Table 5.
Mental health impacts of COVID-19 and household characteristics at South African Population Research Infrastructure Network nodes from April to August 2020a.
| Variable | Individual PHQ-2b screened positive (n=17,257), prevalence ratio (95% CI) | Individual GAD-2c screened positive (n=17,256), prevalence ratio (95% CI) | |||
| Node | |||||
|
|
AHRId (KwaZulu-Natal) | 1.00 (N/Ae) | 1.00 (N/A) | ||
|
|
Agincourt (Mpumalanga) | 0.35 (0.30-0.41) | 0.46 (0.39-0.54) | ||
|
|
DIMAMOf (Limpopo) | 1.16 (0.93-1.44) | 2.83 (2.23-3.60) | ||
| Month of interview | |||||
|
|
April | 1.34 (0.87-2.05) | 2.15 (1.55-2.99) | ||
|
|
May | 1.00 (N/A) | 1.00 (N/A) | ||
|
|
June | 0.37 (0.27-0.50) | 0.72 (0.57-0.91) | ||
|
|
July | 0.93 (0.76-1.14) | 0.61 (0.50-0.75) | ||
|
|
August | 0.59 (0.47-0.74) | 0.41 (0.33-0.51) | ||
|
|
September | 0.56 (0.44-0.71) | 0.25 (0.19-0.33) | ||
|
|
October | 0.34 (0.26-0.46) | 0.06 (0.04-0.10) | ||
|
|
November | 0.60 (0.47-0.77) | 0.32 (0.25-0.43) | ||
|
|
December | 0.88 (0.67-1.16) | 0.63 (0.47-0.84) | ||
| Household members (per person) | |||||
|
|
Children | 1.01 (0.97-1.05) | 1.03 (0.98-1.07) | ||
|
|
Working-aged adults | 1.03 (0.99-1.07) | 1.01 (0.97-1.05) | ||
|
|
Pension-aged adults | 0.88 (0.76-1.02) | 0.91 (0.78-1.07) | ||
| Maximum education of household members | |||||
|
|
Less than complete secondary | 1.00 (N/A) | 1.00 (N/A) | ||
|
|
Complete secondary | 1.07 (0.91-1.27) | 0.89 (0.75-1.07) | ||
|
|
Diploma/certificate/degree | 1.33 (1.08-1.65) | 1.16 (0.92-1.46) | ||
| Household income sources in 2019 | |||||
|
|
Full/part-time employees | 0.97 (0.92-1.02) | 0.98 (0.93-1.03) | ||
|
|
Pension grants | 1.11 (0.95-1.29) | 0.99 (0.84-1.16) | ||
|
|
Non-pension grants | 0.98 (0.94-1.02) | 0.99 (0.95-1.03) | ||
|
|
Household asset index quintiles | 1.01 (0.96-1.06) | 1.06 (1.00-1.11) | ||
aEach column (knowledge, the number of visitors, and members leaving home) is on a single Poisson regression with household-level random effects and robust SEs for calculating prevalence ratios and 95% CIs. In total, 17,384 observations have complete covariate data and 5704 observations from 1475 households have missing covariate values; 33 households surveyed by AHRI, 103 surveyed by Agincourt, and 1339 households surveyed by DIMAMO are missing data. The remaining missing observations reflect missing outcome values.
bPHQ-2: Patient Health Questionnaire 2-item.
cGAD-2: Generalized Anxiety Disorder 2-item.
dAHRI: Africa Health Research Institute.
eN/A: not applicable.
fDIMAMO: Dikgale-Mamabolo-Mothiba