1 Montreal Behavioural Medicine Centre, Montreal North Island Integrated Health and Social Services University Centre (CIUSSS- NIM), Montréal, H3G 1M8, Canada.
2 Department of Psychology, Concordia University, Montréal, Canada.
3 Department of Psychology, University of Quebec at Montreal (UQAM), Montréal, Canada.
4 Montreal Behavioural Medicine Centre, Montreal North Island Integrated Health and Social Services University Centre (CIUSSS- NIM), Montréal, H3G 1M8, Canada. simon.bacon@concordia.ca.
5 Department of Health, Kinesiology & Applied Physiology, Concordia University, Montréal, Canada. simon.bacon@concordia.ca.

Purpose: Depression is a multifaceted psychiatric condition that has been associated with suboptimal weight loss following bariatric surgery. Previous variations of Beck Depression Inventory-Second Edition (BDI-II) subscales been proposed, including those identified within bariatric populations; however, it is unclear whether the BDI-II items contained within these subscales would remain consistent after the procedure considering the physical and lifestyle changes that occur following bariatric surgery.

Materials and methods: A two-step analytic approach that comprised of exploratory (EFA) and confirmatory factor analyses (CFA) that aimed to identify a stable factor structure using pre- and 6-month post-surgical BDI-II responses. Baseline BDI-II responses of 149 patients (Group 1) were used to identify an initial EFA model. Five BDI-II models underwent CFA using BDI-II responses of a comparable pre-surgical group (Group 2; n = 142), and 6-month post-surgical data from Group 1.

Results: EFA generated a two-factor solution. Of the five CFA models performed, the three-factor model that was initially identified by Hayes (2015) among patients undergoing bariatric surgery demonstrated superior fit across time and between groups.

Conclusions: Although the EFA initially identified a two-factor model, CFA determined that a previously defined three-factor model reliably fit both pre- and post-surgical BDI-II responses. This study supports using the Hayes (2015) subscales when monitoring pre- and post-bariatric surgery facets of depression specific to this population. Being able to accurately and reliably monitor depressive symptoms of patients that are undergoing bariatric surgery will allow for the provision and monitoring of targeted interventions aimed at improving their mental and physical health outcomes.