To Identify Relatives at Risk for Prolonged Grief Disorder Symptomatology: A Cross-Sectional Study.

Abstract

While most bereaved can navigate a death-related loss, some face persistent and disabling challenges, a condition known as Prolonged Grief Disorder (PGD). Early identification of PGD can prevent adverse health outcomes. The effectiveness of healthcare providers' clinical intuition versus relatives' self-report surveys in estimating the risk of developing PGD remains uncertain. The study evaluates the accuracy of healthcare providers' subjective risk estimates and self-reported surveys from relatives, conducted close to the cancer patient's death, in predicting ICD-11 PGD symptomatology seven months postloss. The study involved specialist palliative care healthcare providers and bereaved relatives of oncology patients in ten hospital and five home settings. Subjective risk estimates from healthcare providers (N = 75) and relatives' self-report surveys (N = 75) using the PreLoss Grief-12 (PG-12) were compared with Traumatic Grief Inventory-Self Report Plus (TGI-SR+) total scores seven months postloss. Diagnostic efficacy was evaluated using the Receiver Operating Characteristic and Area Under Curve. In the sample of 75 relatives, 8% had ICD-11 PGD symptomatology. The Receiver Operating Characteristic analysis of the PG-12 (N = 75) yielded an Area Under Curve of 0.891, 95% CI [0.782-1], while the healthcare providers' subjective estimates (N = 75) resulted in an Area Under Curve of 0.549, 95% CI [0.299-0.799]. The optimal PreLoss Grief-12 cut-off score for identifying relatives at risk for PGD was 30 with a sensitivity of 100% and a specificity of 64.7%. The PG-12 accurately predicted PGD symptomatology seven months postloss, while clinical intuition did not perform better than chance. Future research should compare the predictive value of self-report data with clinical interviews.