Validation of the Traumatic Grief Inventory – Self-Report (TGI-SR) in a population of Dutch adults with psychotrauma related problems

Abstract

This study determined the reliability, content and construct validity of the Traumatic Grief Inventory – Self-Report (TGI-SR) in a sample of patients with psychotrauma related problems (age 18-75 years, n = 364). Between October 2013 and May 2014, assessment took place with the BSI-53, WHOQOL-BREF, HTQ-16 and the TGI-SR at Foundation Centrum '45 in the Netherlands. Cronbach’s α showed that the TGI-SR has a good internal consistency (.96). A principal component analysis (PCA) produced a two-factor solution. Pearson correlation showed that the TGI-SR is positively correlated with the BSI-53 (r = .58, p < .01) and the HTQ-16 (r = .55, p < .01). The following domains from the WHOQOL-BREF were negatively correlated with the TGI-SR: physical health (r = -.46, p < .01), social relationships (r = -.33, p < .01), environment (r = -.42, p < .01), psychological health (r = -.42, p < .01) and the general quality of health (r = -.38, p < .01). Post hoc comparisons (Tukey HSD) showed that refugees and asylum seekers had a higher score (p < .05) on the TGI-SR compared to veterans, first and second generation survivors of WWII, public service employees and people traumatised by other events. Results indicate that the TGI-SR is a reliable instrument and has good construct validity. Despite the fact that the items were not loading on the factors as predicted, the TGI-SR has good content validity as all items add substantially to the TGI-SR. Limitations and suggestions for further research are proposed in this study.