Between 4 November 2021 to 1 December 2021, we executed a cross-sectional online survey targeting three specific groups; (1) those stranded from their partners; (2) those stranded from immediate families; and (3) temporary visa holders unable to migrate or cross international borders. We collected data on respondents’ demographic profile; the nature of COVID-19-related travel impacts; depression, anxiety, and stress levels (using the validated DASS-21); and finally, data on respondents financial, employment and accommodation situation.
Ethical approval for this study was granted by the UNSW Human Research Ethics Committee (#210418). All respondents indicated their consent to participate.
Descriptive analysis involved the calculation of means, standard deviations, confidence intervals and standard errors for continuous variable data, and the calculation of counts and proportions for categorical variables data. To test significance, we used one-way ANOVA analyses (for continuous variables) and Chi-square analyses or a Fishers exact test for smaller cell sizes (for categorical variables). Any variables found to show a significant association with DASS severity (p<.2) was chosen to be included in the model as predictor variables. We dichotomised DASS scores to indicate either no-mild symptom severity or moderate-to-extremely severe symptom severity. Finally, we conducted multivariable binary logistic regression to determine demographic, financial, and occupational predictors of moderate to extremely severe depression, anxiety, and stress. All analyses were conducted with SPSS (22). No errors, influential outliers or multicollinearity amongst variables was identified. Statistical significance was p > 0.05.