This study, conducted across multiple centers, was a randomized controlled trial conducted over a 7-year period (recruitment period: 2015 – 2021) designed to investigate the effectiveness of a DT intervention in patients and their FCs (partners or close family members) in three groups – patients receiving standard palliative care and DT (the DT group), patients receiving standard palliative care and DT along with their FCs (DT +) – relative to a control group consisting of patients who received standard palliative care without DT (SPC). Patients’ FCs were included into the study and asked to fill in a set of questionnaires. Written informed consent was obtained from each participant (patients and their FCs) prior to study participation. Participants were made aware of their freedom to withdraw from the study at any point in time with no unfavorable consequences to them. This study was approved by the Ethics Committee of the Canton Zurich (KEK), Switzerland (KEK-ZH-Nr. PB_2016-01275). The study was also registered with ClinicalTrials.gov (Protocol Record NCT02646527). Furthermore, all procedures adhered to the World Health Organization's Declaration of Helsinki. The study implemented the CONSORT (Consolidated Standards of Reporting Trials) statement as a framework for reporting and conducting the research (the CONSORT 2010 checklist can be found in an Additional file 1) [47].

Participants

Study participants were recruited from three study sites, which included the University Hospital Zurich with various medical institutions: the Competence Center for Palliative Care (JJ/DB), the Department of Medical Oncology and Hematology (BP), the Department of Gynecology (PI), and the Department of Dermatology (RD). Additionally, study participants were recruited from the Susenberg Clinic (CN) and the Zurich Lighthouse Hospice (JJ). The study procedures and adherence to ethical guidelines at each study site were overseen by a designated study advisor (see author abbreviation in brackets). All three clinics are situated in close proximity. To be eligible for the study, individuals had to be at least 18 years of age, be diagnosed with terminal cancer (expected life-expectancy ≤ 6 months), be in a relationship with a partner for at least two years or have an informal caregiver, and have a Hospital Anxiety and Depression Scale (HADS) score of at least eight points during pre-screening. Participants with cognitive impairments such as dementia or delirium, those who were too ill to fulfill the study requirements, and those who were unable to read and speak German were excluded from the study.

Implementation of dignity therapy

The Dignity Therapy was implemented across the three study sites using a systematic and structured approach. The implementation process encompassed several key steps, including training and education of various health professionals, including physicians, nurses, chaplains, social workers, clinical psychologists, physiotherapists, and occupational therapists at each study site. The training involved a one-hour teaching lecture on the Dignity Therapy intervention, followed by a presentation of the study and its procedures. During these training sessions, essential resources, such as study flyers and the Dignity Therapy inventory were provided to each study site.

Procedures

While being mindful of the patient's vulnerability, in an initial encounter with the patient and their FC, DT was introduced briefly as a potential supportive care option by a physician or a nurse working on one of the three study sites. Potential patients and their FCs were provided with complete information regarding the study by the study coordinator. Because it might be difficult to talk about the end of life or the possibility of death, eligible study participants were given enough time (at least 24 h) to consider their study participation and to think about the content of the DT interview. If the patients and their FCs showed interest in DT, the research coordinator followed up with them to schedule a research visit. Once the patient and their FC provided their written consent to participate, a pre-assessment was performed to screen for each patient’s anxiety and depression levels. Only those patients with clinically-significant stress levels, defined as a total HADStot ≥ 8 were considered eligible for the study. Study participants were randomly assigned to either DT + , DT, or SPC in a 1:1:1 ratio using the extended stratified block without list algorithm offered by secuTrial [49].

Prior to and after the DT intervention, participants and their FCs were required to complete a standardized set of questionnaires. Primary and secondary outcomes were evaluated in patients and FCs at baseline (T0) and one-week post intervention (T1). In addition, FCs were recontacted two weeks (T2) and three months (T3) following their partner’s death (Fig. 1).

Fig. 1Dignity therapy

Dignity Therapy is a brief, individual, dignity-promoting intervention that enables persons to perform a guided appraisal of meaningful moments and memories in their life while creating a unique written legacy for their families (i.e., ‘generativity document’) [10]. In our study, the DT intervention included three sessions: a brief introduction, the main session, and a final session. During the main session (approximately 60 min in duration), patients, either alone or accompanied by their partner or caregivers, were engaged in a dignity-enhancing conversation guided by a trained therapist, employing a semi-structured interview protocol to elicit the desired thoughts and expressions that patients wished to communicate to their loved ones. In the DT + group, the patients and their family caregivers were instructed that the patient's narrative holds primary importance. However, if the patient desired, the family caregiver could contribute to the patient’s narrative for entirety of the patient’s story. The semi-structured interview guideline is adaptable to patients’ individual needs and specifications and may involve reflections on personal history, important achievements, roles, hopes, wishes, dreams, and anything else the study participants wished to be remembered or shared about themselves. The patients’ narrations were recorded on audiotape, transcribed verbatim, and edited to produce a readable and clear narrative: the generativity document. During the final session (approximately 60 min in duration), the generativity document was read aloud to the patient to ensure accuracy and make any necessary corrections. The final version of the document was then given to each patient to share with their significant other or family.

Training and education of DT intervention therapists

One of the authors, AS, completed a DT training workshop with the originator, Harvey Max Chochinov (University Hospital Basel, Switzerland, 2018) and with Sandra Mai and Jan Gramm (Mainz, Frankfurt, 2015), respectively. AS trained a total of four psychologists with more than 3 years of clinical experience in the field of psycho-oncology during a one-day workshop including role plays, training videos, and lecture material. In addition, AS provided continued supervision during the recruitment process. DT therapists used a standardized protocol that clearly outlined the steps, procedures, and techniques of the DT intervention to ensure fidelity and consistency of intervention delivery across different sessions and DT therapists.

Outcomes on acceptability and feasibility

Acceptability of DT was assessed using a series of items with 5-point Likert response scales, which allowed participants to rate their DT experience. These items were specifically crafted to evaluate participants' perceptions regarding how helpful DT was for themselves and their partners or caregivers, its capacity to enhance personal meaning, their inclination to recommend DT to other patients and families, and the benefits they obtained from their participation in DT. Likert scales have been utilized extensively in research as robust and valid instruments to measure the attitudes, opinions, and experiences of study participants [54].

The feasibility of DT was assessed by measuring the number of visits made by therapists, the time taken by therapists to conduct the interviews, transcribe and edit the generativity documents, and the participants’ dropout rate. The number of visits and time taken by therapists provided insights into the potential burden that DT may impose on healthcare resources. Meanwhile, the drop-out rate was a crucial factor in assessing the feasibility of DT among terminally-ill adults with cancer. This set of feasibility metrics provided valuable information to aid in determining the practicality of integrating DT into routine palliative care.

Measures

Sociodemographic and clinical data of patients and their partners or caregivers were collected at baseline, including patient age, gender, relationship status, educational level, employment, and religious preferences. Patients’ medical data, including time since their cancer diagnosis, type of cancer, cancer stage, and current therapy regimen, were obtained by reviewing medical records. In addition, patients were asked about their inclination towards hastened death, utilizing a seven-point Likert scale (0 = none, 1 = minimum, 2 = mild, 3 = moderate, 4 = strong, 5 = severe, 6 = extreme).

Primary outcome

The Hospital Anxiety and Depression Scale (HADS) is a validated and widely-used self-report 14-item questionnaire that assesses individuals’ self-perceived levels of depression and anxiety [60]. It contains seven items each for depression and anxiety, each item accompanied by response options ranging from zero to three, with zero being the least and 3 the most indicative of symptoms. It can be used to identify patients with clinically-relevant symptoms of depression and anxiety (probable anxiety: HADS anxiety score > 7; probably depression: HADS depression score > 7). In cancer patients, a HADS total score of ≥ 13 detected 76% with a specificity of 0.60, while a score of ≥ 6 detected 95%, albeit with a lower specificity of 0.21. In clinical studies, the choice between high detection rates or low misclassification rates is crucial when using the HADS [51]. To be eligible for this study, we opted for a middle ground, selecting a cut-off score of a HADS total score ≥ 8.

Secondary outcomes

The Distress Thermometer (DTherm) is a unidimensional tool that utilizes a visual analog scale (VAS) with values ranging from 0 (no distress) to 10 (extreme distress) to assess an individual's self-perceived level of overall distress [44].

Functional Assessment of Chronic Illness Therapy Palliative Care (FACIT-Pal) and Spiritual Wellbeing Scale (FACIT-SP) are short, validated instruments used to measure quality of life in palliative care patients. The FACIT-Pal explores symptoms that frequently occur in advanced illness, family and friend relationships, life closure issues, and decision-making and communication abilities [59]. The FACIT-SP [39] specifically assesses spiritual components of quality of life. The present study utilized two subscales from the FACIT-SP: ‘meaning/peace’ and ‘faith’.

Patient Dignity Inventory—German Version (PDI-G) is a 25-item questionnaire that has been demonstrated to possess both validity and reliability measuring issues related to dignity at the end of life, including physical symptom distress, loss of sense of worth and meaning, loss of autonomy, anxiety, and uncertainty [13]. Recently, the questionnaire was translated into German [45].

WHO Quality of Life Questionnaire (WHOQOL-BREF) is a generic, cross-cultural instrument to measure quality of life. It consists of 26 items covering the domains of physical and psychological health, social relationships, and environment, as well as overall quality of life and general health [2].

PRISM (Pictorial Representation of Illness and Self Measure) was developed and validated by Büchi et al. [8] as a simple instrument to visualize illness burden and suffering.

Statistical methods

The sample size required to achieve an anticipated effect size of a mean difference in HADS total score of 3.0 (with an SD of change 7.7), based on an 80% power and a 5% error, was calculated to be 42 patients in each group. All analyses were performed using the Statistical Package for the Social Sciences (SPSS) version 26.0 software [24]. Descriptive statistics were reported as means (M) and standard deviations (SD), or as counts and percentages (%), as appropriate. All continuous variables were tested for normality using the Shapiro–Wilk’s test. Inter-group comparisons of continuous outcomes were performed using Student's t-tests for parametric or Mann–Whitney U tests for non-parametric/non-normally distributed data, while categorical variables were compared using Pearson’s-χ2 or Fisher’s exact test, as appropriate.

Linear mixed model analysis was utilized to investigate the impact of group (DT, DT + , SPC) and time (baseline T0, post-intervention T1, 2-week FU1, and 3-month follow-up FU2). Our approach to missing data was the assumptions that data was missing at random [5]. Because issues of multiplicity can arise due to repeated measures of the same outcome, a group*time interaction term was also included in the analysis [31]. Post-hoc analyses were performed using estimated marginal means to compare time points within groups and between groups at different times. The interaction effect between group and time was evaluated to assess whether improvement varied depending on group membership. The use of linear mixed model analysis with all available outcome data corresponds to intention-to-treat analysis (ITT), which includes all patients who intended to receive treatment at baseline.

In a subsequent step, we merged the DT and DT + groups at each measurement time point to increase statistical power. Doing so, we then utilized the same methodology as before, including linear mixed model analysis for all available outcome data. Post-hoc comparisons using estimated marginal means were performed to compare time points within the merged group. All tests were two-tailed and a p value < 0.05 was considered statistically significant.