Global Health
The Association between Minimum Dietary Diversity and Unhealthy Food Consumption among Women in Ghana David Owiredu* David Owiredu Owiredu Owiredu Owiredu Owiredu Department of Epidemiology and Biostatistics, Arnold School of Publich Health, University of South Carolina.
Background: The Minimum Dietary Diversity for Women (MDD-W) is a proxy for micronutrient adequacy, but its association with diet quality in settings undergoing a nutrition transition is underexplored. We examined the association between MDD-W and consumption of unhealthy foods among females aged 15-49 years in Ghana.
Methods: We analyzed cross-sectional data from the 2022 Ghana Demographic and Health Survey. Using the Food and Agriculture Organization’s definition, meeting MDD-W threshold was operationalized as the consumption of ≥5 of 10 food groups in the prior 24 hours. Unhealthy food consumption was defined as the intake of sweetened beverages, sugary snacks, fried foods, or processed packaged foods. We used 1:1 nearest-neighbor propensity score matching with a caliper of 0.1 on age, education (none, primary, secondary, higher), household wealth (asset-based quintiles), marital status, region, rural/urban residence, employment, and media exposure, followed by logistic regression in the matched sample. E-values were calculated to assess sensitivity to unmeasured confounding.
Results: Among 15,008 women, 47.9% met the MDD-W threshold, and 7.5% reported consuming at least one unhealthy food. Consumption of starchy staples (98.9%) and green vegetables (71.8%) was high, but intake of dairy products (15.4%) was low. Matching produced 5,846 balanced pairs, and achieving MDD-W was associated with higher odds of consuming unhealthy foods (OR: 1.33, 95% CI: 1.15-1.54). Based on the E-value, an unmeasured confounder would need to be associated with both the exposure and outcome by an odds ratio of at least 1.99 to fully account for the observed association.
Conclusion: Women who achieved minimum dietary diversity were more likely to consume unhealthy foods. Our findings suggest that dietary diversity alone may not adequately capture overall diet quality in this setting. However, given the moderate E‑value, residual confounding cannot be ruled out.