced credibility through triangulation, suggesting that patterns identified were not site-specific.
Secondly, the age threshold of 50+ may appear low. However, the inclusion of participants aged 50+ was deliberate to capture maximum variation (Palinkas et al. Reference Palinkas, Horwitz, Green, Wisdom, Duan and Hoagwood2015). This was chosen because our interest was precisely in capturing patterns that emerge across diverse contexts, given that life situations vary considerably not only across age groups but also within them. Rather than controlling for age-related differences, the broad age range allowed the study to explore how resilience following spousal loss develops and manifests across the diversity of experiences in later life. Our design therefore did not allow us to systematically disentangle these age-related differences, and future research could examine how resilience processes after spousal loss vary across specific phases of later life.
Thirdly, the role of co-researchers also warrants reflection. Unlike participatory studies that train co-researchers to code data (Schaefer and Bär Reference Schaefer and Bär2019; Perry et al. Reference Perry, Adams, Harland, Broadbridge, Giles, McGeechan, O’Donnell and Ramsay2021), this project prioritized group dialogue sessions. This created experiential spaces in which resilience could be explored conceptually while also being enacted in practice. Through dialogue, interview participants and co-researchers navigated and reflected on the relational and often challenging dynamics through which resilience takes shape in everyday life (Kralik et al. Reference Kralik, van Loon and Visentin2006; Abma et al. Reference Abma, Banks, Cook, Dias, Madsen, Springett and Wright2019b), deepening interpretation and grounding it in their lived experiences. This approach aligns with authors who call for careful consideration of how and whether to involve co-researchers in coding, given that it is time-intensive and risks becoming tokenistic rather than supporting meaningful participation (Nind Reference Nind2011; Locock et al. Reference Locock, Kirkpatrick, Brading, Sturmey, Cornwell, Churchill and Robert2019; Jackson et al. Reference Jackson, Pinnock, Liew, Horne, Ehrlich, Fulton, Worth, Sheikh and De Simoni2020).
Related to this, interviews were not transcribed, as this was deemed unnecessarily burdensome for the co-researchers; instead, they listened back to audio recordings when writing their interview reports to ensure accuracy and quality. Co-authorship was also considered but not pursued, as peer interviews and dialogue sessions were prioritized for their direct relevance to participants’ and co-researchers’ own lives. Co-authorship also exceeded the co-researchers’ initial roles and expectations, boundaries they repeatedly expressed appreciation for upholding, and risked introducing power imbalances given the level of English proficiency required for scientific publication. Furthermore, neither co-researchers nor interview participants were paid, to maintain equality between groups and preserve intrinsic motivation; co-researchers were, however, reimbursed for travel and material costs and received a small gift at the final session in recognition of their contribution.
Each of these decisions made regarding co-researcher involvement should not be regarded as setting a precedent for future work. Data analysis with co-researchers, co-authorship and payment each warrant careful consideration in light of a project’s specific goals and context. Several authors provide guidance on the ethical and practical implications of these decisions for participatory dynamics in PAR (Head Reference Head2009; Castleden et al. Reference Castleden, Morgan and Neimanis2010; Grady Reference Grady2019; Dedding et al. Reference Dedding, Goedhart, Broerse and Abma2021). Lessons learned from implementing PAR in this study are also explored in a separate publication (Beeris, Niemeijer and Machielse Reference Beeris, Niemeijer and Machielse2025b).
In terms of analysis, this article focuses on the interview data, strengthening traceability and accountability. The most frequent codes are prioritized, reflecting both the article’s low-inference qualitative descriptive approach (Sandelowski Reference Sandelowski1995/Reference Sandelowski2000) and the PAR commitment to foreground participants’ experiential expertise, preserving their words and avoiding further abstraction. However, frequency does not indicate significance. Less-common codes may capture marginalized or structurally important experiences that shape resilience differently. Limited inclusion may therefore understate rare but consequential trajectories. Future research could address this by focusing deliberately on under-represented experiences, for instance through single-case studies. Additionally, the analytic focus of this article is resilience processes in the context of spousal loss. It does not aim to cover detailed theory on grief and loss. For a complementary theoretical account, the integrated process model may be of interest (Guldin and Leget Reference Guldin and Leget2024).
Finally, participants’ characteristics also shaped the findings. Men were in the minority among both co-researchers and interviewees. Several authors have, however, identified important differences between female and male perspectives, showing how the experiences of each group and the social meaning of widowhood for each gender significantly shaped participants’ experiences and understanding. For examples, see Beeris, Niemeijer, Bisseling et al. (Reference Beeris, Niemeijer, Bisseling and Machielse2025a), Bennett et al. (Reference Bennett, Hughes and Smith2003) and David and Ali (Reference David and Ali2024). Many participants also experienced their partner’s death during the Covid-19 pandemic, an exceptional period when health and social care capacities were under pressure, which is important when interpreting accounts of professional support. The examples presented are not intended as a general critique of services; rather, they illustrate how small acts of recognition, or their absence, can have lasting effects on experiences of loss and resilience. Several co-researchers and participants were also living with illness, hardship or limited networks, broadening the understanding of resilience beyond relatively stable circumstances and adding nuance and ecological validity.
Conclusion
This PAR study examined how older adults make sense of resilience after spousal loss through a co-designed process foregrounding their experiential expertise. Findings follow a low-inference qualitative descriptive approach, staying close to co-researchers’ and participants’ own words. They depict resilience as a non-linear, multi-dimensional process in which grief and adaptation unfold together across life domains. The findings illustrate how the metaphor of ‘bouncing forward’ can be read as a movement through alternating highs, lows and pauses, unfolding non-linearly and involving uneven gains and losses across domains. Clark’s holonic framework of expansion and contraction, together with Arendt’s ideas of action and natality, help specify the relational, material and institutional conditions shaping these trajectories. For research and practice, recognizing periods of expansion (taking up activities, reaching out, relocating) and contraction or pause (withdrawal, ‘no man’s land’) may help tailor support to each phase. Future studies could investigate how older adults describe expansions, contractions and pauses in possibilities for action, and when new beginnings take hold or falter after spousal loss.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0144686X26100944
Acknowledgements
The authors wish to extend their gratitude to the co-researchers and research participants who generously contributed to this study. Additionally, we express our appreciation to the organizations that were instrumental in facilitating our connection with the co-researchers and research participants.
Author contributions
All authors contributed to developing the research outline and design. As the executive researcher, CB was responsible for contacting partner organizations, co-researchers and respondents, as well as gathering and analysing data. She consulted with AN, RB and AM throughout these steps via discussion rounds. CB also drafted the manuscript, incorporating ongoing feedback from AN, RB and AM. Consequently, all authors actively contributed to shaping the research, analysing the data and critically revising the manuscript. The final version of the manuscript was unanimously approved by all authors for publication.
Financial support
This research was supported by ZonMw, the Netherlands Organisation for Health Research and Development, project number: 05550032110010. The sponsor had no role in the design, execution, interpretation or writing of the study.
Competing interests
The authors have no conflicts of interest to report.
Ethical standards
Ethical approval for this study was obtained from the Ethics Committee of the University of Humanistic Studies, Utrecht, the Netherlands, dossier 2023-06.
Open access
