Introduction
Bereavement support for families is not ancillary to oncology; it is embedded in the ethical and professional mandate of comprehensive cancer care and is articulated in contemporary guidance and training standards.
The American Society of Clinical Oncology (ASCO) recommends early, systematic integration of palliative care principles, including family- and caregiver-oriented support, across the cancer trajectory, with implications for competencies expected of oncology teams (Sanders et al. Reference Sanders, Temin and Ghoshal2024).
Complementing practice guidance, the European Society of Medical Oncology (ESMO) and ASCO Global Curriculum define communication, supportive and palliative care, and professionalism as core domains for medical oncology training and include end-of-life and grief-related competencies among expected residency outcomes (Cufer and Kosty Reference Cufer and Kosty2023).
Together, these frameworks set a clear expectation: programs should deliberately prepare trainees to engage with families before and after a patient’s death, recognizing bereavement as routine care rather than a discretionary add-on (Cufer and Kosty Reference Cufer and Kosty2023; Sanders et al. Reference Sanders, Temin and Ghoshal2024).
Despite these expectations, bereavement-specific education remains variably embedded across medical training. A scoping review of grief support training across undergraduate, postgraduate, and continuing professional development describes a fragmented landscape characterized by heterogeneous objectives, inconsistent content, and limited outcome assessment, with many initiatives outside oncology (Sikstrom et al. Reference Sikstrom, Saikaly and Ferguson2019). Within oncology, gaps persist at the residency level, even as repeated grief exposure is linked to moral distress, compassion fatigue, and burnout, making preparedness a workforce issue as well as an educational one (Granek et al. Reference Granek, Tozer and Mazzotta2012; Hlubocky et al. Reference Hlubocky, Symington and McFarland2021).
The educational sciences offer a pragmatic path forward: serious illness communication skills are teachable, and training changes behavior. In oncology and internal medicine trainees, the Oncotalk program and a randomized trial in residents and nurse practitioners demonstrated improvements in observable communication behaviors and better patient and family reported outcomes after structured, feedback rich curricula (Back et al. Reference Back, Arnold and Baile2007; Curtis et al. Reference Curtis, Back and Ford2013), Compact oncology tailored curricula have also shown benefit, including the PALLI-COM trial that improved oncologists’ competencies for early palliative and end of life discussions, as well as experience based curricula in radiation oncology residents that improved directly observed skills, both pointing to scalable formats that fit busy training environments (Harnischfeger et al. Reference Harnischfeger, Rath and Alt-Epping2022; Christensen et al. Reference Christensen, Kumar and Wang2023). These findings resonate with faculty development and implementation work emphasizing coached practice, deliberate feedback, and longitudinal reinforcement as the ingredients most consistently associated with durable behavior change (Eysenbach Reference Eysenbach2004; Back et al. Reference Back, Fromme and Meier2019).
Yet it remains less well characterized, especially on a global scale, how often bereavement-specific content is present in oncology residency curricula, how prepared residents feel to support grieving caregivers, and how education relates to emotional sequelae that may influence workforce sustainability. This cross-sectional 25-country survey addresses that evidence gap by quantifying the prevalence of formal bereavement education, describing self-perceived preparedness and emotional impact, and testing the association between education and preparedness, with the goal of informing pragmatic, competency-based curricular strategies that programs can implement now (Back et al. Reference Back, Arnold and Baile2007; Curtis et al. Reference Curtis, Back and Ford2013; Christensen et al. Reference Christensen, Kumar and Wang2023; Cufer and Kosty Reference Cufer and Kosty2023; Sanders et al. Reference Sanders, Temin and Ghoshal2024).
Materials and methods
Study design and setting
We conducted an anonymous, web-based cross-sectional survey of medical oncology residents and fellows across 25 countries. The questionnaire was developed and delivered as a secure, web-based instrument using Google Forms (Google LLC) configured to avoid capturing directly identifying information, such as names, emails, or IP addresses. Web-based surveys are widely used in health profession research when appropriate safeguards and transparent reporting standards such as the CHERRIES checklist are applied (Eysenbach Reference Eysenbach2004).
Recruitment, eligibility, and sampling
To preferentially reach eligible participants and minimize ineligible responses, the survey link was disseminated through international oncology working groups and trainee networks, including national young oncologist groups, ESMO and ASCO affiliated trainee channels, and institutional residency program listservs. On the landing page, respondents self-attested that they were current medical oncology residents or fellows; responses without attestation were excluded. Because dissemination occurred through multiple professional channels and listservs, a response rate could not be calculated.
Measures
The instrument captured: demographics and training context (age, gender, country, years in training, clinical setting); presence of formal bereavement education during residency (yes or no); frequency of interactions with grieving caregivers (Rarely, Occasionally, Often, Always); self-perceived preparedness to support grieving caregivers (Not at all, Not very, Somewhat, Very); multi select challenges (emotional fatigue, fear of overstepping boundaries, lack of training, time constraints, other free text); emotional impact of patient death (Not, Rarely, Occasionally, Deeply affected); consideration of leaving residency due to emotional demands (No, briefly, seriously); and endorsement of bereavement education as integral to training (yes or no).
Outcomes
The primary outcome was self-perceived preparedness (4-level ordinal). The key independent variable was formal bereavement education (yes/no). Secondary outcomes were training prevalence; distribution of challenges and emotional impact; and consideration of leaving residency.
Statistical analysis
We summarized data using descriptive statistics. The association between training and the 4-level preparedness outcome used χ 2 (df = 3). A binary contrast (≥ “Somewhat prepared” vs lower) was analyzed by 2 × 2 tables with odds ratio (OR) and Wald 95% CI. Correlation between caregiver-interaction frequency (coded 1–4) and preparedness (1–4) used Pearson’s r. Two-sided p < 0.05 indicated statistical significance. Analyses used complete cases unless otherwise specified.
Ethical considerations
This study surveyed health care professionals using an anonymous questionnaire and collected no patient data and no directly identifying information. Per coordinating center policy, the study met criteria for exemption or did not constitute human subjects research; where required, local determinations or waivers were obtained. Completion of the survey constituted informed consent.
Results
Respondents and training context
We analyzed 165 oncology residents from 25 countries. Female participants comprised 91 of 165 (55.2%), and 79 of 165 (47.9%) worked in academic or university hospitals (Table 1). Seniority was 5 years or more for 73 of 165 (44.2%). Countries with more than one respondent included Italy (n = 90), India (n = 28), the United Kingdom (n = 9), Pakistan (n = 6), Kenya (n = 5), Romania (n = 4), Saudi Arabia (n = 3), Türkiye (n = 2), and Portugal (n = 2); 16 additional countries contributed one respondent each (Table 2).
Participant characteristics (N = 165)
Table 1 Long description
Participant characteristics are reported as counts under N and percentages under percent for age group, sex, training years, and setting. Age 20 to 25 includes 3 participants, 1.8 percent; age 26 to 30 includes 55, 33.3 percent; age 31 to 35 includes 49, 29.7 percent; age 36 to 40 includes 29, 17.6 percent; and age 41 or more includes 29, 17.6 percent. Female includes 91 participants, 55.2 percent, and male includes 74, 44.8 percent. Training years less than 1 includes 8 participants, 4.8 percent; 1 to 2 includes 14, 8.5 percent; 3 to 4 includes 70, 42.4 percent; and 5 or more includes 73, 44.2 percent. Setting, Academic or University includes 79 participants, 47.9 percent; Public hospital includes 61, 37.0 percent; and Private hospital includes 25, 15.2 percent.
Countries represented and formal bereavement training during residency
Table 2 Long description
Country counts are listed with N and the number and percent reporting formal training, labeled, Formal training, n over N percent. Italy has N 90 with 16 reporting training, shown as 16 of 90 and 17.8 percent. India has N 28 with 3 of 28 and 10.7 percent; the United Kingdom has N 9 with 3 of 9 and 33.3 percent; Pakistan has N 6 with 1 of 6 and 16.7 percent; Kenya has N 5 with 1 of 5 and 20.0 percent; Romania has N 4 with 1 of 4 and 25.0 percent. Saudi Arabia has N 3 with 0 of 3 and 0.0 percent; Turkiye has N 2 with 0 of 2 and 0.0 percent; Portugal has N 2 with 1 of 2 and 50.0 percent; France has N 1 with 1 of 1 and 100.0 percent. Other single respondent countries marked with an asterisk have N 15 with 1 of 15 and 6.7 percent. The total row reports N 165 with 28 of 165 and 17.0 percent.
a Other single-respondent countries (n = 15): Australia, Bosnia and Herzegovina, Brazil, Egypt, Georgia, Germany, Ghana, Iraq, Ireland, Malaysia, Mexico, Norway, Russia, Ukraine, and Yemen.
Prevalence of formal bereavement education
Only 28 of 165 respondents (17.0%) reported formal bereavement education during residency, with cross-country variability (Table 2).
Preparedness and association with training
Preparedness distribution was: Somewhat prepared 80 (48.5%), Very prepared 8 (4.8%), Not very prepared 67 (40.6%), and Not at all prepared 10 (6.1%). Formal training was associated with higher preparedness (chi-square = 13.03; df = 3; p = 0.0046). In the binary contrast, training increased the odds of being at least “Somewhat prepared” (OR 3.94; 95% CI, 1.51–10.33) (Table 3).
Self-perceived preparedness to support grieving caregivers by training status
Table 3 Long description
Preparedness levels are reported for two groups, no training with 137 people and training with 28 people, using counts and percentages. In the no training group, not at all prepared is 10 people at 7.3 percent, not very prepared is 61 at 44.5 percent, somewhat prepared is 62 at 45.3 percent, and very prepared is 4 at 2.9 percent. In the training group, not at all prepared is 0 at 0.0 percent, not very prepared is 6 at 21.4 percent, somewhat prepared is 18 at 64.3 percent, and very prepared is 4 at 14.3 percent. Not very prepared is higher in the no training group than in the training group. Very prepared is higher in the training group than in the no training group. Preparedness is self reported.
a Preparedness is self-reported.
Exposure to grieving caregivers and preparedness
Exposure to bereavement situations was frequent: Always 25 (15.2%), Often 66 (40.0%), Occasionally 49 (29.7%), and Rarely 25 (15.2%). Interaction frequency correlated weakly with preparedness (r = 0.15), suggesting exposure alone is insufficient without structured training (Table 4).
Frequency of interactions with grieving caregivers (N = 165)
Table 4 Long description
Interaction frequency is reported with counts under N and percentages under percent, across four categories. Always includes 25 caregivers, 15.2 percent. Occasionally includes 49 caregivers, 29.7 percent. Rarely includes 25 caregivers, 15.2 percent. Always and rarely have the same count and percentage.
Correlation analyses treat ordinal categories as ordered (1–4) and are descriptive.
Challenges and emotional impact
The most frequent challenges were emotional fatigue (107 of 165; 64.8%), fear of overstepping professional boundaries (93 of 165; 56.4%), lack of training (88 of 165; 53.3%), and time constraints (83 of 165; 50.3%) (Table 5). Emotional impact was substantial: deeply affected 45 (27.3%), occasionally affected 90 (54.5%), rarely affected 28 (17.0%), and not affected 2 (1.2%). Consideration of leaving residency due to emotional demands was reported by 59 of 161 (36.6%), with 4 missing responses (Table 6). Nearly all respondents (164 of 165; 99.4%) endorsed integrating bereavement education into oncology training.
Most challenging aspects of bereavement care (N = 165)
Table 5 Long description
Responses are grouped by challenging aspect, with columns for N and percent. Fear of overstepping professional boundaries has N 93 and 56.4 percent. Lack of training has N 88 and 53.3 percent. Time constraints has N 83 and 50.3 percent. Other, free text has N 3 and 1.8 percent. Multiple responses were allowed, so percentages can add to more than 100 percent.
Multiple responses allowed; percentages may exceed 100%.
Emotional impact of patient death and consideration of leaving residency
Table 6 Long description
Outcomes are reported with columns for N and percent for emotional impact of patient death and for considering leaving residency due to emotional demands. Emotional impact, deeply affected: N 45, 27.3 percent. Emotional impact, occasionally affected: N 90, 54.5 percent. Emotional impact, rarely affected: N 28, 17.0 percent. Emotional impact, not affected: N 2, 1.2 percent. Considered leaving residency due to emotional demands, yes, seriously considered: N 21, 13.0 percent. Considered leaving residency due to emotional demands, yes, briefly considered: N 38, 23.6 percent. Considered leaving residency due to emotional demands, no: N 102, 63.4 percent. Considered leaving residency due to emotional demands, missing: N 4, percent not reported.
Emotional impact percentages are based on N = 165. Consideration of leaving percentages is based on N = 161 (4 missing responses).
Discussion
In this multinational cohort, formal bereavement education during oncology training was uncommon, and half of respondents reported being not very or not at all prepared to support grieving caregivers. Importantly, training was associated with higher perceived preparedness, whereas exposure frequency alone correlated only weakly with preparedness, suggesting that experience without structure may be insufficient.
These findings have immediate implications for training. Programs can act now by mapping curricula to ESMO and ASCO Global Curriculum domains and by making bereavement competencies explicit, including distinguishing normal from prolonged grief, recognizing risk factors, structuring condolence and follow-up, navigating boundaries and referral pathways, and supporting clinician self-care and mentorship (Cufer and Kosty Reference Cufer and Kosty2023). Teaching should be experiential, with short iterative workshops that use role play and coached feedback, embedded in rotations and reinforced through case-based debriefings, approaches most consistently associated with behavior change (Back et al. Reference Back, Arnold and Baile2007, Reference Back, Fromme and Meier2019; Curtis et al. Reference Curtis, Back and Ford2013; Christensen et al. Reference Christensen, Kumar and Wang2023). These educational efforts should align with palliative care guideline recommendations for routine, team-based, family-oriented care at the end of life (Sanders et al. Reference Sanders, Temin and Ghoshal2024).
Strengths of this study include a multinational sample spanning 25 countries, a focus on bereavement-specific rather than generic communication skills, and convergence across indicators of training prevalence, preparedness, and emotional sequelae. Limitations include convenience sampling via professional networks with overrepresentation of some regions, reliance on self-reported cross-sectional measures without external validation of preparedness, inability to calculate a response rate, potential unmeasured confounding, and absence of downstream outcomes such as caregiver satisfaction, documented condolence practices, or clinician retention. These constraints support prospective implementation studies that embed brief, coached curricula and measure behavioral outcomes and system outcomes.
If oncology is to be sustainably practiced, bereavement care must be taught deliberately rather than absorbed incidentally. The curricular content is well described; the remaining gap lies in delivery and evaluation.
Conclusions
Bereavement care is a routine component of oncology practice, yet in this 25-country survey most residents reported no formal training and half felt insufficiently prepared. Formal education was associated with higher preparedness, while exposure alone showed only a weak relationship. These findings support embedding brief, skills-based bereavement curricula integrated into existing communication and palliative care training, reinforced by mentorship and reflective debriefing, and evaluating implementation strategies with outcomes that include clinician well-being and retention, as well as the quality and consistency of caregiver support.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1478951526103587.
Data availability statement
De-identified data and analysis code are available upon reasonable request to the corresponding author, subject to institutional policies and data-sharing agreements. The complete survey instrument is provided at the following link: https://forms.gle/h17r49gNUPnzB7jy5.
Acknowledgments
We thank the oncology residents and program leaders who shared and completed the survey.
Funding
None.
Open access
