# Daily meal frequency and its associated factors among children aged 6–23 months in Ethiopia: a Bayesian hierarchical Poisson model

## TL;DR

This study in Ethiopia finds that daily meal frequency among young children is low and varies by region, age, and maternal education, highlighting the need for targeted nutrition interventions.

## Contribution

The study introduces a Bayesian hierarchical Poisson model to analyze daily meal frequency and its determinants in young children at the national level in Ethiopia.

## Key findings

- Children of mothers with secondary/higher education had a 17% higher daily meal frequency compared to those with no education.
- Daily meal frequency increased significantly with age, with children 18–23 months having 90% more meals than those 6–8 months.
- Regional disparities were observed, with children in Afar and SNNPR having 27–33% lower meal frequency than those in Tigray.

## Abstract

Inadequate feeding frequency during the early childhood period is responsible for more than two-thirds of global child deaths. Evidence on the rate of daily meal frequency among infants and young children at the national level is crucial for developing targeted interventions to improve feeding practices. Hence, this study aimed to identify factors associated with the rate of daily meal frequency (DMF) among children aged 6–23 months in Ethiopia.

We retrieved secondary data from the Kids record (KR) of the Ethiopian Mini Demographic and Health Survey (MDHS) dataset. A total of 1,264 children aged 6–23 months were included in the study. A Bayesian hierarchical Poisson model was employed. Model convergence was checked via Rhat, effective sample size, density plots, terrace plots, and autocorrelation plots, and all the results were confirmed. We used the widely applicable information criterion (WAIC) and leave-one-out cross-validation (LOO) for model comparison. The model parameters were estimated via special Markov chain Monte Carlo (MCMC) simulation techniques called Hamiltonian Monte Carlo (HMC) and its extension, the no-U-turn sampler (NUTS). An adjusted incidence rate ratio (AIRR) with a 95% credible interval (CrI) in the multivariable model was used to select variables that had a significant association with the rate of daily meal frequency. The data were analyzed via R software version 4.3.1.

The mean and standard deviation of the DMF were 3.36 and 1.60, respectively. The rate of DMF was 1.17 times greater (AIRR = 1.17, 95% CrI: 0.997, 1.381) in children whose mothers had a secondary/higher educational level than in those whose mothers had no education. Kids currently being breastfed have a lower rate of DMF (AIRR = 0.88, 95% CI: 0.798, 0.979) by 10% than those who are not currently breastfeeding. Compared with children between the ages of 6–8 months, those between 9 and 11 months (AIRR = 1.55 95% CrI: 1.374, 1.754), 12–17 months (AIRR = 1.72, 95% CrI: 1.543, 1.911), and 18–23 months (AIRR = 95% CrI: 1.90, 1.692, 2.125) had 55, 72 and 90% higher rates of DMF, respectively. In the Afar region (IRR = 0.77, 95% CI: 0.615, 0.982), Somalia (AIRR = 0.83, 95% CrI: 0.682, 1.01), Benishangul (AIRR = 0.8, 95% CrI: 0.639, 0.994), Southern nation nationality and people’s region (SNNPR) (AIRR = 0.73, 95% CrI: 0.596, 0.894), and (AIRR = 0.73, 95% CrI: 0.572, 0.925) decrease the daily meal frequency by 33, 17, 20, 27 and 27%, respectively, compared with that of children from Tigray.

The rate of DMF was low in Ethiopia and exhibited a significant clustering pattern across the country. These findings stress the need for tailored interventions addressing regional inequities, promoting age-specific nutrition, supporting maternal education, and empowering working women to improve children’s nutritional intake and ensure more equitable access to meals across Ethiopia.