Answers are grounded in this paper and cite their sources.
Objectives Prolonged grief disorder (PGD) affects approximately 10% of bereaved adults worldwide. Structured pre-death risk assessment may help allocate bereavement care resources. A previous study demonstrated good to excellent inter-rater reliability for the German version of the Bereavement Risk Assessment Tool (Multiprofessioneller Fragebogen zur Trauerverarbeitung, MFT). Because reliability alone does not establish accuracy, the present study examined the MFT’s validity against expert-derived criteria. Methods Fifty-five palliative care and hospice professionals in Germany assessed 4 case studies featuring 10 family members and caregivers of palliative care patients, using the 40-item MFT. Participants provided item-by-item responses, an intuitive overall risk level, and an algorithmically calculated risk level derived from their responses. Each index was compared with expert consensus risk levels (percent exact agreement, Krippendorff’s α ); item-level responses were additionally compared with an expert-derived model solution. Results Item-level responses agreed closely with the expert model solution ( M = 89.5% correct). Overall risk-level agreement was weaker: estimated risk levels matched expert consensus in 34.2% of assessments, with negative Krippendorff’s α for most cases ( M α = –0.30). Calculated risk levels agreed more often ( M = 47.6%), though α remained negative for all cases ( M α = −0.18). Estimated and calculated risk levels converged only weakly ( M α = 0.10). Significance of results Participants recognized individual risk factors accurately. Algorithmic classifications corresponded more closely to expert ratings than intuitive judgments, though neither achieved consistently high agreement. RICD-11 eliable item-level recognition need not imply valid overall risk classification. Embedding the scoring algorithm in clinical decision-making and establishing predictive validity against outcomes are promising next steps.
No takes yet. Share an insight, caveat, or question.
Klug et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: