Feasibility, acceptability and adaption of dignity therapy: a mixed methods study achieving 360° feedback

In accordance with the EORTC Translation Procedure [12], two English and two German native speakers conducted forward-backward translations of the DTQP. These translations were discussed at round tables by physicians and psychologists experienced in palliative care and, where necessary, checked back with H.M. Chochinov to obtain a pilot version.

In June and July 2015, two experienced researchers (S.S.M., qualified psychologist, and PhD S.G., sociologist) conducted focus groups with HCPs in meeting rooms at the university hospitals of M and W; the meetings lasted 60 and 70 min respectively. Both researchers were clinical and scientific colleagues of the participants from M; however, prior to study commencement, they were not acquainted with the HCP participants from W. Participating HCPs were all highly experienced in palliative care and were purposively recruited by e-mail. The number of participants represents the multi-professionalism of palliative care teams, consisting of five HCPs in M (psychologist, chaplain, nurse and two physicians) and seven HCPs in W (two chaplains, two nurses and three physicians). All participants were informed of the aim to achieve a German translation of the DTQP and they all gave written informed consent to participate. Each participant was attributed an ID to anonymize the qualitative data. As none of the 12 HCPs had previously conducted DT, their statements can be regarded as unbiased. Following a comprehensive interview guideline (developed by S.G. according to Willis’ recommendations on cognitive interviewing [13], Additional file 1), the HCPs were requested to express spontaneous associations regarding the title of the intervention, the wording of the DT questions, and the adaption of DT to the German cultural context. As well as visually summarizing and recording the participants’ statements on flip charts during the sessions, audio recordings were also made. S.S.M. and S.G. then transcribed and analysed the qualitative data based on the principles of qualitative content analysis as described by Kuckartz [14]. This method enables the elaboration of core messages from the interview transcriptions, by using inductive and deductive approaches in a systematic manner. Only parts of the interviews pertaining to the research questions have been considered, numbered, and arranged within an Excel sheet. Analysis units from the first interview transcript were paraphrased and generalized, and redundancies were identified, and inductively subsumed into main categories and subcategories. This resulting code structure was used to analyse the second interview deductively, and new codes were added inductively to the code structure, as needed. Typical phrases were identified. In the first instance both researchers coded independently. To achieve interrater reliability, categories were discussed and any coding conflicts were resolved. This process was repeated until the final structure of categories was formed and all relevant parts of the text were coded appropriately. The research group reviewed the results to ensure inter-subjectivity and coherence within the code structure. The results constituted the first step in the linguistic and cultural adaption of the DTQP.

Leaning on the experiences and data of the Danish DT feasibility study [10], we decided to include 30 patients from two PCUs in our study. Each patient admitted to the PCUs in M and W was attributed an ID and was screened for eligibility. Inclusion criteria were: having a terminal illness with a clinically estimated life expectancy of up to 12 months; age ≥ 18 years; ability to give informed consent; possibility to nominate a relative for the feedback questionnaire; and appropriate German language skills. Exclusion criteria were: having cognitive impairments or an estimated life expectancy below 2 weeks, especially where the dying process had begun. After giving written informed consent, patients participated in the audio-recorded DT interview led by DT therapists (M: psychologist and male nurse; W: psychologist) trained by H.M. Chochinov. All steps of the DT protocol were fulfilled (interview session, transcribing, editing, reading out to the patients, finalizing the document and taking it to the recipient). Detailed information about the duration of the DT interview and DT intervention was logged. Furthermore each question asked during the DT interview was counted, repetitions were listed, and rewording as well as patients’ noteworthy reactions were noted.

After the DT interview using the DTQP, the participants took part in semi-structured cognitive interviews with their therapists to report their experience of the implementation of the DT interview (the interview guideline was produced by S.S.M. and S.G. according to Willis’ recommendations on cognitive interviewing [13], Additional file 2). All sessions took place in the patients’ single rooms at the PCUs. The cognitive interviews were recorded, transcribed and analysed by S.S.M. and S.G. using qualitative content analysis [14] as described above. The basic coding tree results from the major themes of the cognitive interview. The resulting data constituted the second step in the adaption of the DTQP.

After delivery of the finalized generativity document, patients and relatives (as nominated by the patients and after having given informed consent) were asked to complete the DT Patient Feedback Questionnaire [15] and the DT Family Feedback Questionnaire [16] respectively. Both questionnaires previously had been translated into German by forward-backward translation (Additional file 3 and Additional file 4). They combine quantitative items (5-point Likert scale; 1 = strongly disagree, 5 = strongly agree) and free text comments.

All data were analysed descriptively with SPSS 22.0 and Microsoft Excel®.

Besides general remarks on the interpretation of the word ‘dignity’, which some HCPs viewed in a religious way and others in a more philosophical way, we observed five main categories of comments on DT and the DTQP: (1) negative aspects of DT; (2) positive aspects of DT; (3) Dignity Therapist; (4) conducting DT; and (5) the DTQP (Table 2). There were no comments on financing DT. Among the focus group conclusions was the opinion that the PCU is an appropriate setting for DT, as the patients have single rooms. HCPs also agreed that DT could be an essential part of the interdisciplinary care for patients if it is conducted with compassion: “If DT is conducted with heart, it could be an essential part of care and it becomes effective through this [compassion] and may be guided less by the questions that are asked but by the issue that lies on the patient’s heart” (M5). One HCP compared DT to the concept of ‘holding spaces’: “You give space for life issues, experiences and suffering, and this is dignifying and a valuable encounter” (W1).

A total of 410 patients were admitted to the PCUs in M (215 patients between June 2015 and October 2015) and W (195 patients between June 2015 and April 2016). Of these, 72 patients (17%) (M: 54; W: 18) met the inclusion criteria, and 30 of them (M: 19; W: 11) participated (7.3% of all patients and 41.7% of eligible patients). Participants’ inpatient stay ranged from 5 days to 98 days (median = 16 days). The mean age was 63 years (range: 38–88); 20 participants were female; the median Eastern Cooperative Oncology Group performance status (ECOG) was 3 with a 40% Palliative Performance Scale (PPS) rating. Table 3 provides further information on the participants’ characteristics.

The main exclusion criteria were: a life expectancy below two weeks (196 patients; 58%), followed by cognitive impairments (74 patients; 22%). A further 10% of eligible patients could not be included due to organizational barriers.

The median time-span of the whole DT process – from introducing it to the patient to returning the generativity document – was 15 days (mean: 24; range: 6–131). During this time, there were four contact sessions between patient and therapist on average. The mean time expenditure for therapists was 7 h (SD: 1.6; range: 4.27–10.38), comprising time to interview, edit, read out to the patient, make final editorial changes and return the finalized document to the recipient (note that this excludes time taken to transcribe the audio recording; this task was conducted by student assistants). The mean time expenditure for the patients was 2.15 h (SD: 0.5; range: 1.25–3.5). The mean time required for the DT interview was 48 min (SD: 19; range: 20–81). Once finalized, 19 of the generativity documents were given to the patients solely (to be handed out to the relatives by themselves or after their death), 3 documents were given to patients and relatives together, and 8 were received directly by relatives.

On average, nine questions were used per interview. Q1 (“Tell me a little about your life history; particularly the parts that you either remember most or think are the most important”) was always asked; the least asked question was Q6 (“Why were [these roles] so important to you and what do you think you accomplished within those roles?”). Q2 (“When did you feel most alive?”) was asked in 24 out of 30 interviews. In one interview, to match the language used by the patient, the wording changed from “alive” to “bloody good” (M 129). In 17 out of 23 interviews, Q3 (“Are there specific things that you would want your family to know about you?”) was responded to with comments like, “I don’t have any secrets”, “I think they know everything”, or “I don’t hide anything”. Q7 (“What are your most important accomplishments, and what do you feel most proud of?”) was used 26 times. We found that Q8 (“Are there particular things that you feel still need to be said to your loved ones?”) was sometimes personalized, e.g. “to your son” (5/16). Q10 (“What are your hopes and dreams for your loved ones?”) was used 27 times and reworded 17 times (from “dreams” to “wishes”). Similarly, Q12 (“What advice or words of guidance would you wish to pass along to your …?”) was used in 11 interviews and reworded (skipping ‘words of guidance’) without prior agreement between the interviewers.

Out of 30 patients, 29 subsequently took part in cognitive interviews to share their views on the DTQP (mean duration: 10 min; SD: 6; range: 2–30). We identified four main categories of feedback: on the title; the DTQP; the wording of individual questions; and the set of questions actually asked during the DT interview (Table 4). In addition, patients gave the following positive remarks and ideas regarding DT: the questions could lead to more openness in talking about cancer; DT preserves pride; DT should be widely offered; DT should be conducted in the presence of relatives; DT is only successful if it is a helpful encounter. More sceptical comments related to the effectiveness or usefulness of DT or to the fear of psychological challenges arising. One patient suggested adding the following question: “What has been your mission in your life?” (M 43). Regarding Q5, she recommended changing “remits” into “roles” (M 43); this was the de facto wording in 8 out of 23 interviews. For comparative purposes, Q7 (“What are your most important accomplishments, and what do you feel most proud of?”) was of particular interest: Danish colleagues found that using the word “proud” was not feasible for the Danish culture [10 Pride was only discussed as negative respectively positive by 2 patients each. Case M 136 illustrates this issue. Within the cognitive interview the patient stated that “pride is too strong” and that “it should be replaced with contentment”. Yet, during the DT interview she had easily answered that question by recounting what she had achieved on her own after getting divorced. The interview atmosphere was clearly not negatively affected by the word “proud”; in contrast it seemed to offer her the opportunity to be proud of herself.

Of the 30 patients who took part in the DT interview, 19 returned feedback questionnaires. In all, 18 patients evaluated DT as helpful (“It was helpful, because DT gave me a feeling of stability” [M 43]) or satisfactory (“It was far more than satisfactory. DT is a gift. It was more than I expected” [M 43]); 14 patients agreed that DT heightened their sense of dignity (“I feel accepted the way I am” [W 42]). All patients were satisfied with their psychosocial care. Statistics and selected comments for all items are presented in Table 5.

Out of 30 nominated relatives, 26 provided feedback and 23 of these evaluated DT as helpful for their loved ones (“To reminisce has brought up many smiles and shown how beautiful his life was and how precious life is.” [M N99]); 20 relatives agreed that DT was as important as any other aspect of care (“To talk about former times reduced spiritual pain.” [W N23]). Further, 24 relatives would recommend DT to other patients and families (“All terminally ill should receive help in this wonderful way.” [M N128]). Statistics and selected comments for all items are presented in Table 6.

The collected data led to the establishment of a consensus version, discussed by a DT expert group (S.S.M., S.G., E.J., and Jan Gramm and Jochen Spang). Slight changes to the intermediate translated version of the DTQP were made for Q3, Q4, Q5, Q8, Q9, Q10 and Q12. As “proud” was not a sensitive issue for HCPs or patients, Q7 remained. The final German version of the DTQP is provided as Additional file 5.