A Systematic Review of Dimensional and Categorical Models of Grief Related Disorders

Abstract

Since the inclusion of Prolonged Grief Disorder (PGD) in the ICD-11 and DSM-5-TR, many studies have employed factor analytic and mixture modeling approaches to explore whether PGD symptoms are best explained by a single dimension, multiple factors, or distinct subgroups. However, no review to date has systematically synthesized these findings. A pre-registered systematic review of psychometric studies of PGD symptoms was conducted by searching Web of Science, PubMed, Scopus, and PsycINFO. Two researchers independently identified eligible studies that met inclusion and exclusion criteria, and methodological rigor was evaluated using a refined COSMIN risk of bias checklist. The retained studies were grouped into four categories: (1) single-factor validation studies, (2) latent structures and dimensional models, (3) diagnostic criteria evaluations, and (4) longitudinal trajectory analyses. Across studies, 10 grief measures were used, with 89% relying on three legacy instruments (ICG, PG-13, TGI-SR) developed pre ICD-11/DSM-5-TR and not consistently implementing current symptom and impairment requirements. Evidence more consistently supported PGD as a graded, multidimensional construct, with class or trajectory findings best interpreted as pragmatic risk strata along an underlying continuum. Methodological quality varied, with only 25% rated very good, alongside sparse longitudinal work and uneven cross-cultural coverage. A criteria-aligned instrument such as the TGI-SR+ that implements ICD-11 and DSM-5-TR algorithms and explicit caseness rules is the strongest current candidate to anchor cross-study comparability and clinical assessment. With a shared standard, dimensional and categorical models, including factor mixture approaches, can test whether categorical distinctions add value beyond continuous symptom severity.