Abstract:
Background: Food insecurity is a pressing social and public health issue that varies in degree
and impact on individuals and social groups, requiring immediate attention from policymakers
and decision-makers. Therefore, this study aimed to identify the spatial distribution of household
food insecurity and associated factors across different zones in Ethiopia.
Methods: This study utilized data from the LSMS 2018/2019. A community-based cross-sectional
study design was employed. A stratified two-stage sampling technique was used to draw a
sample of 6680 households. Spatial and Ordinal logistic regression models were fitted to
determine factors associated with household food insecurity. STATA 14, SAS version 9.4, and
ARC GIS version 10.8 statistical software were used for data management, logistic models, and
spatial respectively.
Result: Of the total households included in this study, (71.8%) of severely food insecure
households would be found in this study. The Amhara region is 38.8%, the Tigray region 32.9%,
and the Somali region 30.5% more likely food insecure as compared to Addis Ababa
respectively. Sex of the household head for females was 49.6% more likely food insecure
compared with males. Households who save in financial institutions were 29.9% less likely food
insecure than households those not saving. Household heads who reside in urban were 57% less
likely severe food insecure compared with rural. From ESS 2018_2019 data North Gondar,
northern and southern zones of Wollo, border of Amhara and Afar zones, unknown zones of
Somali, and southern and eastern Tigray zones were high-risk areas.
Conclusion: from the model mentioned in the method, the partial proportional odds model was
preferable due to the low value of AIC. Residence, savings, religion, region, number of
household members, marital status, sex, educational level, and average rainfall were significant
factors in food insecurity among households. Western Tigray, Waghimra, south Wollo, west
Gojjam, unknown Somali zone, zone4 of Amhara, the northern and southern part of Gondar, and
Oromia zone zone5 were a high proportion of food insecurity. Therefore, the government gives
spatial attention to the above regions.
Keywords: household level, ordinal logistic regression, partial proportional odds model,
Ethiopia, spatial, zones