Data was collected twice among the same food markets and HHs during enforcing and partial lifting of COVID-19 lockdown. A total of 65 and 62 staple food sellers in food markets closet to slum and non-slum HHs, respectively participated in providing information on food prices during the study. A total of 208 and 202 slum and non-slum HHs respectively participated in data collection during COVID-19 lockdown. However, 3 of 208 and 2 of 202 slum and non-slum HHs respectively were lost to follow-up after COVID-19 lockdown was partially lifted. Therefore, data for 205 and 200 slum and non-slum HHs were available for analysis during the period of enforcing and partial lifting of the COVID-19 lockdown.
Socio-demographic characteristic of the study sample
A total of 127 food sellers (65 and 62 from HHs closest to slum and non-slum HHs respectively) participated in the study. Out of the 127 food sellers, 75 (59%) and 52 (41%) were females and males, respectively. Furthermore, a total of 405 HH heads participated in the HHFI assessment, of which 205 (50.6%) and 200 (49.4%) HHs were slum and non-slum residents. Out of the 405 HH heads, 185 (45.7%) were female headed. The mean age ± standard deviation (SD) of HH heads was 40 years ± 11 SD. Table 2 shows the social demographic characteristics of HHs that participated in the HHFI assessment.
The average family size was 5, with a range of 1 to 9. Out of 405 HH heads the majority, 176 (43.5%) were daily wage earners, whilst 96 (23.7%) and 133 (32.8%) were unemployed and employed, respectively. Out of the 176 daily wage earners, 55 (31.3), 32 (18.2%), 77(43.8%), 4 (2.3%) and 8 (4.5%) were boda-boda (motorbike) transport riders, public passenger tax drivers, bar/hotel maids, sex workers and food vendors, respectively. Of the 205 slum HHs, 179 HHs had received food aid of staple common beans and maize flour to sustain them for one month during the period of enforcing the COVID-19 lockdown. Figure 1 shows Uganda’s military force distributing food and a woman receiving food aid in a slum located in urban Uganda.
Staple food prices before, during enforcing and partial lifting of COVID-19 lockdown
Maize flour and common beans are major staple foods in Uganda. The mean price of maize flour significantly increased from 2030 Uganda shillings (Ug Shs) before COVID 19 lockdown to 2383 Ug Shs during COVID19 lockdown (p <0.00001), an increase equivalent to 17.4%. However, the mean price of maize flour significantly (p <0.00001) reduced from 2383 Ug Shs during lockdown to 1823 Ug Shs during partial lifting of the lockdown, a reduction rate equivalent to 23.5%. Currently 1 United States Dollar is equivalent to 3719 Ug Shs. Figure 2 shows staple food prices before, during enforcing and partial lifting of COVID-19 lockdown lock down in Nansana municiparity, Uganda .
The price of common beans significantly (p<0.00001) increased from 2558 Ug Shs before COVID-19 lockdown to 4333Ug Shs during lockdown, an increase of 69.4%. However, on partial lifting of the lockdown, the common bean price significantly (p<0.00001) reduced from 4333 to 2847 Ug Shs, a reduction rate of 34.3%.
Association between socio-demographic characteristics and food consumption scores during COVID-19 lockdown and partial lifting of COVID-19 lockdown in Nansana Municipality, Uganda
The mean food consumption scores (FCS) during the lockdown were at borderline (22.8-22.9) for either slum (22.8) or non-slum (22.9) HHs and were not significantly different from each other (p=0.06). However, during partial lifting of the lockdown, FCS among slum HHs were significantly lower at 20.1 (poor) compared to non-slum HHs at 22.7 (borderline) (p=0.01). During lockdown, both female and male headed HHs had borderline FCS. However, mean FCS was significantly lower (p<0.00001) in female HHs (21.7) compared to male headed HHs (24.7). In contrast, when the COVID-19 lockdown was partially lifted, mean FCS for female headed HHS were poor (20.3) and significantly lower (p<0.00001) compared to borderline mean FCS (23.4) for male headed HHs. Mean FCS were significantly lower among HH heads who were unemployed and daily wage-earners compared to the employed both during the COVID-19 lockdown and partial lifting of the lockdown (p<0.05). Table 3 shows the association between socio-demographic characteristics and mean FCS during COVID-19 lockdown and during partial lifting of COVID-19 lockdown in Nansana municipality, Wakiso district, Uganda.
Household heads with advanced level education and above had significantly higher mean FCS (p<0.05) compared to HH heads who were uneducated, had primary and secondary education both during lockdown and partial lifting of the lockdown. Household heads who were tenants had significantly lower (p< 0.05) FCS during lockdown (21.0) and during partial lifting of lockdown (20.9) compared to owners of houses at FCS of 22.9 and 22.7, respectively. Food aid distribution was only done among slum HHs during the COVID-19 lockdown. The mean FCS was significantly higher (p<0.0001) at borderline (24.5) among HHs that received food aid compared to poor FCS (18.2) among HHs that did not receive food aid.
Food insecurity status and its association with social inequalities during COVID-19 lockdown in Nansana municipality, Wakiso district Uganda
A binary outcome of food insecure and food secure HHs was created to assesses the social demographic characteristics of food insecurity during lockdown and partial lifting of local down. Table 4 shows bivariate and multivariate logistic regression analysis of factors associated with food insecurity during COVID-19 lockdown.
Households were food insecure and food secure if their FCS were 0 to < 35 (poor to borderline) and FCS > 35 (acceptable), respectively. In bivariate logistic analysis, HH heads who were daily wage earners and unemployed were 9.2 times (95% CI: 21-39.9) and 7.6 times (95% CI: 1.02-57.3) at risk to HHFI, respectively compared to the employed HH heads. In contrast, employment of HH heads prevented against food insecurity by 0.06 times (95% CI: 0.002-0.2) compared to wage earners and the unemployed HH heads. All other variables such as food aid distribution, size of HH members, age, place of residence, sex, house of ownership of HH head were not associated with food insecurity during the COVID-19 lockdown. In multivariate (adjusted) logistic regression analysis (table 4), being a wage earner and employed HH head significantly remained positively (AOR: 8.3, 95% CI: 1.9-36.2) and negatively (AOR: 0.07, CI: 0.02-0.2) associated with HHFI, respectively after adjusting for food aid, age, sex, place of residence, size of HH members, and HH head ownership of house.
Food insecurity status and its association with social inequalities during partial lifting of COVID-19 lockdown in Nansana municipality, Wakiso district Uganda
During partial lifting of COVID-19 lockdown, bivariate logistic analysis showed that slum HHs (COR: 12.3, 95%CI:1.6-93.8), female headed HHs (COR:13.5, 95% CI:1.8-102.9), daily wage earners (COR: 12.3, 95% CI:1.6-93.9) and tenant HH heads (COR: 4.6, 95%CI:1.3-16.5) were positively associated with food insecurity. In multivariate (adjusted) logistic regression analysis (table5), slum HHs (AOR: 11.8, 95% CI: 1.5-91.3), female headed HHs (AOR: 11.9, 95%CI: 1.5-92.7), HH head being a wage earner (AOR: 10.7, 95% CI: 1.4-82.9) and tenant (AOR: 4.0, 95% CI: 1.1-14.7) significantly remained positively associated with HHFI after adjusting for other covariates. On the other hand, non-slum HHs (AOR: 0.2, CI: 0.07-0.9), male headed HHs (AOR: 0.1, CI: 0.01-0.7), employed HH heads (AOR: 0.2, 95% CI: 0.06-0.7), having own house (AOR: 0.2, 95% CI: 0.7-0.9), significantly remained negatively associated with HHFI after adjusting for other covariates. Table 5 shows the bivariate and multivariate logistic regression analysis of factors associated with food insecurity during partial lifting of COVID-19 lockdown in Nansana municipality, Wakiso district, Uganda.