Impact of pre-existing depression and food insecurity on the trajectory of depressive symptomatology during the COVID-19 pandemic outbreak in South Africa: A panel analysis of nationally representative South African data

We investigated the trajectory of depressive symptoms (“depression”) from the start of the COVID-19 pandemic in South Africa (March 2020) until 2021, between individuals with and without pre-pandemic depression, specifically regarding the role of food security. Our investigation used publicly available panel data (N = 6,930) from the South African National Income Dynamics Study Coronavirus Rapid Mobile Survey (SA-NIDS-CRAM from 2020–2021) on those who had also participated in the pre-pandemic South African National Income Dynamics Study (SA-NIDS, 2017) depression interview. We investigated trends in depressive symptomatology (based on a 2-item Patient Health Questionnaire) at SA-NIDS-CRAM Wave 2 (July 2020), Wave 3 (February 2021) and Wave 5 (May 2021). Generalized estimating equations (GEE) with post-estimation linear combinations of estimators were fitted to investigate the roles of pre-pandemic depression (based on 2017 SA-NIDS data) and food insecurity during the pandemic on depressive symptomatology. During the pandemic, the highest levels of depression were observed consistently among those with pre-pandemic depression and food insecurity; and were lowest among those without pre-pandemic depression and food security. Depressive symptomatology rose in nearly equal magnitude during the early phases of the pandemic in two groups: those without pre-pandemic depression but food insecure during the pandemic; as well as those with pre-pandemic depression but food secure during the pandemic. However, this dynamic changed later in the pandemic, when higher depressive symptomatology was observed in the group with both pre-pandemic depression and food insecurity, widening the gap between them from Wave 3 (adj β = 0.63, p