Table 3

Relation between assets, COVID-19 financial stressor exposure and probable depression among US adults

 Model 1 Model 2 Model 3 Model 4 OR (95% CI) OR (95% CI) OR (95% CI) Beta OR (95% CI) Assets Low 3.0 (2.1 to 4.2) 2.6 (1.8 to 3.7) 1.4 n/a High – – – – COVID-19 financial stressors Low – – – – High 2.8 (2.1 to 3.9) 2.3 (1.6 to 3.2) 1.4 n/a Assets*COVID-19 financial stressor indicator −1 n/a Low assets, low stressors 6.4 (4.0 to 10.3) Low assets, low stressors 4.3 (2.6 to 6.9) High assets, high stressors 4.2 (2.5 to 7.2) High assets, low stressors 1
• Data: COVID-19 and Life stressors Impact on Mental Health and Well-being study collected from 31 March 2020 through 13 April 2020 (n=1441). Complete case analysis used for multiple logistic regression resulting in n=1327 for this model. Nationally representative of US adults ages 18 years and older.

• All models controlled for age, race/ethnicity, sex and household size.

• Probable depression defined as Patient Health Questionnaire-9 score cut-off of ≥10.

• Interaction indicator term: high or low assets*COVID-19 financial stressor exposure. High assets defined by having three or more of the following: household income (above $45 000), household savings (above$5000), home ownership, education (college or more) and being married. High COVID-19 financial stressor exposure defined by experiencing one or more of the following COVID-19 stressors: losing a job, having a family member lose a job, having financial difficulties and having difficulty paying rent. Low COVID-19 financial stressor exposure refers to the absence of endorsement of COVID-19-related financial stressor exposure.

• Model 1 shows the odds of depression with no assets (not controlling for stressors). Model 2 shows the odds of depression with stressors (not controlling for assets). Model 3 shows the odds of depression with stressors and assets (no interaction term). Model 4 shows the odds of depression with stressors and assets and an interaction term between the two.

• OR not given for variables where interactions are present since ORs are uninterpretable. Model parameters used to determine the relative odds of the relation between specific variables and depression.

• n/a, not applicable.