Introduction
Dignity is among the most invoked yet least operationalized values in end-of-life care. It is widely recognized as intrinsic to the human condition, yet its meaning is plural and context-dependent, shaped by individual narratives, institutional practices, and the quality of the relationships between patients, caregivers, and health professionals (Chochinov et al. Reference Chochinov, Hack and McClement2002; Pleschberger Reference Pleschberger2007; Leget Reference Leget2013). For people approaching death, feeling dignified typically means continuing to be recognized as a person, preserving the coherence of one’s own life story, and maintaining a sense of belonging and identity continuity even amid physical and existential vulnerability (Chochinov Reference Chochinov2006; Sinclair et al. Reference Sinclair, Beamer and Hack2017; Juliao et al. Reference Juliao, Courelas and Costa2018).
Chochinov’s Dignity Therapy opened a decisive path for integrating existential dimensions into palliative care, but its focus remains largely psychological and individual (Chochinov et al. Reference Chochinov, Kristjanson and Breitbart2011). It offers limited articulation of the biographical time, institutional context, and reciprocal human encounter that, we argue, sustain the experience of dignity in practice. DiRePal – Relational–Temporal Dignity in Palliative Care – was proposed to address this gap, suggesting that dignity emerges at the intersection of lived time (biographical, clinical, and symbolic) and relation (between persons, contexts, and systems): a dynamic process of recognition and narrative reconstruction rather than a fixed possession of the individual (García Abejas et al. Reference García Abejas, Santos and Mota-Filipe2025; García Abejas Reference García Abejas2026).
This theoretical architecture – anchored in Ricoeur’s narrative identity, Levinas’s ethics of alterity, Nussbaum’s capabilities approach, Tronto’s ethics of care, and Aristotelian phronesis – has so far been developed and defended philosophically (García Abejas et al. Reference García Abejas, Santos and Mota-Filipe2025; García Abejas Reference García Abejas2026). What has been missing is a systematic empirical account, gathered directly from the people whose dignity, or its absence, the framework attempts to describe: patients, caregivers, and the physicians who share the clinical encounter with them. This paper reports such an account, offered not as confirmatory validation of a pre-established model but as the empirical fieldwork from which its core categories were, in part, inductively informed and against which they can now be examined.
The present study asked: How is dignity at the end of life understood and lived by patients, caregivers, and physicians? Which relational, narrative, and temporal factors contribute to its preservation or compromise? How do organizational and digital transformations interfere with the experience of dignity? And what ethical and practical implications follow for clinical practice and professional formation in palliative and oncological care? By examining the 3 perspectives simultaneously, this study aims to offer an integrated, empirically grounded account of dignity that can inform both the philosophical and the clinically applicable dimensions of the proposed framework.
Methods
Design
This qualitative study followed a constructivist–relational paradigm, centered on the meanings attributed to dignity at the end of life from the interconnected perspectives of patients, caregivers, and physicians. The study was conducted within a doctoral research program underpinning the DiRePal model, with methodological collaboration from IQVIA Portugal, which supported the technical and operational design of the qualitative fieldwork (IQVIA Portugal 2025a, 2025b).
Participants and procedure
Six focus groups (FGs) were conducted between April and November 2025: 2 with oncology patients (10 participants: 4 in FG1 and 6 in FG2), 2 with informal caregivers of oncology patients (14 participants: 8 in FG1 and 6 in FG2), and 2 with pulmonology/oncology physicians (15 participants: 7–8 per group). Given the high mortality and vulnerability of patients with advanced lung cancer, the patient sample also included a small number of patients with other tumors metastatic to the lung to preserve clinical comparability and experiential validity.
Inclusion criteria were: for patients, an incurable lung cancer diagnosis (or other pulmonary-metastatic tumor), preserved communication capacity, informed consent, and absence of severe cognitive decline; for caregivers, direct experience accompanying a family member or close person in palliative care, with emotional and cognitive availability to share their experience; for physicians, more than 1 year of clinical experience in palliative care or respiratory oncology, recruited from different regions and institutional contexts (public and private hospitals). Sampling was purposive, ensuring geographic, professional, and socioeconomic diversity, and was complemented by theoretical sampling as new categories emerged during coding. The main sociodemographic and professional characteristics of the participants are summarized in Table 1.
Sociodemographic and professional characteristics of the study participants (N = 39)
Table 1 Long description
Participant group, n, and characteristics are listed for patients, informal caregivers, and physicians. Patients numbered 10, with mean age 59.0 years, standard deviation 11.3, and age range 41 to 79; sex was 9 women and 1 man. Informal caregivers numbered 14, with mean age 51.0 years, standard deviation 10.3, and age range 30 to 66; sex was 10 women and 4 men. Caregiver relationship to the patient included spouse or partner 6, parent 5, grandmother 1, cousin 1, and friend 1. Physicians numbered 15, with specialties pulmonology 12 and oncology 3. Physician clinical experience had a mean of 8.5 years, standard deviation 9.6, and range 1 to 27 among 14 physicians with an exact duration available; one physician was in specialty training, postgraduate year 1 to 5.
SD = standard deviation.
Sessions lasted approximately 90 minutes, were held on a secure digital platform with professional moderation and scientific supervision, and followed a semi-structured guide covering 6 axes: the personal and professional meaning of dignity; concrete experiences of dignity preservation or loss; the importance of life history and narrative; perception of time and decline; the role of the patient–caregiver–physician relationship; and the impact of digital technologies (patient-reported outcome measures (PROMs), patient-reported experience measures (PREMs), and artificial intelligence (AI)). All sessions were audio/video-recorded with participants’ authorization and transcribed verbatim.
Analysis
Transcripts were anonymized and analyzed following reflexive thematic analysis (Braun and Clarke Reference Braun and Clarke2019), supported by NVivo 14 and ATLAS.ti. Coding proceeded in 2 connected stages: first, an inductive, data-driven familiarization and open coding of the full corpus by 2 researchers working independently, generating an initial set of descriptive codes; second, an interpretive, more deductively informed stage in which these codes were organized into candidate categories and refined into the final themes reported below, through iterative comparison across the 3 participant groups. Rather than being treated as separate methods, this 2-stage process is presented here as a single reflexive analytic pathway, consistent with Braun and Clarke’s view of theme development as an actively interpretive, researcher-shaped process rather than a mechanical or purely emergent one. Divergent readings between the 2 coders were resolved through discussion and consensus rather than through a formal inter-rater reliability statistic, which is not required by, and would sit uneasily with, a reflexive thematic analysis approach.
Rigor and ethics
Methodological quality was addressed through the trustworthiness criteria proposed by Lincoln and Guba (Reference Lincoln and Guba1986) and operationalized for qualitative health research by Amin et al. (Reference Amin, Nørgaard and Cavaco2020): credibility, through triangulation of data, researchers, and sources (physicians, caregivers, and patients) and through peer debriefing between the 2 coders and cross-validation of emerging categories; dependability, through a systematic audit trail documenting analytic decisions; transferability, through thick description of the clinical, organizational, and interpersonal context (Denzin 2001; Ponterotto Reference Ponterotto2006); and confirmability, reinforced by researcher reflexivity and independent external supervision. Member checking – returning findings to participants for confirmation – was deliberately not undertaken. Given the advanced illness, emotional vulnerability, and, in several cases, limited residual life expectancy of the patient and caregiver samples, re-exposing participants to their own transcribed accounts of suffering, or requesting a further round of engagement, was judged to carry a disproportionate burden relative to its expected contribution to credibility; this decision is consistent with published concerns about the feasibility and appropriateness of member checking in vulnerable populations (Amin et al. Reference Amin, Nørgaard and Cavaco2020). Its absence is stated here explicitly, as a limitation rather than as an oversight, and credibility instead rested on the multiple triangulation and audit strategies described above. The study additionally integrated the authenticity criteria of fairness, ontological, educative, catalytic, and tactical authenticity (Lincoln and Guba Reference Lincoln and Guba1986; Amin et al. Reference Amin, Nørgaard and Cavaco2020). The principal investigator maintained a reflexive diary throughout (Finlay and Gough Reference Finlay and Gough2008).
Results
Six interconnected themes emerged: personal understanding of dignity, narrative flexibility, negotiating technical rationality and relational care, temporal sensitivity, professional reciprocity, and digital and structural integration. Rather than describing patients, caregivers, and physicians as 3 separate blocks, each theme below is presented as a single dialogical phenomenon seen from 3 vantage points, closing with a short interpretive synthesis. Quotations were translated from the original Portuguese into English independently by 3 researchers, with discrepancies resolved by consensus discussion to preserve idiomatic and emotional register as closely as possible; FG and group identity are indicated for each.
Personal understanding of dignity
Across all 3 groups, dignity converged on a single axis – being recognized as a person rather than reduced to a diagnosis – even though each group approached that axis from a different position of vulnerability. For patients, this meant respect and physical/psychological integrity above all; illness trajectories varied by posture (optimist, resigned, apathetic, revolted, or fragile), but all rejected being treated differently because they were ill, repeatedly citing the diminutive “coitadinho” (“poor little thing”) as language that collapses identity into diagnosis:
For me, basically and fundamentally, it’s about respecting my integrity, both physical and psychological. It’s not treating me with discrimination — you know, ‘oh, poor thing, she had that problem, she’s a poor thing.’ (FG2, Patient)
Physicians largely agreed in principle, structuring dignity around respect for the patient’s life history, identity, and values:
Always remember we are in the presence of a human being, a person who has a particular experience, a life, loved ones they share it with — and to keep that as the focus… The illness is just one more experience in that person’s life. (FG2, Physician)
Yet physicians also acknowledged that, in daily practice, this principle tends to collapse into autonomy, comfort, and the pursuit of a “tranquil death,” confirming dignity as “difficult to operationalize” and locating its threats mainly outside their own practice – in the hospital system, or in paternalistic attitudes attributed to others rather than to themselves.
Caregivers described the demands of the role – silent suffering, concealed emotion, exhaustion – as a route to a different kind of erasure: even when directly asked about their own dignity, the conversation consistently redirected to the patient.
It’s very hard, because we must be very cold, we have another side that’s suffering. And knowing there’s no… much more hope, we must keep our composure and not let ourselves cry. (FG1, Caregivers)
Caregivers wished to be recognized as part of the care process, noting, paradoxically, that the person who most worried about their dignity was often the patient, while professionals, family, and friends assumed they were “only doing their duty.” Taken together, the 3 groups articulate the same underlying claim from 3 unequal positions: dignity is inseparable from being seen as a whole person, but each group is differently exposed to the risk of not being seen – patients by depersonalizing care, caregivers by structural invisibility, and physicians by their own difficulty in translating a value they endorse into daily practice.
Narrative flexibility
Illness reorganized priorities and relationships for patients and caregivers alike: minor problems were relativized, personal goals gained urgency, and a kind of “natural selection” made visible who truly offered support.
It’s like being born again… I was terrified of flying. When I got sick, I thought I’ve missed out on so much because I was afraid, and now this awful illness comes and undoes me. So, I started taking more risks than I used to, stepping outside my comfort zone. (FG2, Patient)
Writing about this transformation was uncommon and sharing it with a physician rarer still – only 1 patient reported having done so, and not with her primary physician. Physicians confirmed this distance directly, attributing it to time rather than to disinterest:
We really don’t have time, day to day we don’t. (…) Sometimes just five minutes of listening is all that’s needed. (FG2, Physician)
I personally try to escape the situation quickly and give objective information (…), because it’s uncomfortable. (FG2, Physician)
Physicians tended to delegate narrative work to palliative care teams and to family, defending emotional distance as a way of protecting clinical objectivity – and, implicitly, of protecting themselves. Caregivers, meanwhile, described a parallel narrowing: concealing their own feelings behind an “apparent positivity” that did not match their inner experience, and feeling invisible to physicians, other professionals, and sometimes family.
I became colder — the way I relate to people changed. I developed this way of hiding my feelings. (FG2, Caregivers)
What emerges is a chain of mutual withdrawal: patients rarely offer their transformed narrative because they do not see physicians as the right interlocutor; physicians rarely ask because they lack time and feel unequipped; caregivers absorb the resulting silence by suppressing their own account altogether. A simple “How are you?” on the rare occasions it was offered was enough to interrupt this chain and register as meaningful.
Negotiating technical rationality and relational care
This theme captured a recurring tension between technical reasoning and ethical presence, most concretely mapped by patients as a contrast between depersonalizing and dignifying moments. Depersonalizing: a diagnosis received at home through a test result with no medical accompaniment or communicated to a family member instead of the patient; 10-minute consultations with no room for feelings; physicians avoiding eye contact, absorbed in the screen; being discussed clinically in front of interns as though absent; being switched between physicians’ mid-course, experienced as abandonment. Dignifying: a physician asking how the patient felt; a consultation that opened with the person rather than the disease; small, shared stories that created proximity; a phone call home simply to ask how treatment had gone.
Physicians described the same tension from the inside, framing emotional distance not as indifference but as a form of care under constraint:
We are paternalistic because we continue to worry about the patient, while respecting their choices. (FG1, Physician)
We try to keep the patient collaborative, without affecting them negatively. (FG1, Physician)
This rationalization – protecting objectivity to protect the patient – was experienced, physicians admitted, with an undertone of guilt. Caregivers occupied the middle ground, appreciating professional competence while identifying the “coldness of distance” as one of the principal threats to their own dignity; analysis distinguished 3 caregiver postures toward this rationality – the invisible, who withdraw; the involved, who feel integrated; and the assertive, who push for inclusion despite resistance. None sought to displace the patient as protagonist; all asked, in different registers, to be let into the room.
Temporal sensitivity
Dignity’s temporality was the theme participants found hardest to articulate directly, yet each group located it differently along the illness trajectory. Patients placed their greatest sense of dignity in the acceptance of finitude and serenity before loss; caregivers described dignity as something that could “grow over time” through open communication and continuity of care. Physicians reported that their own understanding of dignity had evolved across their careers, from a disease-centered to a more person-centered view – yet, in the moment, many still equated dignities most directly with physical fragility itself:
In practice, I think it [dignity] is put at stake precisely as the patient becomes more fragile. The more fragile, the more their dignity is at risk. (FG1, Physician)
The contrast is instructive: patients and caregivers describe dignity gaining depth over time through relationship, while physicians most readily notice its erosion at moments of acute physical decline – a difference in where, in the arc of illness, each group’s attention to dignity is triggered.
Professional reciprocity
This proved the densest theme. Patients and caregivers converged in recognizing that oncology professionals face intense strain from the disease’s aggressiveness and frequently tragic outcomes, compounded by overburdened services – and largely excused the resulting distance, even while experiencing it as a loss of accompaniment:
I was talking to a doctor who works at the hospital, and I asked her: ‘Doctor, how do you live with the context of losing lives?’ And what she told me was that what was hard was the early years, because now it’s like eating a sandwich… because they do everything they can to help this person. (FG2, Patient)
Physicians confirmed the picture from within, describing themselves as frequently vulnerable and under-equipped, with intense emotional overload and no institutional psychological support:
It takes time, it takes energy, it requires family support even for us, and we’re not always prepared to play that role. (FG2, Physician)
Our coldness ends up being a form of defence; it’s more comfortable to be pragmatic. (FG1, Physician)
Physicians described relying on family and on a colder professional “armor” as protective strategies, in the near-total absence of systematic psychological support, reserved in practice for last-resort situations. Patients, for their part, valued brief but genuine moments of humanity – “just a look, a pause, a touch” – as sufficient to restore a sense of accompaniment. The reciprocity is direct: when physicians have no support for their own vulnerability, the armor they build to survive becomes the coldness that patients and caregivers experience as a loss of dignity.
Digital and structural integration
Patients and caregivers showed limited recognition of digital tools’ potential for communication and continuity of care, and considerable distrust of how testimonies might be used:
At the hospital I attend, they send an email after every appointment asking us to give feedback. Obviously, I never write anything I think or feel, because I’m afraid, they’ll treat me even worse if I say something. (FG2, Patient)
Some caregivers, conversely, valued the idea of an AI-based platform to clarify doubts without direct contact with professionals, particularly those already familiar with tools such as ChatGPT, though reliability concerns persisted. Physicians displayed a broadly parallel skepticism, but for different reasons – generalized unfamiliarity with PROMs and PREMs and doubt about their clinical applicability:
PROMs and PREMs are very fashionable, but in real clinical practice there’s no time for that. AI could help with the bureaucratic part, but I don’t know how it could take care of dignity. (FG1, Physician)
Across groups, the barrier was less technological than interpersonal: patients distrust who reads their words and why; physicians distrust whether they will ever have time to read them at all.
Discussion
Across all 6 themes, the empirical pattern is consistent with a broader international literature that has long identified dignity as relational rather than possessed by the individual – from Chochinov’s dignity-conserving care (Chochinov et al. Reference Chochinov, Hack and McClement2002, Reference Chochinov, Kristjanson and Hack2006) to Honneth’s theory of recognition (Honneth Reference Honneth1996), in which identity is constituted in the mirror of the other, and the phenomenological accounts of brief, dignifying encounters offered by Butler (Reference Butler2004) and Leget (Reference Leget2013). The caregivers’ near-total self-erasure documented here – their tendency to displace their own needs onto the patient while still craving recognition of their labor – sits squarely within Tronto’s ethics of care (Tronto Reference Tronto1993), which frames caregiving as an inherently relational practice organized around attentiveness, responsibility, competence, and responsiveness, each of which our participants described being denied to them as caregivers even while striving to enact it toward the patient. What this fieldwork adds to that literature is not a new claim that dignity is relational – that claim is well established – but a granular account of where, exactly, the relation succeeds or fails in everyday oncological care, and of the fact that this failure is symmetrical: patients, caregivers, and physicians each described themselves as, at different moments, insufficiently seen.
A first conceptual tension concerns technical rationality. The international literature on clinician detachment (Halpern Reference Halpern2001) and on vulnerability as a condition of openness to the other (Lévinas Reference Lévinas1972; Nussbaum Reference Nussbaum2001) tends to treat emotional distancing primarily as a deficit to be corrected. Our data complicate this reading: physicians did not describe distance as an absence of care but as its paradoxical expression – “we are paternalistic because we continue to worry about the patient.” These findings suggest that emotional distancing may be better understood not as a deficit to correct but as an imperfect form of prudential judgement – closer to Aristotelian phronesis (Aristotle 2009) than to indifference: not the elimination of clinical distance, but its calibration in response to the concrete other rather than to protocol.
A second tension concerns narrative. Chochinov’s model establishes that narrative coherence sustains dignity (Chochinov et al. Reference Chochinov, Hack and McClement2002, Reference Chochinov, Kristjanson and Hack2006); Pleschberger frames professionals’ discomfort with patients’ stories as an institutional rather than individual failure (Pleschberger Reference Pleschberger2007). Our findings extend this by showing a self-reinforcing cycle across all 3 groups – patients withholding their transformed narrative, physicians not asking, caregivers absorbing the resulting silence – that a purely individual account of narrative competence would not capture. These findings suggest that narrative is distributed across the caring triad rather than privately held; it survives only if someone – patient, caregiver, or clinician – is willing to carry it forward (Ricoeur Reference Ricoeur1992).
A third tension concerns time. Walter and Ricoeur describe grief and narrative as fundamentally temporal processes, yet our data show that patients, caregivers, and physicians inhabit markedly different temporalities within the same clinical episode – reconciliation, apprenticeship in loss, and scarcity, respectively (Ricoeur Reference Ricoeur1992; Walter Reference Walter1996). Institutional time, compressed and fragmented, collides with the existential time of dying; listening to the other’s time, rather than imposing one’s own, becomes itself an act of dignification. The professional reciprocity theme suggests why institutional time is compressed in the first place: physicians report acute emotional overload and near-total absence of institutional psychological support, and the coldness this produces circles back to compromise the listening that patients and caregivers need (Truog et al. Reference Truog, Campbell and Curtis2008; Puchalski et al. Reference Puchalski, Vitillo and Hull2014). Clinician and patient dignity are, on this evidence, interdependent rather than separately optimizable.
Finally, the digital and structural theme extends earlier work on AI and post-biographical dignity: patients and caregivers fear surveillance and the reduction of suffering to data, while physicians doubt clinical applicability more than they resist the technology itself (García Abejas et al. Reference García Abejas, Santos and Mota-Filipe2025; García Abejas Reference García Abejas2026). Read together with Floridi’s ethics of information, this suggests digital tools are not resisted uniformly, but for divergent reasons that a single design solution is unlikely to satisfy – an ethics of mediation must attend to trust as much as to usability (Floridi Reference Floridi2013).
Three implications follow. Clinically, listening warrants recognition as a therapeutic act with its own value, embedded in training rather than treated as a courtesy contingent on spare time (Chochinov et al. Reference Chochinov, Hack and McClement2002; Leget Reference Leget2013). Organizationally, consultation length and scheduling should be governed by the same prudential logic, so that institutional urgency does not routinely override the patient’s subjective time, and psychological support for clinicians should be built into standard practice rather than reserved for crises. Technologically, digital and AI-based tools should be designed to extend presence rather than substitute for it, with governance that returns data meaningfully to patients instead of merely extracting it. At the policy level, this argues for person-centered and narrative indicators (respect, empathy, and inclusion) alongside biomedical quality metrics, and for interdisciplinary formation that treats dignity as a shared clinical competence rather than an individual disposition (Garcia Abejas et al. Reference Garcia Abejas, Geraldes Santos and Leite Costa2025).
Limitations
Six FGs conducted over several months, with a bounded number of participants, constrain generalizability; patients in very advanced physical or cognitive decline could not be included, narrowing representativeness at the far end of the trajectory the study addresses. Self-report within a group setting may have shaped disclosure of sensitive material. A specific limitation concerns credibility strategy: member checking was deliberately not performed, given the advanced illness and emotional vulnerability of the patient and caregiver samples, but its absence removes a recognized safeguard against researcher misinterpretation. Credibility instead relied on triangulation across 3 participant groups, independently coded and reconciled through discussion, peer debriefing, and an audit trail (Amin et al. Reference Amin, Nørgaard and Cavaco2020) – strategies that mitigate, without fully replacing, the confirmatory function member checking would otherwise have served.
Conclusion
Across all groups, dignity was described not as an attribute residing solely within the dying person but as something sustained – or withheld – through relationships: through listening, through time shared rather than merely elapsed, and through a life story that someone else is willing to carry when the patient no longer can. These findings ground and inform, rather than confirm, a central claim of the proposed framework: that dignity is simultaneously relational and temporal. They also show that clinician dignity is not separable from patient dignity: physicians reporting moral distress and absent institutional support are structurally less able to offer the presence patients and caregivers need. Rather than protecting dignity as an abstract ethical principle, palliative care ultimately protects it through everyday acts of recognition, listening, and shared time.
Data availability statement
The interview guide and coding framework are available from the corresponding author upon reasonable request. To protect participant confidentiality, given the sensitivity of the topic and the vulnerability of the sample, full anonymized transcripts are not publicly available; de-identified excerpts beyond those quoted in this article may be shared with qualified researchers under a data-sharing agreement, subject to approval by the Ethics Committee of the Universidade da Beira Interior.
Acknowledgments
The authors sincerely thank the patients, family caregivers, and physicians who generously shared their experiences and perspectives during the focus groups. The authors also acknowledge IQVIA Portugal for providing methodological and operational support in the design and conduct of the qualitative fieldwork. During the preparation of this manuscript, the corresponding author used ChatGPT (OpenAI) for language refinement and editorial assistance. Generative AI was not used for data collection, coding, qualitative analysis, interpretation of findings, or generation of scientific conclusions. All AI-assisted content was critically reviewed and revised by the authors, who take full responsibility for the final content of the manuscript.
Author contributions
A.G.A. conceived and designed the study, led data collection and analysis, and drafted the manuscript. D.G.S. contributed to the philosophical framing and critically revised the manuscript. H.M.F. contributed to methodological design and critically revised the manuscript. F.L.C. and A.C.B. contributed to study conception and critically revised the manuscript. All authors approved the final version and agreed to be accountable for all aspects of the work.
Funding
This research was conducted within a doctoral research program at the Universidade da Beira Interior. The qualitative fieldwork received methodological and operational support from IQVIA Portugal. This research received no external funding.
Competing interests
The authors declare no conflicts of interest. IQVIA Portugal provided methodological and operational support for the qualitative fieldwork under a commercial service agreement. IQVIA had no role in the study design, data interpretation, manuscript preparation, or the decision to submit the manuscript for publication. The authors declare that this commercial relationship does not constitute a conflict of interest.
Ethical approval
This study was approved by the Ethics Committee of the Universidade da Beira Interior (reference CE-UBI-Pj-2025-042; favorable opinion issued 19 September 2025) and was conducted in accordance with the Declaration of Helsinki and the General Data Protection Regulation (EU) 2016/679.
Ethics and consents
All participants provided written and verbal information and digital informed consent prior to participation, with the right to withdraw at any time without consequence.
Open access
