A Novel Food-Based Diet Quality Score Is Associated with Nutrient Adequacy and Reduced Anemia Among Rural Adults in Ten African Countries

Abstract

Abstract Objectives We aimed to develop and evaluate an easily-tabulated metric that is sensitive to diet quality in diverse settings. In this analysis, we examined associations between a novel food-based metric – the Global Diet Quality Score (GDQS) - and diet quality using data from rural Africa. Methods The GDQS gives points for higher intake of 16 healthy food groups and lower intake of 9 unhealthy groups, based on 3 intake ranges for each group. We scored the GDQS using food frequency questionnaire (FFQ) data from 1613 men and 1710 nonpregnant nonlactating women ages 15–49 (median: 30) in 12 rural African villages participating in the Millennium Villages Project in 2006 to 2009. We evaluated associations between the GDQS and nutrient intakes calculated from the same FFQ, body mass index (BMI), mid-upper arm circumference (MUAC), hemoglobin, and an overall nutrient adequacy score ranging from 0 to 7 (1 point given for each of 7 nutrients meeting average requirements). Associations were also derived for GDQS submetrics (GDQS+ and GDQS−) computed using only healthy or unhealthy food groups, respectively, and simplified versions of the GDQS and submetrics (scored using 2 intake ranges for each food group). Results Moderate rank correlations were observed between the GDQS+ and energy-adjusted intakes of fiber (women: 0.43, men: 0.33), folate (0.40, 0.30), vitamin A (0.40, 0.34), and zinc (0.38, 0.30), exceeding correlations with the Minimum Dietary Diversity Score for Women (p for difference between metrics <0.05 for fiber and folate in both sexes). The simplified GDQS- correlated moderately with energy-adjusted saturated fat intake (women: 0.35, men: 0.29). Rank correlations between the GDQS+ and overall nutrient adequacy score differed notably by country (range: 0.34–0.75), but not by age or season. Adjusting for age, interview month, and household size, the GDQS and GDQS+ were associated with hemoglobin and anemia (p for difference between metrics >0.05): women and men in the highest GDQS+ quintile had an OR of anemia of 0.25 (95% CI 0.09, 0.68) and 0.16 (95% CI 0.04, 0.46), respectively, relative to those in the lowest. Metrics were not associated with BMI or MUAC in adjusted analyses. Conclusions The GDQS and submetrics were associated with nutrient intakes, hemoglobin, and reduced anemia in rural African adults. Funding Sources Intake - Center for Dietary Assessment at FHI Solutions.