Multiple-Indicator, Multiple Cause Modelling to Examine the Relationship between Foods Consumed and Non-Communicable Diseases

Laina Mbongo, University of Namibia
Lillian Pazvakawambwa, University of Namibia
Lawrence Kazembe, University of Namibia

Non-Communicable diseases are commonly associated with the dietary patterns of an individual. quantifying the disease’s burden over a household’s or individual’s health has been a topic of great interest to researchers as well as policymakers. Principal Component Analysis was used as a data reduction method to derive dietary patterns. Furthermore, this chapter applied a Multiple-Indicator, Multiple-Cause (MIMIC) model in which NCD’s is dealt with as an unobserved construct or latent variable to be determined by its causes and indicators and to be estimated in a system of structural equations. SEM was used to assess the association between the prevalence of NCD’s and food types consumed. Fruits, foods such as condiments/tea/coffee and potatoes, yams, cassava, or any foods made from roots and tubers accounted for majority of the variation. The SEM showed that food types such as local grains, meat and food made from oil or were found to be significant at 5% level.