Increase Font Size Decrease Font Size View as PDF Print

Plain Language Summary

Combinations of food intake (assessed using reduced rank regression) that explain the most variation in risk of cardiovascular disease

Researchers have previously looked at the relationship between individual foods and nutrients and health. Today, there is interest in looking at how combinations of foods and beverages intake, or dietary patterns, influence health by applying different scientific methods. A statistical method called reduced rank regression analysis can be used to describe the patterns of foods and beverages people eat based on a set of “response variables” that are known to be related to the health outcome of interest. This summary of a NEL review presents what research evidence currently exist when reduced rank regression analysis is the method used to study the dietary patterns of groups of people and their likelihood of developing the risk for cardiovascular diseases such as high blood lipids, high blood pressure, and heart disease. 
 
Conclusion
Insufficient evidence, due to a small number of studies, was available to examine the relationship between dietary patterns derived using reduced rank regression and risk of cardiovascular disease. The disparate nature of the methods used made it difficult to compare results, and therefore, no conclusions were drawn. 
 
What the Research Says
  • Four studies looked at dietary patterns found using reduced rank regression analysis and the risk of developing cardiovascular disease. However, these studies had some key issues that make it hard to make any recommendations:
  • There were few studies available.
  • There were many differences in how the studies were done.
The populations studied were different between studies.

Technical Abstract

Background
The goal of this systematic review project is to identify patterns of food and beverage intake that promote health and prevent disease. Historically, most dietary guidance has been based on research conducted on individual food components or nutrients. Dietary patterns are defined as the quantities, proportions, variety, or combination of different foods, drinks, and nutrients (when available) in diets, and the frequency with which they are habitually consumed. Reduced rank regression (RRR) is a statistical method that determines dietary patterns (combinations of food intake) that explain as much variation as possible among a set of response variables related to a health outcome of interest. It is an a posteriori method since it uses both existing evidence and exploratory statistics. The objective of this systematic review was to assess the relationship between patterns of food and beverage intake identified using reduced rank regression, and risk for cardiovascular disease.
 
Conclusion Statement
Insufficient evidence, due to a small number of studies, was available to examine the relationship between dietary patterns derived using reduced rank regression and risk of cardiovascular disease. The disparate nature of the methods used made it difficult to compare results, and therefore, no conclusions were drawn. (Grade: Not Assignable)
 
Methods
Literature searches were conducted using PubMed, Embase, Navigator (BIOSIS, CAB Abstracts, and Food Science and Technology Abstracts), and Cochrane databases to identify studies that evaluated the association between dietary patterns derived using reduced rank regression analysis and risk of cardiovascular disease (CVD). Studies that met the following criteria were included in the review: Human subjects; Ages: 2 years and older; Populations: Healthy and those with elevated chronic disease risk; individuals with chronic disease; published in English in a peer-reviewed journal; Sample size: Minimum of 30 subjects per study arm; Dropout rate less than 20 percent; Study assesses dietary intake using reduced rank regression analysis; study considered cardiovascular disease and risks of  cardiovascular disease; subjects from countries with high or very high human development (based on the 2011 Human Development Index). The date range for the conduct of the studies was unlimited.
 
The results of each included study were summarized in evidence worksheets (including a study quality rating) and evidence table. A group of subject matter experts were involved in a qualitative synthesis of the body of evidence, development of a conclusion statement, and assessment of the strength of the evidence (grade) using pre-established criteria including evaluation of the quality, quantity and consistency, magnitude of effect, and generalizability of available evidence.
 
Findings
  • Four prospective cohort studies examined dietary patterns derived using reduced rank regression analysis and their association with CVD risks and incidence. The studies ranged in size from 981 to 26,238 subjects, and one study each was conducted in the United States, United Kingdom, and Germany, and one included subjects from Europe.  The follow-up for these studies ranged in duration from 6 to 25 years.
  • Comparison across studies was limited by the small number of studies, differences in methodologies used, and in the populations studied. Therefore no conclusions were drawn.
  • More U.S. population-based research is needed to examine dietary patterns and risk of cardiovascular disease using reduced rank regression, preferably with more consistent methods and response variables.
Discussion
The ability to draw a gradable conclusion was limited due to the following issues:
  • Three out of the four studies used biomarkers and the fourth study used nutrients as response variables in the reduced rank regression analyses. Among the three studies that used biomarkers as response variables, there were differences in the type of biomarkers chosen, leading to the identification of dietary patterns that differed from study to study. One study used change in BMI, mean arterial pressure, total cholesterol, HDL-cholesterol, triglycerides (mg/dl), fasting glucose, and uric acid. The second study used C-reactive protein, Interleukin (IL)-6, and Interleukin (IL)-18, while the third study used total cholesterol, HDL cholesterol, and triglycerides. The fourth study used nutrients, including total fat, total carbohydrate, and fiber, as response variables. Because the dietary patterns described in each study are directly linked to response variables chosen, the variation in the response variables used means that the resulting dietary patterns may not be comparable.
  • Dietary assessment methods were different across the studies. One study used 3-day diet records; another used a self-administered FFQ; a third used a 127-item validated FFQ, and the fourth study used a 7-day dietary record. It is unclear what impacts different dietary assessment methods have on the derivation of dietary patterns using reduced rank regression.
  • The studies were not consistent in their use of confounders in analyses. In particular, physical activity was not included as a confounder in the analyses by one study, and another did not include smoking as a confounder.
  • The studies were conducted in different countries, representing populations in different regions of the world, which limited the ability to compare and interpret the results due to potential differences in dietary patterns between these regions. From that perspective, the results may not be generalizable to some U.S. populations.