Background: Prices of foods can influence purchase and, therefore, overall quality of diet. However, a limited number of studies have analyzed the cost of diets according to the overall quality of diets taking into account sociodemographic characteristics.

Objective: Our aim was to estimate cost variations according to diet quality and to identify sociodemographic characteristics associated with such cost differences in adults' diets in Belgium.

Design: This cross-sectional study used nationally representative data from Belgium.

Participants/settings: Participants were adults (aged 18 to 64 years; n = 1,158) included in the 2014-2015 Belgian National Food Consumption Survey.

Main outcome measures: Dietary assessment was based on two 24-hour dietary recalls and a food frequency questionnaire. The Mediterranean Diet Score and the Healthy Diet Indicator were used to assess diet quality. Daily diet cost was estimated after linking the consumed foods with the 2014 GfK ConsumerScan Panel food price data.

Statistical analyses performed: Associations were estimated using linear regressions.

Results: The mean daily diet cost was US$6.51 (standard error of mean [SEM] US$0.08; €5.79 [€0.07]). Adjusted for covariates and energy intake, mean (SEM) daily diet cost was significantly higher in the highest tercile (T3) of both diet quality scores than in the T1 (Mediterranean Diet Score: T1 = US$6.29 [US$0.10]; €5.60 [€0.09] vs T3 = US$6.78 [US$0.11]; €6.03 [€0.10]; Healthy Diet Indicator: T1 = US$6.09 [US$0.10]; €5.42 [€0.09] vs T3 = US$7.13 [US$0.11]; €6.34 [€0.10]). Both diet quality and cost were higher in 35- to 64-year-old respondents (vs 18- to 34-year-olds), workers (vs students), and those with higher education levels (vs the lowest). The association between quality and cost of diets was weaker in men and among individuals with higher education levels.

Conclusions: In Belgium, a high-quality diet was more expensive than a low-quality diet. These findings can be used to inform public health policies.