QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Barr, S. I. Articles by Johnson, R. K. Search for Related Content PubMed PubMed Citation Articles by Barr, S. I. Articles by Johnson, R. K.

---

Letters to the Editor

Effect of Added Sugars on Dietary Quality

Human Nutrition Faculty of Agricultural Sciences University of British Columbia

Rachel K. Johnson

Human Nutrition Faculty of Agricultural Sciences University of British Columbia Nutrition and Food Sciences College of Agriculture and Life Sciences University of Vermont

Dear Editor:

We were intrigued by Forshee and Storey’s (1) critique of the approach used by the Institute of Medicine (IOM) (2) to analyze the effect of added sugars on diet quality. They suggested that the statistical model used was flawed because it did not control for total energy in the diet, and they also criticized the use of the ratio variable "percent energy from added sugars" (%E_AS) on the grounds that "ratio-variables in general create serious statistical analysis and interpretation problems ..." (1). They conducted a reanalysis using a different model, and concluded that consumption of added sugars had little or no association with diet quality. As discussed below, their analysis does not appear to be adequate to support this conclusion.

We agree that failure to control for total energy in the diet could theoretically confound the relation between %E_AS and nutrient intake. This could occur if %E_AS was negatively correlated with total energy intake. Because higher total energy intakes are associated with higher nutrient intakes, lower nutrient intakes at high %E_AS could be due to lower total energy intakes. They provided some support for this suggestion by showing that total energy intakes tended to be lower among those with the highest %E_AS (although this was also true for those with the lowest %E_AS). Accordingly, it would have been more appropriate for the IOM analysis to have been conducted with energy intake as a covariate, or by adjusting for total energy using the residual method. Forshee and Storey’s analysis did not use either of these methods to control total energy intake; accordingly, their conclusions are flawed.

The regression analyses they conducted, examples of which are depicted in Figures 3 and 4 of their paper, examined the effect of increasing energy from added sugars and from other sources of energy. The figures show that the predicted values for calcium consumption do not change when energy from added sugars increases from the 10th to the 90th percentile and energy from other sources is held constant at its mean. They also show the converse, i.e., that calcium intake increases when energy from other sources increases from the 10th to the 90th percentile, and energy from added sugars is held constant at its mean. Neither of these observations is surprising. They simply illustrate the intuitive concepts that 1) if you eat the same amount of everything else, and just add more sugars, nothing happens to nutrient intake; and 2) if you eat the same amount of added sugars, and add greater intakes of other nutrient-containing foods, you consume more nutrients. However, this analysis fails to control appropriately for total energy intake. In their model, total energy intake at the 90th percentile for added sugars would be above the mean total energy intake. If total energy intake was controlled, calcium intake would actually decrease from the 10th to the 90th percentiles for added sugars, because the available energy from other sources (that provide calcium) would be reduced. Thus, their model does not illustrate what occurs when added sugars displace other micronutrient-dense sources of energy.

Forshee and Storey further suggested that the spurious nature of the relation is revealed by the fact that both total energy and absolute energy from added sugars were positively associated with nutrient intake, whereas %E_AS correlated negatively, and explained less of the variance than total energy. We do not disagree with the observation that nutrient intakes tend to increase as total energy intake increases. However, this is unrelated to why the analysis presented in Appendix J of the IOM report (2) was completed in the first place. It is unreasonable to suggest that people who meet their energy requirements with appropriate physical activity increase their total energy intake to increase their nutrient intakes. Instead, one must examine the effect of different sources of energy on nutrient intakes, and that is the question addressed in Appendix J. In this regard, although the ratio variable %E_AS may have problems associated with its use, it is nevertheless the most meaningful variable to use in the context of the overall diet, particularly if total energy is controlled in the model.

We also note an additional factor that was not considered in either analysis, i.e., the food source of the added sugars. When added sugars are consumed as part of a nutrient-dense food (e.g., presweetened breakfast cereals, sweetened yogurt), they may be associated with an increase in the nutrient density of the diet (3). This association may contribute to the finding that individuals with 0–5% of energy from added sugars tend to have lower intakes of some nutrients. However, the majority of added sugars in the U.S. diet are provided by sweetened beverages such as soft drinks and fruitades, which do not contribute to nutrient density (4).

In summary, we suggest that the reanalysis essentially confirms the point made by the IOM. In the context of their usual pattern of use in the overall diet, added sugars do one of two things: 1) add energy, or 2) dilute micronutrients.

Manuscript received 15 December 2004.

LITERATURE CITED

1. Forshee, R. A. & Storey, M. L. (2004) Controversy and statistical issues in the use of nutrient densities in assessing diet quality. J. Nutr. 134:2733-2737.[Abstract/Free Full Text]

2. Institute of Medicine, Food and Nutrition Board (2002) Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids 2002 National Academy Press Washington, DC.

3. Frary, C. D., Johnson, R. K. & Wang, M. Q. (2004) Children’s choices of food and beverages high in added sugars impact their diet quality. J. Adolesc. Health 34:56-63.[Medline]

4. Guthrie, J. F. & Morton, J. F. (2000) Food sources of added sweeteners in the diets of Americans. J. Am. Diet. Assoc. 100(51):43-48.[Medline]

This article has been cited by other articles:

				R. A. Forshee and M. L. Storey Reply to Barr and Johnson J. Nutr., May 1, 2005; 135(5): 1337 - 1337. [Full Text] [PDF]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Barr, S. I. Articles by Johnson, R. K. Search for Related Content PubMed PubMed Citation Articles by Barr, S. I. Articles by Johnson, R. K.