Participants were recruited between April and May 2023 in two settings of the Health Department of Distrito Federal, Brazil: the Hospital Regional de Ceilândia (HRC) and the Hospital de Apoio de Brasília (HAB). The HRC is a general hospital that provides care across various medical specialties, while the HAB specializes in neuromuscular rehabilitation and oncological and geriatric palliative care (Secretaria de Saúde do Distrito Federal, 2022a; 2022b).

Non-probabilistic, purposive sampling was employed, with the inclusion criterion being healthcare professionals who had been working for at least six months in one of the research settings. Professionals on leave for any reason were excluded. Selected participants were approached individually and directly, provided with a detailed explanation of the research, after which they gave their free and informed consent.

The data were collected through semi-structured interviews, beginning with socio-professional information. Subsequently, participants were presented with four guiding questions, constructed based on the A, B, C, and D of Dignity Conserving Care (Chochinov, 2007). The interviews were conducted individually and in person. The responses were audio-recorded and later transcribed in full.

The quantitative data and responses to the guiding questions were subjected to statistical analysis and thematic analysis, using Bardin’s (2011) approach. Descriptive statistical analyses and inferential tests were conducted using the R programming language. The qualitative data were processed using the IRaMuTeQ (Interface de R pour les Analyses Multidimensionnelles de Textes et de Questionnaires) software, version 0.7 alpha 2. For this purpose, the audio recordings of the interviews were transcribed using Microsoft Word 2013® software. The resulting material was consolidated into a file, with each interview separated by a command line. Each command line contained six variables considered relevant to the study design: age, sex, marital status, workplace, academic background, and experience in palliative care.

The embedding technique (Creswell & Plano Clark, 2014) was employed to combine the qualitative and quantitative datasets, using contiguous presentation of the results in separate sections to facilitate further analyses and interpretations (Fetters et al., 2013).

The research was approved by the Research Ethics Committee of the Universidade Católica de Brasília (protocol No 5.680.693) and the Research Ethics Committee of the Fundação de Ensino e Pesquisa em Ciências da Saúde do Distrito Federal (protocol No 5.862.376), in accordance with Resolution No 466/2012 of the National Health Council.

A total of 111 healthcare professionals were approached, of which 101 met the eligibility criteria. The majority were female (78.22%), married, and aged between 30 and 50 years. The average age of the group was 39 years, ranging from 21 to 67 years (Table 1).

Most participants were nurses, nursing technicians, physicians, and psychologists (70.29%). While nearly half of the healthcare professionals had specializations, only one (32.67%) had additional training in geriatrics or palliative care. More than half reported having experience in palliative care (Table 2).

The examination of temporal variables revealed that the average length of employment for participants was 6.92 years, while the average length of employment in the specific sector within the hospital was 5.83 years. The study found that participants obtained their academic degree, on average, 14.15 years ago and had 13.72 years of experience in their field of study. The majority reported having experience in palliative care (53.47%), with a range from 6 months to 17 years (x̅ = 6.01) (Table 3).

Chi-square and Fisher’s tests were conducted to examine associations between the terms used to define dignity and the socio-professional variables. A statistically significant association was found at the 10% significance level between the variable “hospital” and the word “comfort” (χ² = 3.6548; p-value = 0.05591), as evidenced by the post-hoc analysis of the standardized residuals table.

Similarly, for the variable “experience in palliative care”, a statistically significant association was observed at a significance level of 5% with the term “nurturing” (χ² = 3.9973; p-value = 0.04557), based on the post-hoc analysis of the standardized residuals table.

Binary logistic regression analyses were conducted to determine predictive variables for a selected binary dependent variable. The odds ratio (OR) was used to compare the probability of occurrence of this variable in two groups. The term “comfort” was examined as the dependent variable, while “hospital” and “sex” were independent variables. The results showed that affiliation with HRC led to the definition of dignity as “comfort” at a significance level of 5% (p-value = 0.0275).

Another dependent variable, “experience in palliative care”, was also examined, with “hospital” and “age group” as independent variables. The results indicated that being part of the HRC reduced the likelihood (OR = 0.1144; p < 0.01) of gaining experience in palliative care compared to affiliation with the HAB, at a significance level of 5%.

Regarding age, participants aged 40 to 50 years had a significantly higher probability of having experience in palliative care compared to other age groups (OR = 14.001; p < 0.01).

When categorizing the data, IRaMuTeQ allowed for the application of three analysis techniques: Descending Hierarchical Classification (DHC), Specificity Analysis, and Word Cloud.

The initial processing revealed that the corpus consisted of 101 texts, from which 56,718 words emerged. Of these, 2,546 were distinct words, and 1,032 occurred only once. The average number of words per text was 561.56.

The hapax coefficient – words or terms that occur only once in the corpus – was used to demonstrate data saturation (Lima et al., 2021). The examined corpus had a low number of these occurrences (3.51%), indicating sufficient sample size and consistency in responses to the guiding questions.

The DHC divided the corpus into 1,614 Text Segments (TS), of which 1,311 (81.23%) were utilized. These relevant TS were partitioned into three classes: (1) Work routines with terminally ill older adult patients, comprising 491 TS (37.45%); (2) Professional competencies, comprising 258 TS (19.68%); (3) Dignity at the end of life, comprising 562 TS (42.87%). In the end, a dendrogram was generated with these classes (Figure 1).

This material was examined using Bardin’s (2011), content analysis method to identify the meaning of the formed classes. These classes were then named and analyzed according to the distribution and association of the words included in them.

This class comprised 37.45% (f = 491 ST) of the analyzed corpus and consisted of words and roots within the range of χ² = 3.85 (case) and χ² = 24.77 (stay). It included terms such as “talk” (χ² = 48.16); “time” (χ² = 32.87); “explain” (χ² = 29.99); “converse” (χ² = 24.49); “alone” (χ² = 22.77), and “companion” (χ² = 21.93).

This class presented content related to the challenges faced by healthcare professionals in hospital care for terminally ill older adult patients. Participants expressed their concerns about family support and providing ideal care to patients. They also highlighted the emotional challenges at work, communication with families, and respecting patients’ privacy.

This class involved 19.68% (f = 258 ST) of the analyzed corpus, composed of words and roots within the range of χ² = 3.86 (listen) and χ² = 120.08 (team). It included terms and expressions such as “palliative care” (χ² = 73.05); “communication” (χ² = 63.74); “humanize” (χ² = 66.67); “knowledge” (χ² = 11.87); “interdisciplinary” (χ² = 4.2).

The content of this class highlighted the importance of training in palliative care and communication between teams, emphasizing the gap in understanding principles of palliative care between specialized and routine teams, often lacking effective communication between them.

This class encompassed 42.87% (f = 562 ST) of the analyzed corpus, composed of words and roots within the range of χ² = 4.01 (biography) and χ² = 105.18 (dignity). It included terms such as “care” (χ² = 52.72); “treat” (χ² = 38.94); “comfort” (χ² = 37.1); “empathy” (χ² = 25.17); “nurturing” (χ² = 24.05); and “respect” (χ² = 23.94).

This class revealed content related to the operationalization of care aimed at preserving the dignity of terminally ill older adult patients, including participants’ beliefs, behaviors, and feelings in delivering this care.

The Specificity Analysis (SA) categorized the corpus based on the categorical variable academic background – which is related to the profession practiced – in order to compare the evocations of different professional categories. In SA, IRaMuTeQ assigns a score to each selected TS, with higher scores indicating greater relevance within the variable under study (Sousa, 2021).

Nurses emphasized the implications of each clinical scenario and highlighted the value of professionalism and empathetic communication, prioritizing patient and family preferences (e.g., “here”, “care”, “provide”, “care for”, and “value”).

Pharmacists highlighted the importance of recognizing patients’ spiritual beliefs but acknowledged the lack of training in this area. Healthcare professionals use religion and spirituality as coping strategies when dealing with individuals near death (e.g., “religion”, “sacred”, “autonomy”, “respect”, and “value”).

Physical therapists emphasized providing care that prioritizes comfort and alleviation of suffering in terminally ill older patients, recognizing the challenges of effective communication in this context. Considering patient preferences, along with empathy, compassion, and teamwork, is fundamental (e.g., “symptom”, “suffering”, “team”, and “compassion”).

Speech therapists emphasized the role of specialized knowledge, active listening, and respectful care. The focus is on ensuring a peaceful and pain-free end of life for terminally ill older patients, while maintaining clear communication with their families. Dignity at the end of life is promoted by providing empathetic and personalized care (e.g., “rights”, “intervention”, “choose”, “conversation”, and “respect”).

Physicians emphasized the importance of preparing for the loss of loved ones. They also highlighted the need for emotional support. When caring for older patients in their final stages of life, they stressed the importance of effective interaction, simple language, and non-intrusive approaches. Empathy, communication, and compassion were seen as essential in end-of-life care (e.g., “father”, “medicine”, “avoid”, “invasive”, and “explain”).

Nutritionists recognized the significance of religion as a coping tool and the importance of respecting patients’ spiritual beliefs. For them, dignity-conserving end-of-life care demands compassion, attention, and addressing the concerns of both the patient and family. This includes ensuring adequate nutrition and honoring the patient’s wishes and values (e.g., “religion”, “Catholic”, “empathy”, and “family”).

Dentists emphasized providing individualized care, taking into account the beliefs, values, and aspirations of terminally ill older patients. Listening, understanding, respect, and family involvement are essential. Multidisciplinary teams and palliative care committees facilitate this care (e.g., “happy”, “respect”, “value”, “will”, and “attend to”).

Psychologists emphasized the importance of understanding patients’ wishes and needs, especially in pain management. Person-centered care is a priority, and validating patients’ beliefs and values is crucial. They are also concerned with communication among healthcare professionals (e.g., “perspective”, “meaning”, “understand”, “belief”, and “communication”).

Social workers emphasized the role of communication in palliative care and the need to ensure patient autonomy. Empathy in dealing with patients and the development of communication skills through training on the subject are crucial (e.g., “communicate”, “person”, “team”, “understand”).

Nursing technicians acknowledged the complex needs of older patients and the necessity of providing empathetic care, taking their religious beliefs into account. The emotional impact of their work was highlighted, as well as the need to maintain a balance between personal and professional responsibilities (e.g., “affection”, “sad”, “assistance”, “feel”, “faith”, “comfort”, and “alone”).

Occupational therapists highlighted the importance of protecting patient dignity, respecting their values, and offering support to the caregiver. The need for early palliative care and attention to the spiritual dimension of care were also mentioned (e.g., “life”, “care”, “spiritual”, “diagnosis”, and “palliative care”).

The Word Cloud (WC) analysis identified that the terms “patient” (f = 880) and “family” (f = 357) were the most frequently mentioned. The WC visually represents the key elements of the corpus, with the size and centrality of the words being directly proportional to their frequency (Sousa, 2021). Figure 2 represents the WC corresponding to the examined corpus.