Populationg-based Validation Of The Nutritional Score For Pre-packed Foods

BackgroundThe nutritional score for pre-packed foods is an index which assesses nutrient quality of food items by ranking them according to their nutrient composition, which is one of the latest nutrient profiling model in China. Validating a nutrient profiling model was one of the most important steps in establishing a model. Population-based validation was widely used, which can realize the ability of the model on validating diet quality and predicting health outcomes. As a nutrient profiling model, the nutritional score for pre-packed foods should be validated by different types of validation methods.ObjectivesOur study aimed to investigate the discriminant validity of the nutritional score for pre-packed foods by using data from the China Health and Nutrition Survey(CHNS) and then to evaluate predictive validity of the model by examining the association between the nutritional score for pre-packed foods and incidence of central obesity among adults aged 18 to 60 in China.MethodsDietary intake data was assessed using consecutive 3-day 24-hour recall and household food weighing from the CHNS in 2011 and the individual diet score was calculated with the nutritional score for pre-packed foods among adults aged 18-60. The associations between the individual diet score (in quintiles) and intakes of various food groups, nutrients and the proportion of participants having nutrient intake less than the estimated average requirement (EAR) were investigated using linear regression model and Chi-square test. Adults without central obesity and pre-central obesity at baseline(1993) were included in our analyses respectively. Central obesity was defined as men with waist circumference less then 90cm and women with waist circumference less than 85cm and pre-central obesity was defined as men with waist circumference less then 85cm and women with waist circumference less than 80cm. Participants becoming central obesity and pre-central obesity in the following surveys were regarded as endpoint incident. The individual diet score and food group scores (in quintiles) were calculated with the nutritional score for pre-packed foods using 1993 dietary intake data. The hazard ratios (HRs) of central obesity were calculated with Cox proportional hazards regression.Results1.The adults in 2011 with higher quartile of the dietary scores were more likely to be female, non-smokers and non-drinkers, have higher education and income, and live in the south part of China.2.The adults in 2011 with higher quartile of the diet scores had higher intakes of cereals and potatoes, and lower intakes of vegetables, fruits, livestock and poultry meat, fish and shellfish, eggs, milk and nuts.3.The adults in 2011 with higher quartile of the diet scores had lower intakes of energy, cereal protein and fat, saturated fat and sodium(P<0.0001) and higher intakes of protein from other sources, other minerals, vitamins(P<0.0001). Among participants with higher quartile of the diet scores, there was lower proportion of participants who had nutrient intakes less than EAR(P<0.01).4.The mean diet score showed increased trend from 1993 to 2011. Participants were female, non-smokers and non-drinkers, had higher education and income, and lived in the south part of China that had higher diet score.5.Higher diet score was associated with lower risk of pre-central obesity and central obesity among the adults. Comparing the extreme quintile, the adjusted HR was 0.66(95%CI:0.50-0.88) and 0.78(95%CI:0.57-1.07) in urban adults,0.74(95%CI; 0.62-0.88) and 0.71(95%CI:0.59-0.87) in rural adults. Higher vegetable and fruit score was associated with lower risk of central obesity with an HR of 0.82(95%CI: 0.59-1.13) in urban adults and 0.72(95%CI:0.59~0.87) in rural adults. Higher cereal and potato score was associated with higher risk of central obesity with an HR of 1.62(95%CI:1.15~2.27) in urban adults and1.42(95%CI:1.15~1.75) in rural adults.ConclusionOur results showed adequately discriminant validity of the nutritional score for pre-packed foods. To some degree, the model can characterize individual diet quality and nutritional status. Findings from our study suggest that the model has the potential to make healthy food choices.