The Impact Of Food Insecurity On Quality Of Life In Windhoek Informal Settlements: A Structural Equation Modelling

ABSTRACT

Namibia is experiencing an increase in its population living in urban areas due to rapid urbanisation. However, this growth has not been the same in all parts of the country. Currently the biggest growth rate is experienced in Windhoek, which is home to about 16% of the total population. Due to rapid population growth, Windhoek has been facing a number of challenges such as high unemployment rate, illegal land occupation, and development of informal settlements among others. As a result, over-population has led to various social problems in informal settlements such as poor health condition, lack of access to clean water, unsustainable livelihood and food insecurity among others. Given the number of challenges faced by residents of informal settlements, there is a need to assess the impact of food insecurity on quality of life, since going without food does have an impact on people’s well-being.

The objective of the study is to model the relationship between food insecurity and quality of life in Windhoek informal settlements using Structural Equation Modelling (SEM) approach. Secondary data from a quantitative cross-sectional study were used. The study measured food insecurity using Household Food Insecurity Scale (HFIAS) indicators. World Health Organization Quality-of-Life Scale (WHOQOL-BREF) and European Quality of life (EQ-5D) instruments were used to measure quality of life. Multivariate statistical analysis using Exploratory Factor Analysis (EFA), Principal Component Analysis (PCA), and Hierarchal

ii

Clustering methods were applied to reduce dimensionality in the data, and to link

food insecurity to quality of life.

EFA results showed that from the set of ten (10) questions asked on food insecurity,

only one component was adequate in explaining food insecurity. Similarly, results

revealed that only four components (social, environmental, physical and

psychological health) accounted for much variation in WHOQOL-BREF from the

set of 21 questions asked. The findings further indicated that four constructs

(pain/discomfort, anxiety/depression, self-care and usual activities) were sufficient

in explaining quality of life based on EQ-5D instrument. All measurement scales

for the constructs met the minimum condition of the Cronbach’s alpha coefficient,

all with α ≥ 0.70. The study further assessed the level of agreement between data

driven approach and existing method of food insecurity using Kappa test of

symmetric. The results showed that there was a substantial agreement between

HFIAS and the data driven, with a Kappa ( ) value of 0.68 (p < 0.001).

Overall, latent models identified with EFA were validated using CFA before fitting

SEM. Measurement models were evaluated based on goodness of fit, level of

significance and coefficient signs, and non-significant models were modified by

removing unstable parameters. The SEM was accepted based on the criteria used

for CFA. The results showed that the model had a significant (1741) 4527.90 2   , p

< 0.001 and SRMR = 0.07 and a RMSEA = 0.07 (90% CI=0.071; 0.077), which

indicated that the model fits the data well. More importantly, the results supported

that food insecurity was a significant predictor of quality of life outcomes (physical

health, environmental health, social health, pain/discomfort, anxiety/depression,

iii

self-care and usual activities) in Windhoek informal settlements. The study further revealed that lack of access to food affect social well-being, environmental well-being and physical well-being in a negative way. Similarly, the study found that pain/discomfort, anxiety/depression, self-care, and usual activities were positively associated with food insecurity.