Poor diet is a leading cause of preventable mortality globally, accounting for an estimated 26%. While individual nutrients are important, overall dietary patterns are more strongly linked to health outcomes. Existing research on global dietary patterns has significant limitations: studies often cover small numbers of countries, rely on inaccurate national food availability data instead of individual-level intake data, exclude children and adolescents, and lack comprehensive assessments across multiple validated dietary quality metrics. This study aimed to address these gaps by characterizing global, regional, and national dietary patterns and trends from 1990 to 2018, using individual-level data from the Global Dietary Database (GDD) for 185 countries. The analysis considered variations by age, sex, education, and urbanicity to provide a more comprehensive understanding of global dietary habits and disparities.

Previous studies on dietary patterns have been limited in scope and rigor, particularly concerning children and adolescents. Many studies used national per capita food availability or sales data, which are known to misestimate actual individual intake. Furthermore, there's a lack of global data on disparities in dietary patterns based on age, sex, education, and urbanicity. Existing studies seldom jointly assessed several validated diet quality metrics, such as the AHEI, DASH, and MED scores. This paper addresses the need for more comprehensive and rigorous data on global dietary patterns.

This study utilized data from the Global Dietary Database (GDD) 2018, a collaborative project compiling and standardizing individual-level dietary data from around the world. The database includes data on 53 foods, beverages, and nutrients. Bayesian modeling methods were employed to estimate dietary intakes, stratified by age, sex, education, and urbanicity for 185 countries between 1990 and 2018. Three validated dietary quality metrics were assessed: the Alternative Healthy Eating Index (AHEI), Dietary Approaches to Stop Hypertension (DASH) score, and Mediterranean Diet Score (MED). The AHEI, the primary metric, incorporates nine components (fruit, non-starchy vegetables, whole grains, sugar-sweetened beverages, legumes/nuts, red/processed meats, seafood omega-3 fat, polyunsaturated fatty acids, and sodium). DASH and MED scores were also calculated using established methods. Population-weighted average dietary pattern scores were computed for each subgroup, considering uncertainties and demographic changes over time. Data were standardized to 2,000 kcal per day to allow for comparability across subgroups. The study employed several validity checks, including five-fold cross-validation and comparisons of predicted versus observed intakes. A varying slopes model, incorporating country-level intercepts and slopes, was used to strengthen time-trend estimates for dietary factors with corresponding food or nutrient availability data.

The global mean AHEI score in 2018 was 40.3 (95% UI 39.4, 41.3), indicating modest overall dietary quality. Regional means ranged from 30.3 in Latin America and the Caribbean to 45.7 in South Asia. Only ten countries had AHEI scores ≥50. Among the 25 most populous countries, scores varied substantially, with Vietnam, Iran, Indonesia, and India having the highest scores and Brazil, Mexico, the United States, and Egypt having the lowest. Globally, AHEI scores were similar for children and adults, except in Central/Eastern Europe and Central Asia, high-income countries, and the Middle East and Northern Africa, where children had lower scores. Women generally had higher AHEI scores than men. Individuals with higher education levels had better diet quality in most regions, with the largest differences seen in Central/Eastern Europe and Central Asia, Latin America and the Caribbean, and South Asia. Urbanicity showed mixed effects on dietary quality, varying by region. Globally, the mean AHEI score increased by 1.5 points between 1990 and 2018, but this trend wasn't consistent across all regions (no significant change in South Asia, and a decreasing trend in Sub-Saharan Africa). Increases in scores for non-starchy vegetables and legumes/nuts were offset by decreases in scores for red/processed meat, sugar-sweetened beverages, and sodium. DASH and MED scores showed similar regional variations and trends, although scores were generally lower than AHEI scores. High correlations existed between the three dietary pattern scores, ranging from 0.5 to 0.8.

This study provides the most comprehensive assessment to date of global dietary quality, including data on children and adolescents across multiple dietary quality metrics. The findings highlight the considerable regional and demographic variations in dietary quality. The modest overall improvement in dietary quality globally since 1990 is not uniform, with some regions showing no improvement or even decline. This underscores the need for region-specific interventions tailored to address unique dietary challenges and promote healthier eating habits. Regional differences in the balance between insufficient intake of healthy foods versus excess intake of unhealthy foods necessitate targeted policies. For instance, low-income countries in South Asia and Sub-Saharan Africa need to focus on increasing the consumption of produce, seafood, and plant oils. Conversely, high-income countries need to focus on both increasing healthy foods and decreasing harmful ones. The identified disparities in dietary quality based on education and urbanicity highlight the need for policies that address health inequities. The association between dietary quality and chronic diseases underscores the significance of these findings for public health.

This study provides a comprehensive assessment of global dietary quality across 185 countries from 1990 to 2018. It reveals modest overall dietary quality with substantial regional and demographic variations. While some improvement is seen globally, significant disparities remain, indicating the need for region-specific and targeted interventions to enhance nutritional health equity. Future research should focus on evaluating the effectiveness of such interventions and exploring the impact of broader social determinants of health on dietary patterns.

While this study used a large dataset and advanced statistical methods, some limitations exist. Individual-level dietary data are subject to measurement errors. Survey data availability was incomplete for some countries, dietary factors, and demographic groups, particularly for children aged 3–9 and adults ≥85. The study standardized dietary intakes to 2,000 kcal per day, which may not accurately reflect the dietary needs of all population groups. The study did not include data on trans fat or alcohol consumption, potentially affecting the comprehensive assessment of dietary quality. Finally, the dietary metrics used, while validated, were initially developed for adult populations in high-income countries, which needs to be considered for interpreting the findings in diverse global populations.