AUTHOR=Vaccarino Anthony L. , Black Sandra E. , Gilbert Evans Susan , Frey Benicio N. , Javadi Mojib , Kennedy Sidney H. , Lam Benjamin , Lam Raymond W. , Lasalandra Bianca , Martens Emily , Masellis Mario , Milev Roumen , Mitchell Sara , Munoz Douglas P. , Sparks Alana , Swartz Richard H. , Tan Brian , Uher Rudolf , Evans Kenneth R. 

TITLE=Rasch analyses of the Quick Inventory of Depressive Symptomatology Self-Report in neurodegenerative and major depressive disorders

JOURNAL=Frontiers in Psychiatry

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2023.1154519

DOI=10.3389/fpsyt.2023.1154519

ISSN=1664-0640

ABSTRACT=Background: Symptoms of depression are present in neurodegenerative disorders (ND). It is important that depression-related symptoms be adequately screened and monitored in persons living with ND.

Methods: De-identified data from the Ontario Neurodegenerative Disease Research Initiative (NCT04104373) and Canadian Biomarker Integration Network in Depression (NCT01655706) were used in the analyses. Five hundred and twenty participants with ND (Alzheimer’s disease or mild cognitive impairment, amyotrophic lateral sclerosis, cerebrovascular disease, frontotemporal dementia and Parkinson’s disease) and 117 participants with major depressive disorder (MDD) were administered the QIDS-SR. Rasch Measurement Theory was used to assess measurement properties of the QIDS-SR, including item-level fit, category ordering, item targeting, person separation index and reliability and differential item functioning.

Results: The QIDS-SR fit well to the Rasch model in ND and MDD, including unidimensionality, satisfactory category ordering and goodness-of-fit. Item-person locations (Wright maps) showed gaps in item difficulties, suggesting poor precision for persons falling between those severity levels. Differences between mean person and item locations in the ND cohort (-1.89 logits) suggest that QIDS-SR items target more severe depression than experienced by the ND cohort. Some items showed differential item functioning between cohorts.

Conclusions: The present study supports the use of the QIDS-SR in MDD and suggest that the QIDS-SR can be also used to screen for depressive symptoms in persons with ND. Rasch analysis highlighted areas of scale improvement, including gaps in item targeting. Future studies would benefit from examination in a more severely depressed ND cohort, including those with diagnosed clinical depression.