Using both principal component analysis and reduced rank regression to study dietary patterns and diabetes in Chinese adults

Abstract

Objective We examined the association between dietary patterns and diabetes using the strengths of two methods: principal component analysis (PCA) to identify the eating patterns of the population and reduced rank regression (RRR) to derive a pattern that explains the variation in hemoglobin A1c (HbA1c), homeostasis model of insulin resistance (HOMA-IR), and fasting glucose. Design We measured diet over a 3-day period with 24-hour recalls and a household food inventory in 2006 and used it to derive PCA and RRR dietary patterns. The outcomes were measured in 2009. Setting Adults (n = 4,316) from the China Health and Nutrition Survey. Results The adjusted odds ratio for diabetes prevalence (HbA1c ≥ 6.5%), comparing the highest dietary pattern score quartile to the lowest, was 1.26 (0.76, 2.08) for a modern high-wheat pattern (PCA; wheat products, fruits, eggs, milk, instant noodles and frozen dumplings), 0.76 (0.49, 1.17) for a traditional southern pattern (PCA; rice, meat, poultry, and fish), and 2.37 (1.56, 3.60) for the pattern derived with RRR. By comparing the dietary pattern structures of RRR and PCA, we found that the RRR pattern was also behaviorally meaningful. It combined the deleterious effects of the modern high-wheat (high intake of wheat buns and breads, deep-fried wheat, and soy milk) with the deleterious effects of consuming the opposite of the traditional southern (low intake of rice, poultry and game, fish and seafood). Conclusions Our findings suggest that using both PCA and RRR provided useful insights when studying the association of dietary patterns with diabetes.

The Diabetes Epidemic in China: An Integrated Review of National Surveys

“…In addition, glycemic response to glucose and rice in people of Chinese ethnicity was 60% greater than in Europeans (39). Very recently, results from the China Health and Nutrition Survey reconfirmed the association between dietary patterns and diabetes or insulin resistance among Chinese adults, using both principal component analysis and reduced rank regression (40). …”

Data mining: Potential applications in research on nutrition and health

“…On the other hand, PCA categorises similarities in variables rather than cases and can be used to reduce the number of variables or determine which variables explain the dissimilarity in the data. This type of analysis is also widely used in establishing dietary patterns . Often, PCA is used for variable reduction prior to considering predictive models.…”

The Difference of Being Overweight or Obese between the North and South of China

“…Dietary patterns were extracted using principal component factor analysis based on 12 main categories of food [20]. Two dietary patterns were obtained from participants' food consumption records with eigenvalues (> 1), and the variance explained by each factor.…”