Serious illness conversations and quality of end-of-life care in patients with hematological malignancies-a retrospective quality improvement study.

Introduction

Despite significant advances in the treatment of hematological malignancies, most remain associated with high mortality, necessitating a strong focus toward end-of-life care [ 1 , 2 ]. Compared to patients with solid tumors, patients with hematological malignancies receive more aggressive treatments and low-quality end-of-life care, including more frequent emergency room visits, end-of-life hospitalizations, anticancer treatments, and less often specialized palliative care services [ 3 – 7 ]. Complicating decision-making, many patients with hematological malignancies overestimate their chances of cure [ 8 – 12 ], leading to lower hospice utilization and increased hospitalizations [ 13 ]. A crucial step in avoiding futile treatments and reducing suffering may be effective communication between patients, their caregivers, and healthcare providers about prognosis, values, and goals of care.

The Serious Illness Care Program [ 14 ] was developed to enhance quality of communication through structured conversation tools, clinician training, and system changes. In a cluster-randomized trial involving 286 cancer patients, the program resulted in more and earlier conversations centered on values, goals, prognostic understanding, and in reductions in symptoms of anxiety and depression [ 15 , 16 ]. Another recent randomized trial, including 25% with lymphoma and multiple myeloma, in the USA demonstrated that increasing the number of serious illness conversations reduced use of out-patient systemic therapy at end of life [ 17 ]. The few retrospective studies that have examined the association of such conversations with quality of end-of-life care, specifically in patients with hematological malignancies, found significantly improved documentation of goals-of-care [ 18 ], lower odds of intensive care unit admission, and earlier hospice enrolment [ 19 ]. However, these studies did not explore an association with anticancer treatment, referral to specialized palliative care, or place of death.

Aim

We utilized data from the Danish adaptation and implementation of the Serious Illness Care Program in a hematological outpatient setting [ 20 , 21 ] to explore the association between receiving a serious illness conversation and quality of end-of-life care, and the timing of serious illness conversations in patients with hematological malignancies.

Methods

The reporting of this study adheres to the Revised Standards for Quality Improvement Reporting Excellence (SQUIRE 2.0) [ 22 ].

Study design and population

We conducted a single-center retrospective quality improvement study at the Department of Hematology, Copenhagen University Hospital, Rigshospitalet. Patients seen by physicians who had received Serious Illness Conversation training (see intervention description below) were screened. This study included all patients ≥ 18 years of age who died from a hematological malignancy between January 2020 and December 2022 and had received anticancer treatment at the department within 12 months before death. Patients were excluded if they had been formally discharged from our hospital’s care and transferred to another hospital before death, received only low-intensity palliative anticancer treatments (e.g., hydroxyurea or mercaptopurine), or underwent allogeneic hematopoietic stem cell transplantation as their final treatment, as transplantation clinicians were not part of the implementation.

Intervention and implementation

In 2020, a structure for serious illness conversations was implemented as the new standard of practice, including diagnosis-specific guidelines for conversation timing, allocated time for conversations, and standardized mandatory documentation templates in medical records. The diagnosis-specific guidelines recommended offering serious illness conversations (1) to patients with multiple myeloma, high-risk myelodysplastic syndrome, acute leukemia, and aggressive lymphoma (e.g., T cell lymphoma and CNS lymphoma) at the time of diagnosis; (2) to patients with myelofibrosis when weekly transfusions were required, and (3) to all patients at relapse and transition to palliative care. To train all clinicians, five separate 8-h interprofessional Serious Illness Care training days were conducted for the department’s physicians and nurses. Each clinician attended one full-day session, which included simulation-based skills practice with professional actors and was guided by a structured Serious Illness Conversation Guide [ 21 , 23 ]. Following the training, clinicians initiated structured conversations with patients and caregivers, addressing prognosis, concerns, hopes, treatment goals, and care preferences. Interprofessional group supervision led by a senior palliative care physician and the hospital chaplain supported conversation, ethical dilemmas, and emotional support. Additional support included preparatory materials for patients and caregivers.

Data collection

In 2023, after 3 years of implementation (2020–2022), records of potentially eligible patients were identified through electronic medical records and systematically assessed for eligibility by authors CBM, SNC, and SBB. In 2024, data were extracted from eligible patients’ records, including serious illness conversation documentation and end-of-life care outcomes. Dates of serious illness conversation documentation were recorded, and the number of days between conversations and death was calculated. Hospital days within the last 30 days of life were recorded, as well as the date of the last intravenous, intrathecal, subcutaneous, or oral administration of anticancer treatment, number of treatment lines given since diagnosis, documentation of do-not-resuscitate, referral to specialized palliative care (including referral to hospices and palliative care units), and place of death (hospital, intensive care unit, home (including nursing homes), specialized palliative care facilities, municipal facility e.g., rehabilitation center, or “unknown” if not documented).

Sex, age, and hematological diagnosis were extracted, and diagnoses were categorized into “multiple myeloma/amyloidosis,” “lymphoma,” and “acute leukemia/myelodysplastic syndrome/myeloproliferative neoplasms.” These covariates have previously been shown to influence end-of-life outcomes [ 24 – 27 ].

Statistical analysis

Frequencies and percentages were used to describe characteristics and end-of-life care outcomes for patients who received a serious illness conversation and those who did not (Table 1 ). To explore associations between receiving a serious illness conversation and end-of-life care outcomes, we ran logistic regression models adjusted for potential confounding covariates (age, sex, and diagnosis), estimating odds ratios and 95% confidence intervals (CI) for the end-of-life outcomes: anticancer treatment, hospitalization, referral to specialized palliative care, do-not-resuscitate and place of death (Table 2 ). Analyses were conducted in R [ 28 ]. Table 1 Patient characteristics and end-of-life care outcomes Characteristics All participants ( N = 311) No serious illness conversation (SIC) ( N = 248) Received serious illness conversation (SIC) ( N = 63) Age in years, median (IQR, min–max) 74 (12, 25–94) 74 (12, 25–91) 76 (13, 25–94) Lines of treatment, median (IQR, min–max) 2 (2, 1–12) 1 (2, 1–10) 3 (3, 1–12) Sex (female), N (%) 134 (43) 104 (42) 30 (48) Diagnosis, N (%) Multiple myeloma/amyloidosis 107 (34) 66 (27) 41 (65) Lymphoma 111 (36) 104 (42) 7 (11) Acute leukemia/MDS/MPN 93 (30) 78 (31) 15 (24) Place of death , N (%) Specialized palliative care 72 (23) 51 (20) 21 (33) Hospital 129 (41) 107 (43) 22 (35) Intensive care unit 28 (9) 26 (10) 2 (3) Home 56 (18) 41 (16) 15 (24) Municipality 12 (4) 10 (4) 2 (3) Unknown 14 (5) 13 (5) 1 (2) Hospitalization within the last 30 days of life, N (%) 260 (84) 214 (86) 46 (73) Days in hospital: 1 ≤ 14 days 137 (53)* 104 (49) 33 (72) Days in hospital: 15–30 days 123 (47)* 110 (51) 13 (28) Anticancer treatment 0 ≤ 30 days before death, N (%) † 167 (54) 151 (61) 16(25) Hereof 1 ≤ 14 days before death 101 (60) ‡ 92 (61) 9 (56) Referral to specialized palliative care, N (%) 139 (45) 99 (40) 40 (63) Do-not-resuscitate documented , N (%) 269 (87) 212 (86) 57 (90) Days to event, median (IQR, min–max) Days from last anticancer treatment to death* 28 (53, 0–362) 23 (43, 0–362) 54 (98, 0–310) Days from “Do-not-resuscitate” to death 26 (57, 0–886) 23 (49, 0–868) 54 (112, 1–886) Days from referred to specialized palliative care to death 18 (44, 1–393) 15 (28, 1–277) 35 (107, 2–393) Days between SIC 1 and death 204 (296, 2–720) Days between SIC 2 and death 129 (113, 12–229) MDS = Myelodysplastic syndrome; MPN = Myeloproliferative neoplasms *Percentage calculated based on the number of patients hospitalized within the last 30 days of life † Palliative anticancer treatment (Hydroxyurea/Thioguanine) excluded ‡ Percentage calculated based on the number of patients who received anticancer treatment ≤30 days before death

Results

A total of 311 patients were included (Fig. 1 ), 43% were female, with a median age of 74 years. Thirty-four percent had multiple myeloma or amyloidosis, 36% lymphoma, 30% acute leukemia, myelodysplastic syndrome, or myeloproliferative neoplasms. Patients had received a median of two lines of treatment since diagnosis (Table 1 ). Among all patients, 54% received anticancer treatment within the last 30 days of life, 84% were hospitalized within the last 30 days, and 45% were referred to specialized palliative care. Sixty-three patients (20%) had documented receipt of at least one serious illness conversation, with a median of 204 days before death. Twenty-two patients received a second conversation, with a median of 129 days before death (Table 1 ). Fig. 1 Flowchart of inclusion

Of the 298 patients with known place of death, the majority died in hospital ( n = 129), followed by specialized palliative care facilities (including hospice) ( n = 72), home ( n = 56), intensive care units ( n = 28), and municipal facilities ( n = 12) (Table 1 ).

Receiving a serious illness conversation was associated with significantly lower receipt of anticancer treatment in the last 30 days (OR 0.19, 95% CI [0.10; 0.37]) and 14 days of life (OR 0.21, 95% CI [0.09; 0.46) and significantly more referrals to specialized palliative care (OR 2.67, 95% CI [1.44; 4.91]) (Table 2 ). The associations between receiving a serious illness conversation and hospitalization within the last 30 days of life and documentation of do-not-resuscitate status or place of death did not reach statistical significance (Table 2 ).

Discussion

In this retrospective cohort study of serious illness conversations in patients with hematological malignancies, we found that 20% of patients received at least one conversation a median of 6.7 months prior to death. Receiving a conversation was associated with significantly lower odds of anticancer treatment in the last 30 and 14 days of life, as well as significantly higher odds of referral to specialized palliative care. Additionally, there was a trend toward more patients who had received a conversation dying at home, in specialized palliative care facilities, or municipal facilities, and fewer in intensive care units.

The high proportion of patients receiving anticancer treatment (54%) in the last 30 days of life and the association of receiving serious illness conversations with less end-of-life anticancer treatment, in our study highlight an important opportunity to reduce treatment intensity at end of life. A previous prospective study in the USA of 160 patients with acute leukemia reported that 65.9% of the 44 patients who died during follow-up received anticancer treatment in the last 30 days of life [ 29 ]. Similar to our study, patients who participated in timely conversations about illness understanding and treatment goals had lower odds of receiving anticancer treatment at end of life [ 29 ]. A retrospective study of 319 deceased patients with various hematological malignancies found that conversations about goals of treatment were an independent predictor of end-of-life outcomes and positively impacted anticancer treatment [ 30 ]. Other retrospective studies of various hematological malignancies reported lower rates (16–45%) of anticancer treatment at end of life, independent of a serious illness conversation based on data from 2009 to 2016 in the USA ( n = 113 [ 31 ]), Canada ( n = 319 [ 30 ]), and France ( n = 46,629 [ 24 ]). Advancements in therapies since that period may explain the higher prevalence of end-of-life treatment in our study. Yet, our study was conducted during the COVID-19 pandemic. Patients with hematological malignancies and COVID-19 experienced higher mortality [ 32 , 33 ], and acute death due to COVID-19 infections may have led to more anticancer treatment close to death. In fact, the pandemic highlighted an increased need for structured approaches, such as serious illness conversations, due to the heightened mortality.

In our study, patients who received a conversation had received more lines of treatment since diagnosis, likely because more patients with multiple myeloma, a malignancy often associated with more available treatment lines than acute leukemia and lymphomas, received conversations. We did not extract data on date of diagnosis and therefore cannot compare survival between groups. Our findings could, however, suggest that conversations were initially offered to patients with longer treatment trajectories, aligning with the natural progression of an implementation process where early adopters may prioritize patients who are more obviously suitable before extending conversations to earlier stages of disease. However, a median of three treatment lines (IQR 1–12) suggests that conversations were often initiated early.

We found a median of 204 days between the first serious illness conversation and death, reflecting a successful shift to earlier conversation timing. Receiving a serious illness conversation was significantly associated with higher odds of referral to specialized palliative care, with a median of 35 days from referral to death compared to 15 days for those without a conversation. This earlier conversation timing and referral may provide patients with more time in specialized palliative care, potentially enhancing the quality of end-of-life care. The role of timely discussions in optimizing end-of-life care has previously been documented in a retrospective study of 383 patients, where goals-of-care conversations conducted in outpatient settings with hematologists more than 30 days before death, compared to those held within 30 days, were associated with fewer intensive care unit admissions and longer enrolment in hospice care [ 19 ]). However, it is noteworthy that we found conversations conducted at a median time of approximately 6 months before death associated with changed end-of-life care. Nevertheless, another retrospective analysis of 332 deceased patients from a prospective multisite study found end-of-life discussions within the last 4 months of life associated with prolonged hospice stays [ 34 ]. In our study, the trend toward more home deaths may be driven by increased specialized palliative care referrals, as 60% of patients who died at home and had serious illness conversations had been referred to specialized palliative care compared to 41% of those without a conversation. Mixed outcomes across studies on other end-of-life outcomes suggest the need for further exploration of the timing and frequency of these conversations to gain insights into optimizing end-of-life care for patients with hematological malignancies.

Barriers among hematological clinicians to early integration of specialized palliative, such as fear of taking away hope, prognostic uncertainty, and misconceptions about palliative care as only end-of-life care, have historically led to underuse of specialized palliative care services [ 35 – 37 ]. Although all clinicians received serious illness communication training, a significant proportion of patients did not receive a serious illness conversation before death. This underscores the complex influence of clinical culture on adopting serious illness conversation practices during implementation and the intensive treatment culture within hematology. A French cross-sectional national study found that while most hematologists recognize the importance of end-of-life discussions, only a minority routinely address these topics, reflecting a knowledge-practice gap [ 38 ]. A qualitative study with 30 clinicians and leaders from five healthcare systems implementing the Serious Illness Care Program in the United States revealed that clinicians often view these conversations as solely concerning end-of-life planning, contributing to hesitancy [ 39 ]. Another qualitative study involving 14 physicians from cardiology, endocrinology, hematology, and palliative care, at two Swedish hospitals implementing the Serious Illness Care Program, highlighted concerns about timing, emotional harm, and undermining hope, which may delay these conversations [ 40 ]. In our study, most patients who received a serious illness conversation were diagnosed with multiple myeloma [65%] and were therefore treated by the same team of clinicians. Some of the clinicians treating multiple myeloma were involved in the process of adapting the Serious Illness Care Program to a Danish cultural setting, including participation in workshops with patients and caregivers, and were the first to receive training, which may partly explain the higher implementation of conversations in this group. Moreover, multiple myeloma is incurable from the time of diagnosis; thus, the treatment approach might generally be more palliative compared to that of acute leukemia or aggressive lymphoma, which may have further facilitated the integration of serious illness conversations into clinical practice. This suggests that we may have identified a subgroup of clinicians who were more engaged in conducting these conversations and might have different practices around end-of-life care. Yet, even within the multiple myeloma group, where uptake of conversations was highest, not all patients received a conversation before death, reflecting variability in implementation. This variation may be influenced by differences in clinician engagement, integration of the training into practice, or patient-level factors such as comorbidities or disease trajectory, although our data do not allow further exploration of these potential explanations. Shifting the focus from “end-of-life planning” to “understanding what matters most to patients” may help bridge the gap between knowledge and practice [ 39 ]. Further research is needed on the use of serious illness conversations in diagnoses like acute leukemia, where prognostic uncertainty is high, and the potential impact may be substantial.

The small percentage of patients receiving serious illness conversations may also be due to patients declining the invitation. A multi-site cross-sectional study from Canada found that patients may decline due to concerns about misaligned values, emotional distress, not feeling mentally ready, or insufficient time to address complex issues [ 41 ]. A qualitative study with 44 patients, caregivers, and clinicians in a randomized trial identified barriers, including patients’ coping strategies, such as focusing on the present and maintaining optimism, and clinicians’ hesitancy and discomfort with discussing prognostic information, which discouraged patient engagement [ 42 ]. Other studies have highlighted racial and ethnic disparities in the willingness to engage in serious illness conversations or advance care planning discussions [ 43 – 46 ], as well as prevalent symptoms of anxiety and depression in the patients, which may affect their willingness to participate [ 47 , 48 ].

Strengths and limitations

Our study has several strengths and limitations. First, the retrospective nature of the data limits our ability to establish causality. While associations have been identified, we cannot determine whether observed differences are due to the conversations or other unmeasured factors. Our study was also limited by the lack of access to data on racial or ethnic information, symptoms of anxiety and depression in patients, and clinicians’ attitudes toward end-of-life care, which may introduce confounding factors. The lack of detailed socio-demographic data limits our ability to fully interpret differences between patients who received a serious illness conversation and those who did not, as these differences may reflect pre-existing patient or clinician characteristics rather than the effects of the conversations themselves. We were, however, able to adjust for age, sex, and subtype of hematological malignancy, enhancing the robustness of our analysis. As a single-center study, the results may have limited generalizability to other settings where the culture regarding engagement in serious illness conversations and end-of-life care may differ. Finally, no fidelity assessment was performed to document the quality of the conversations, reducing the ability to attribute associations to specific components of the conversations. Despite this, the implementation had robust management support, enabling comprehensive training for, and participation of all clinicians at the Department of Hematology. We have previously documented in a qualitative study that physicians and nurses experienced that interprofessional delivery strengthened their collaboration [ 21 ]. Additionally, the implementation of system changes, including allocated time for conversations and standardized documentation templates, further reinforced the integration and implementation of the conversations. Moreover, our inclusion of various hematological subtypes provides a more comprehensive understanding of the potential impact of serious illness conversations across a broad spectrum of hematological patients. Finally, our study is among the few to examine the associations between implementing serious illness conversations in real-world settings and end-of-life outcomes among patients with diverse hematological malignancies.

Conclusion

Serious illness conversations are associated with a reduction in end-of-life anticancer treatment and an increase in referrals to specialized palliative care.

Building on these hypothesis-generating results, we have initiated a randomized controlled trial to validate and further explore these findings.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Annika von Heymann and Stine Novrup Clemmensen have contributed equally as the last authors.

Acknowledgements

We would like to express our gratitude to Klaus Kaae Andersen for his invaluable statistical expertise and support and to the leadership team at both the Center for Cancer and Organ Diseases and the Department of Hematology, Copenhagen University Hospital, for their encouragement and support in implementing the Danish adaptation of the Serious Illness Care Program.

Author contribution

CBM was the primary author of this manuscript, with critical revisions for intellectual content by CJ, RB, JK, JJ, SBB, MJ, AVH, and SNC. CBM, SBB, and SNC collected the data. RB, JJ, JK, SNC, AVH, and CBM developed the statistical analysis plan, and CBM conducted the statistical analysis. SNC and AVH contributed equally as senior authors.

Funding

Open access funding provided by National Hospital

Data availability

No datasets were generated or analysed during the current study.

Declarations

Research ethics

The study was approved by the legal team at Copenhagen University Hospital, Rigshospitalet, (no. 22072599) and conducted in compliance with Danish regulations, the Helsinki II Declaration, and the European General Data Protection Regulation (GDPR). As this is a retrospective observational study, approval from the Regional Ethics Committee of the Capital Region of Denmark is not required as per Danish law [ 49 ].

Consent to participate

As this is a retrospective observational study, informed consent was not required as the study involved only medical record review of deceased patients [ 49 ].

Competing interests

The authors declare no competing interests.

References

1. Pulte D Jansen L Brenner H Changes in long term survival after diagnosis with common hematologic malignancies in the early 21st century Blood Cancer J 2020 10 56 10.1038/s41408-020-0323-4 32404891 PMC7221083 Pulte D, Jansen L, Brenner H (2020) Changes in long term survival after diagnosis with common hematologic malignancies in the early 21st century. Blood Cancer J 10:56. 10.1038/s41408-020-0323-4 32404891 10.1038/s41408-020-0323-4 PMC7221083

2. Hemminki K, Hemminki J, Försti A, Sud A. Survival in hematological malignancies in the Nordic countries through a half century with correlation to treatment. Leukemia [Internet]. 2023 Apr 1 [cited 2024 Sep 22];37(4):854–63. Available from: https://pubmed.ncbi.nlm.nih.gov/36828868/ 10.1038/s41375-023-01852-w PMC10079539 36828868

3. Sexauer A, Cheng MJ, Knight L, Riley AW, King L, Smith TJ. Patterns of hospice use in patients dying from hematologic malignancies. J Palliat Med [Internet]. 2014 Feb 1 [cited 2024 Sep 22];17(2):195–9. Available from: https://pubmed.ncbi.nlm.nih.gov/24383458/ 10.1089/jpm.2013.0250 PMC3997144 24383458

4. O’Connor NR, Hu R, Harris PS, Ache K, Casarett DJ. Hospice admissions for cancer in the final days of life: independent predictors and implications for quality measures. J Clin Oncol [Internet]. 2014 Oct 1 [cited 2024 Sep 22];32(28):3184–9. Available from: https://pubmed.ncbi.nlm.nih.gov/25154824/ 10.1200/JCO.2014.55.8817 PMC4171361 25154824

5. Hui D, Didwaniya N, Vidal M, Shin SH, Chisholm G, Roquemore J, et al. Quality of end-of-life care in patients with hematologic malignancies: a retrospective cohort study. Cancer [Internet]. 2014 May 15 [cited 2024 Oct 28];120(10):1572–8. Available from: https://pubmed.ncbi.nlm.nih.gov/24549743/ 10.1002/cncr.28614 PMC4013204 24549743

6. El-Jawahri AR, Abel GA, Steensma DP, LeBlanc TW, Fathi AT, Graubert TA, et al. Health care utilization and end-of-life care for older patients with acute myeloid leukemia. Cancer [Internet]. 2015 Aug 1 [cited 2024 Sep 22];121(16):2840–8. Available from: https://pubmed.ncbi.nlm.nih.gov/25926135/ 10.1002/cncr.29430 PMC4418225 25926135

7. Ho TH, Barbera L, Saskin R, Lu H, Neville BA, Earle CC. Trends in the aggressiveness of end-of-life cancer care in the universal health care system of Ontario, Canada. J Clin Oncol [Internet]. 2011 Apr 20 [cited 2024 Sep 22];29(12):1587–91. Available from: https://pubmed.ncbi.nlm.nih.gov/21402603/ 10.1200/JCO.2010.31.9897 PMC3082976 21402603

8. El-Jawahri A, Nelson-Lowe M, VanDusen H, Traeger L, Abel GA, Greer JA, et al. Patient-clinician discordance in perceptions of treatment risks and benefits in older patients with acute myeloid leukemia. Oncologist. 2018/08/25. 2019;24(2):247–54. 10.1634/theoncologist.2018-0317 PMC6369944 30139841

9. El-Jawahri A, Traeger L, Kuzmuk K, Eusebio J, Vandusen H, Keenan T, et al. Prognostic understanding, quality of life and mood in patients undergoing hematopoietic stem cell transplantation. Bone Marrow Transplant. 2015/05/12. 2015;50(8):1119–24. 10.1038/bmt.2015.113 PMC4526323 25961772

10. Gray TF, Temel JS, El-Jawahri A. Illness and prognostic understanding in patients with hematologic malignancies. Blood Rev. 2020/04/15. 2021;45:100692. 10.1016/j.blre.2020.100692 32284227

11. Loh KP, Xu H, Back A, Duberstein PR, Gupta Mohile S, Epstein R, et al. Patient-hematologist discordance in perceived chance of cure in hematologic malignancies: a multicenter study. Cancer. 2019/12/07. 2020;126(6):1306–14. 10.1002/cncr.32656 PMC7050385 31809566

12. Shimer S, Allen OS, Yang C, Canavan M, Westvold S, Kim N, et al. Prognostic understanding, goals of care, and quality of life in hospitalized patients with leukemia or multiple myeloma. J Palliat Med [Internet]. 2024 Jul 1 [cited 2024 Oct 13];27(7):879–87. Available from: https://pubmed.ncbi.nlm.nih.gov/38990245/ 10.1089/jpm.2023.0530 38990245

13. Loh KP, Seplaki CL, Sanapala C, Yousefi-Nooraie R, Lund JL, Epstein RM, et al. Association of prognostic understanding with health care use among older adults with advanced cancer: a secondary analysis of a cluster randomized clinical trial. JAMA Netw Open [Internet]. 2022 Feb 18 [cited 2024 Nov 25];5(2). Available from: https://pubmed.ncbi.nlm.nih.gov/35179585/ 10.1001/jamanetworkopen.2022.0018 PMC8857680 35179585

14. Bernacki R, Hutchings M, Vick J, Smith G, Paladino J, Lipsitz S, et al. Development of the Serious Illness Care Program: a randomised controlled trial of a palliative care communication intervention. BMJ Open. 2015/10/08. 2015;5(10):e009032. 10.1136/bmjopen-2015-009032 PMC4606432 26443662

15. Bernacki R, Paladino J, Neville BA, Hutchings M, Kavanagh J, Geerse OP, et al. Effect of the Serious Illness Care Program in outpatient oncology: a cluster randomized clinical trial. JAMA Intern Med. 2019/03/15. 2019;179(6):751–9. 10.1001/jamainternmed.2019.0077 PMC6547155 30870563

16. Paladino J, Bernacki R, Neville BA, Kavanagh J, Miranda SP, Palmor M, et al. Evaluating an intervention to improve communication between oncology clinicians and patients with life-limiting cancer: a cluster randomized clinical trial of the Serious Illness Care Program. JAMA Oncol. 2019/03/15. 2019;5(6):801–9. 10.1001/jamaoncol.2019.0292 PMC12634133 30870556

17. Manz CR Zhang Y Chen K Long Q Small DS Evans CN Long-term effect of machine learning-triggered behavioral nudges on serious illness conversations and end-of-life outcomes among patients with cancer: a randomized clinical trial JAMA Oncol 2023 9 3 414 418 10.1001/jamaoncol.2022.6303 36633868 PMC9857721 Manz CR, Zhang Y, Chen K, Long Q, Small DS, Evans CN et al (2023) Long-term effect of machine learning-triggered behavioral nudges on serious illness conversations and end-of-life outcomes among patients with cancer: a randomized clinical trial. JAMA Oncol 9(3):414–418 36633868 10.1001/jamaoncol.2022.6303 PMC9857721

18. Hui D, Ahmed S, Nortje N, George M, Andersen CR, Wilson K, et al. Differential impact of a multicomponent goals-of-care program in patients with hematologic and solid malignancies. cancers (Basel). 2023/03/12. 2023;15(5). 10.3390/cancers15051507 PMC10001115 36900298

19. Odejide OO Uno H Murillo A Tulsky JA Abel GA Goals of care discussions for patients with blood cancers: association of person, place, and time with end-of-life care utilization Cancer 2020 10.1002/cncr.32549 31593321 PMC6980246 Odejide OO, Uno H, Murillo A, Tulsky JA, Abel GA (2020) Goals of care discussions for patients with blood cancers: association of person, place, and time with end-of-life care utilization. Cancer. 10.1002/cncr.32549 31593321 10.1002/cncr.32549 PMC6980246

20. Borregaard Myrhøj C, Novrup Clemmensen S, Sax Røgind S, Jarden M, Toudal Viftrup D. Serious illness conversations in patients with multiple myeloma and their family caregivers—a qualitative interview study. Eur J Cancer Care (Engl). 2021/11/05. 2021; 10.1111/ecc.13537 34734446

21. Myrhøj CB, Viftrup DT, Jarden M, Clemmensen SN. Interdisciplinary collaboration in serious illness conversations in patients with multiple myeloma and caregivers—a qualitative study. BMC Palliat Care. 2023 Dec 1;22(1). 10.1186/s12904-023-01221-5 PMC10339578 37438765

22. Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2016 Dec 1;25(12):986–92. 10.1136/bmjqs-2015-004411 PMC5256233 26369893

23. Bernacki RE, Block SD. Communication about serious illness care goals: a review and synthesis of best practices. JAMA Intern Med. 2014/10/21. 2014;174(12):1994–2003. 10.1001/jamainternmed.2014.5271 25330167

24. Beaussant Y Daguindau E Chauchet A Rochigneux P Tournigand C Aubry R Hospital end-of-life care in haematological malignancies BMJ Support Palliat Care 2018 8 3 314 324 10.1136/bmjspcare-2017-001446 29434048 Beaussant Y, Daguindau E, Chauchet A, Rochigneux P, Tournigand C, Aubry R et al (2018Sep 1) Hospital end-of-life care in haematological malignancies. BMJ Support Palliat Care 8(3):314–324 29434048 10.1136/bmjspcare-2017-001446

25. Malhotra C, Bundoc F, Chaudhry I, Teo I, Ozdemir S, Finkelstein E, et al. A prospective cohort study assessing aggressive interventions at the end-of-life among patients with solid metastatic cancer. BMC Palliat Care [Internet]. 2022 Dec 1 [cited 2024 Nov 12];21(1). Available from: https://pubmed.ncbi.nlm.nih.gov/35578270/ 10.1186/s12904-022-00970-z PMC9109395 35578270

26. Wang R Zeidan AM Halene S Xu X Davidoff AJ Huntington SF Health care use by older adults with acute myeloid leukemia at the end of life J Clin Oncol 2017 35 30 3417 3424 10.1200/JCO.2017.72.7149 28783450 PMC5648174 Wang R, Zeidan AM, Halene S, Xu X, Davidoff AJ, Huntington SF et al (2017) Health care use by older adults with acute myeloid leukemia at the end of life. J Clin Oncol 35(30):3417–3424 28783450 10.1200/JCO.2017.72.7149 PMC5648174

27. Odejide OO, Cronin AM, Earle CC, Lacasce AS, Abel GA. Hospice use among patients with lymphoma: impact of disease aggressiveness and curability. J Natl Cancer Inst. 2016 Jan 1;108(1). 10.1093/jnci/djv280 26438575

28. R Core Team R: a language and environment for statistical computing 2024 Vienna, Austria R Foundation for Statistical Computing R Core Team (2024) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria

29. El-Jawahri A, LeBlanc TW, Kavanaugh A, Webb JA, Jackson VA, Campbell TC, et al. Effectiveness of integrated palliative and oncology care for patients with acute myeloid leukemia: a randomized clinical trial. JAMA Oncol. 2020/12/18. 2021;7(2):238–45. 10.1001/jamaoncol.2020.6343 PMC7747042 33331857

30. Korsos V, Ali A, Malagon T, Khosrow-Khavar F, Thomas D, Sirhan S, et al. End of life in haematology: quality of life predictors - retrospective cohort study. BMJ Support Palliat Care [Internet]. 2023 [cited 2024 Dec 4]; Available from: https://pubmed.ncbi.nlm.nih.gov/37068924/ 10.1136/spcare-2023-004218 37068924

31. Hui D, Didwaniya N, Vidal M, Shin SH, Chisholm G, Roquemore J, et al. Quality of end-of-life care in patients with hematologic malignancies: a retrospective cohort study. Cancer [Internet]. 2014 May 15 [cited 2024 Sep 21];120(10):1572–8. Available from: https://pubmed.ncbi.nlm.nih.gov/24549743/ 10.1002/cncr.28614 PMC4013204 24549743

32. Vijenthira A, Gong IY, Fox TA, Booth S, Cook G, Fattizzo B, et al. Outcomes of patients with hematologic malignancies and COVID-19: a systematic review and meta-analysis of 3377 patients. Blood [Internet]. 2020 Dec 17 [cited 2025 Feb 10];136(25):2881–92. Available from: https://pubmed.ncbi.nlm.nih.gov/33113551/ 10.1182/blood.2020008824 PMC7746126 33113551

33. Hardy N Vegivinti CTR Mehta M Thurnham J Mebane A Pederson JM Mortality of COVID-19 in patients with hematological malignancies versus solid tumors: a systematic literature review and meta-analysis Clin Exp Med 2023 23 6 1945 1959 10.1007/s10238-023-01004-5 36795239 PMC9933827 Hardy N, Vegivinti CTR, Mehta M, Thurnham J, Mebane A, Pederson JM et al (2023Oct 1) Mortality of COVID-19 in patients with hematological malignancies versus solid tumors: a systematic literature review and meta-analysis. Clin Exp Med 23(6):1945–1959 36795239 10.1007/s10238-023-01004-5 PMC9933827

34. Wright AA, Zhang B, Ray A, Mack JW, Trice E, Balboni T, et al. Associations between end-of-life discussions, patient mental health, medical care near death, and caregiver bereavement adjustment. JAMA [Internet]. 2008 Oct 8 [cited 2024 Aug 27];300(14):1665–73. Available from: https://pubmed.ncbi.nlm.nih.gov/18840840/ 10.1001/jama.300.14.1665 PMC2853806 18840840

35. LeBlanc TW, El-Jawahri A. When and why should patients with hematologic malignancies see a palliative care specialist? Hematology Am Soc Hematol Educ Program [Internet]. 2015 Dec 5 [cited 2024 Dec 16];2015(1):471–8. Available from: https://pubmed.ncbi.nlm.nih.gov/26637760/ 10.1182/asheducation-2015.1.471 26637760

36. LeBlanc TW Roeland EJ El-Jawahri A Early palliative care for patients with hematologic malignancies: is it really so difficult to achieve? Curr Hematol Malig Rep 2017 12 4 300 308 10.1007/s11899-017-0392-z 28639084 LeBlanc TW, Roeland EJ, El-Jawahri A (2017Aug 1) Early palliative care for patients with hematologic malignancies: is it really so difficult to achieve? Curr Hematol Malig Rep 12(4):300–308 28639084 10.1007/s11899-017-0392-z

37. Fadul N, Elsayem A, Palmer JL, Zhang T, Braiteh F, Bruera E. Predictors of access to palliative care services among patients who died at a comprehensive cancer center. J Palliat Med [Internet]. 2007 Oct 1 [cited 2024 Dec 16];10(5):1146–52. Available from: https://pubmed.ncbi.nlm.nih.gov/17985971/ 10.1089/jpm.2006.0259 17985971

38. Serey K, Cambriel A, Pollina-Bachellerie A, Bay JO, Bouleuc C, Ladrat L, et al. Hematologists’ perspective on advance directives, a French national cross-sectional survey—the ADORE-H study. BMC Med Ethics [Internet]. 2024 Nov 27 [cited 2024 Dec 5];25(1):142. Available from: https://pubmed.ncbi.nlm.nih.gov/39605054/ 10.1186/s12910-024-01146-5 PMC11600615 39605054

39. Paladino J, Sanders JJ, Fromme EK, Block S, Jacobsen JC, Jackson VA, et al. Improving serious illness communication: a qualitative study of clinical culture. BMC Palliat Care [Internet]. 2023 Dec 1 [cited 2024 Dec 12];22(1). Available from: https://pubmed.ncbi.nlm.nih.gov/37481530/ 10.1186/s12904-023-01229-x PMC10362669 37481530

40. Baxter R, Pusa S, Andersson S, Sandgren A. Perils and payoffs for patients in serious illness conversations as described by physicians: a qualitative study. BMJ Open Qual [Internet]. 2024 May 23 [cited 2024 Dec 12];13(2). Available from: https://pubmed.ncbi.nlm.nih.gov/38782485/ 10.1136/bmjoq-2023-002734 PMC11116881 38782485

41. Bernard C, Tan A, Slaven M, Elston D, Heyland DK, Howard M. Exploring patient-reported barriers to advance care planning in family practice. BMC Fam Pract [Internet]. 2020 May 25 [cited 2024 Dec 5];21(1). Available from: https://pubmed.ncbi.nlm.nih.gov/32450812/ 10.1186/s12875-020-01167-0 PMC7249389 32450812

42. Hanley S, Cotner CE, Fenton A, Wright AA, Manz CR. Barriers to serious illness conversations among patients with advanced cancer: a qualitative study. J Pain Symptom Manage [Internet]. 2024 [cited 2024 Dec 31]; Available from: https://pubmed.ncbi.nlm.nih.gov/39515596/ 10.1016/j.jpainsymman.2024.10.034 39515596

43. Abbas M, Chua IS, Tabata-Kelly M, Bulger AL, Gershanik E, Sheu C, et al. Racial and ethnic disparities in serious illness conversation quality during the COVID-19 Pandemic. J Pain Symptom Manage [Internet]. 2024 Sep 1 [cited 2024 Dec 5];68(3):205–213.e6. Available from: https://pubmed.ncbi.nlm.nih.gov/38782305/ 10.1016/j.jpainsymman.2024.05.016 38782305

44. Smith AK, McCarthy EP, Paulk E, Balboni TA, Maciejewski PK, Block SD, et al. Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences. J Clin Oncol [Internet]. 2008 [cited 2024 Dec 5];26(25):4131–7. Available from: https://pubmed.ncbi.nlm.nih.gov/18757326/ 10.1200/JCO.2007.14.8452 PMC2654372 18757326

45. Perkins HS, Geppert CMA, Gonzales A, Cortez JD, Hazuda HP. Cross-cultural similarities and differences in attitudes about advance care planning. J Gen Intern Med [Internet]. 2002 [cited 2024 Dec 5];17(1):48–57. Available from: https://pubmed.ncbi.nlm.nih.gov/11903775/ 10.1046/j.1525-1497.2002.01032.x PMC1494998 11903775

46. Lenko R, Voepel-Lewis T, Robinson-Lane SG, Silveira MJ, Hoffman GJ. Racial and ethnic differences in informal and formal advance care planning among U.S. older adults. J Aging Health. 2022 Dec 1;34(9–10):1281–90. 10.1177/08982643221104926 PMC9633341 35621163

47. McMahan RD Barnes DE Ritchie CS Jin C Shi Y David D Anxious, depressed, and planning for the future: advance care planning in diverse older adults J Am Geriatr Soc 2020 68 11 2638 2642 10.1111/jgs.16754 32783199 PMC7719568 McMahan RD, Barnes DE, Ritchie CS, Jin C, Shi Y, David D et al (2020) Anxious, depressed, and planning for the future: advance care planning in diverse older adults. J Am Geriatr Soc 68(11):2638–2642 32783199 10.1111/jgs.16754 PMC7719568

48. Arch JJ Bright EE Finkelstein LB Fink RM Mitchell JL Andorsky DJ Anxiety and depression in metastatic cancer: a critical review of negative impacts on advance care planning and end-of-life decision making with practical recommendations JCO Oncol Pract 2023 19 12 1097 1108 10.1200/OP.23.00287 37831973 PMC10732500 Arch JJ, Bright EE, Finkelstein LB, Fink RM, Mitchell JL, Andorsky DJ et al (2023Dec) Anxiety and depression in metastatic cancer: a critical review of negative impacts on advance care planning and end-of-life decision making with practical recommendations. JCO Oncol Pract 19(12):1097–1108 37831973 10.1200/OP.23.00287 PMC10732500

49. Retsinformation.dk. Bekendtgørelse af lov om videnskabsetisk behandling af sundhedsvidenskabelige forskningsprojekter og sundhedsdatavidenskabelige forskningsprojekter [Internet]. LBK nr 1338. Indenrigs- og Sundhedsministeriet; 2020. Available from: https://www.retsinformation.dk/eli/lta/2020/1338