Adaptation and psychometric validation of the Prolonged Grief Disorder scale among widows in central Nepal

Background: Symptoms of grief vary by culture and societal reactions to death may be gender specific. We aimed to validate a Nepali language version of the Prolonged Grief-13 item scale (PG-13) among widows. Methods: We tested two adapted versions of a Prolonged Grief Disorder (PGD) instrument with 204 Nepali-speaking widows: one was a Nepali translation of the original PG-13 items, while the other contained five additional items derived from qualitative research. We evaluated internal consistency, factor structure, and construct and criterion validity. Results: Participants were on average 44 years old (SD=9.3), completed 6.7 years of school (SD=3.3) and had survived their husbands by 10 years (SD=8.1). Thirteen percent met global criteria for PGD. The removal of one original PG-13 item (felt emotionally numb) from both versions due to poor discriminant validity resulted in 12-and 17-item versions. Exploratory factor analysis supported a one-factor structure for the PG-12 and PG-17. Both versions of the scale exhibited high internal consistency (0.89 and 0.93 respectively). Confirmatory factor analysis suggested that symptoms of PGD were distinct from post-traumatic stress disorder (PTSD), anxiety and depressive symptoms. The PG-12 had lower sensitivity (74.1%) but higher specificity (83.6%) compared to the PG-17 (81.5% and 73.5% respectively). Limitations: Psychosocial counselors' clinical interview global ratings were used as the standard for comparison in criterion validity analyses. Generalizability to other socio-cultural (e.g. non-widowed, low-caste) populations and men in Nepal cannot be assumed. Conclusions: Results indicate satisfactory psychometric properties and validity of both versions of the PG instruments, supporting their use with Nepali speaking widows.