Abstract
To examine minimum dietary diversity (MDD) trends and determinants among children aged 6-23 months. Secondary analysis of the Indonesia Demographic and Health Surveys (IDHS) between 2007 and 2017. The primary outcome was MDD, the consumption of at least five out of eight food groups (MDD-8). We included a total of 5015 (IDHS 2007), 5050 (IDHS 2007) and 4925 (IDHS 2017) children aged 6 to 23 months to estimate trends of MDD-8 and to identify factors associated with MDD-8. We used multiple logistic regression analysis adjusted for the complex sampling design to investigate the association between the study factors and MDD-8. Indonesia. A total of 14 990 children aged 6-23 months. Over the 10 years, the percentage of children who consumed a diversified diet was 53·1 % in 2007, 51·7 % in 2012 and 53·7 % in 2017. Multivariate analyses showed that older age children, higher maternal education, maternal weekly access to media, paternal non-agricultural occupation, history of at least four antenatal care visits and wealthier households were associated with the increased odds of MDD-8. Children living in rural areas, Sulawesi and Eastern Indonesia, were less likely to eat a diversified diet. The proportion of children meeting MDD-8 has stagnated in the last decade. Child, parental, health care, household and community factors are associated with MDD-8. Therefore, nutrition education programmes and behaviour change communication activities should target mothers and families from socio-economically and geographically disadvantaged populations.
Highlights
The percentage of minimum dietary diversity (MDD)-8 declined steadily among [6,7,8,9,10,11] months old children, children whose mothers working in agriculture, wealthier households and urban residents
Compared to MDD-8, the proportion of children meeting MDD-7 was higher across the three surveys but showed a similar pattern across the study variables
Interaction analyses on the factors associated with MDD-8 We found significant interactions between the mother’s education and household wealth (P = 0·006) and between the mother’s access to media and household wealth (P < 0·001) during our investigation on the model assessing factors related to MDD-8
Summary
Data sources and study population We used data from the IDHS collected in 2007, 2012 and 2017(13,14,20). The census blocks or sample clusters were selected from each subdistrict using systematic random sampling in the second stage. The survey team identified twenty-five households in each cluster by simple random sampling and interviewed all ever-married women in the selected households(13,14,20). The current indicator was an updated version of MDD-7 in 2008(10) that defined the consumption of at least four of seven food groups without adding breastmilk. For this study analysis of MDD-8, we constructed the breastmilk item based on the separated question of current breast-feeding status. Since the 2007 IDHS collected information on flesh foods and eggs into one single question, we could not assess flesh foods and eggs trend separately in this survey. We found a similar trend for each ‘flesh foods’ and
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
