We examine the performance of the Diagnostic and Statistical Manual of Mental Disorders-fifth edition (DSM-5) persistent complex bereavement-related disorder (PCBD) criteria in bereaved adults to identify prolonged grief cases determined prospectively.

Bereaved adults (n = 138) were assessed at 8, 21, 32, 67, and 90 months after the sudden death of a spouse or close relative. We used latent class growth analysis to identify the longitudinal trajectories of grief assessed using the Inventory for Complicated Grief. To validate the trajectory that corresponded to prolonged grief, we examined the baseline predictors of these trajectories and their relationship with functional impairment.

We found three distinct trajectories of grief reactions. One of these trajectories (13.8%) showed high and sustained grief reactions that persisted for almost 7.5 years after the death. Participants with prolonged grief showed greater functional impairment [relative risk ratio (RRR) = 0.82, 95% confidence interval (CI): 0.70 to −0.97; p = 0.02] and higher self-reported depression (RRR = 1.21, 95% CI 1.09 to 1.96; p = 0.001) than participants whose grief reactions subsided over time. The original PCBD (requiring 6 criterion C symptoms) criteria correctly identified cases (57.9–94.7%) with perfect specificity (100%) but low to high sensitivity (5.6–81.3%); however, its sensitivity increased when revising criterion C to require ⩾3 (45.5–94.1%). The dimensional approach showed high sensitivity (0.50–1) and specificity (0.787–0.97).

We recommend revisions to the PCBD criteria, which are overly restrictive and may exclude cases with clinically significant grief-related distress and impairment. In the meantime, clinicians need to monitor grief symptoms over time using available dimensional approaches to reduce the burden of grief.