Defining Diet Quality: A Review of Diet Metrics as Applied to the Double Burden of Malnutrition (OR17-01-19)

Meeting most of the UN Sustainable Development Goals (SGDs) will require a strong focus on tackling all forms of malnutrition─ addressing maternal and child health (MCH) as well as diet-related non-communicable diseases (NCDs). Yet, the optimal metrics to define a healthy diet remain unclear. Our aim was to comprehensively review diet metrics and assess the evidence on each metric's association with MCH and NCDs. Using comprehensive searches and expert discussions, we identified metrics that i) are used in ≥3 countries to link diet to health, ii) quantify the number of foods/food groups consumed and/or iii) quantify recommended nutrient intakes. We reviewed and summarized each metric's development, components and scoring. For each identified metric, we systematically searched PubMed to identify meta-analyses or narrative reviews evaluating these metrics with nutrient adequacy and health outcomes. We assessed validity by grading the number of studies included and the consistency of the diet metric-disease relationship. We identified 6 MCH, 13 NCD and 0 MCH/NCD metrics. Most were developed for describing adherence to dietary guidelines or patterns, and others were developed for predicting micronutrient adequacy. On average, the metrics included 14 food groups/nutrients (range 4–45), with 10 food-group only metrics and 0 nutrient-only metrics. The most frequent metric components were grains/roots/tubers, fruits and vegetables. We identified 16 meta-analyses and 14 narrative reviews representing 102 metric-disease relationships (98 metric-NCD and 4 metric-MCH relationships, respectively). We found 5 metrics that have been consistently validated in meta-analyses and narrative reviews for NCDs, 1 metric with limited evidence for MCH, but 0 metrics for both. Of the metrics, the Alternative Healthy Eating Index (aHEI), Dietary Approaches to Stop Hypertension (DASH), Healthy Eating Index (HEI), and Mediterranean Diet Score (MED) were most commonly validated, especially for all-cause mortality and cardiovascular disease (Figure 1). Few diet metrics have been used in multiple countries to define a healthy diet. This suggests a serious gap in global analyses of diet quality relating to malnutrition in all its forms, which hinders effective policy action. Gates Foundation.