Journal of the American College of Nutrition, Vol. 25, No. 2, 108-116 (2006)
Published by the American College of Nutrition
Changes in Calcium Intake and Association with Beverage Consumption and Demographics: Comparing Data from CSFII 1994–1996, 1998 and NHANES 1999–2002
Richard A. Forshee, PhD,
Patricia A. Anderson, MPP and
Maureen L. Storey, PhD
Center for Food, Nutrition, and Agriculture Policy, University of Maryland—College Park, College Park, MD
Address reprint requests to: Maureen L. Storey, PhD, 0220 Symons Hall, College Park, MD 20742. E-mail: storey{at}umd.edu
 |
ABSTRACT
|
|---|
Background: Consumption of soft drinks has been hypothesized to be negatively associated with calcium intake. However, fortification of some foods and beverages may have affected calcium intake.
Objective: The purpose of this study was to examine changes in calcium intake and the association of milk consumption with key beverage consumption and demographic variables using the most current data available.
Design: Several techniques were used to describe how age, gender, race/ethnicity, and beverage consumption were associated with milk and calcium intake using the Continuing Survey of Food Intake by Individuals 1994–1996, 1998 (CSFII) and the National Health and Nutrition Examination Survey 1999–2002 (NHANES). Using bivariate and multivariate regression analyses, we examined the independent relationships of total non-beverage energy intake, fluid milk consumption, non-milk beverage consumption, and demographics with calcium intake.
Results: During the time period between CSFII and NHANES, milk consumption decreased and RCSD consumption increased among children 6–11 y. Calcium intake was unaffected. Among other age categories, milk consumption either did not change or increased (females 40–59 y), while RCSD consumption increased. Calcium intake either did not change or increased in most age-gender categories, including adolescent females. Fluid milk consumption exhibited the strongest association with calcium intake. Fruit juice consumption was also positively associated with calcium intake in most age-gender categories. Consumption of other beverages, including RCSD, had little or no association with calcium intake.
Conclusions: Consumption of low-fat milk should be encouraged, but calcium fortification of certain foods and beverages and calcium supplementation may be needed to further increase calcium intake.
Key words: CSFII 1994–1996, 1998, NHANES 1999–2002, beverages, carbonated beverages, milk, calcium intake
Abbreviations: CSFII = Continuing Survey of Food Intakes by Individuals 1994–96, 98 • DCSD = diet carbonated soft drinks • NHANES = National Health and Nutrition Examination Survey • RCSD = regular carbonated soft drinks • 24 HR = twenty-four hour recall
|
INTRODUCTION
|
|---|
Adequate calcium intake is critical for maintaining bone mass [1, 2] and may have other health benefits. Some subpopulations in the United States, most notably adolescent females and young women, have an average calcium intake well below recommended levels. Inadequate calcium intake, especially among adolescent females, has remained a problem at least since the 1970s [3–5]. Calcium intake has been a focus of the 2000 and 2005 Dietary Guidelines Advisory Committees and the Institute of Medicine Dietary Reference Intake Report on Macronutrients [1, 6, 7].
The need to address low calcium and milk consumption has sparked considerable controversy. Some scientists argue that calcium intake and milk consumption are negatively associated with either added sugars or sweetened beverages [8–11], while others have found no negative associations among these variables [3, 5, 12].
Food disappearance data show that over the last several years, per capita availability of sweeteners increased while fluid milk availability declined [13]. In addition, a study by Guthrie and Morton found that the largest source of added sugars in the diet was regular carbonated soft drinks (RCSD) [14]. Although suggestive, these ecological studies cannot detect actual consumption at the individual level. They simply suggest which foods and beverages may be the biggest contributors of certain nutrients. Survey data also indicate a decrease in milk consumption and an increase in RCSD consumption [15, 16]. Since the late 1990s, however, more foods and beverages have been fortified with calcium to address the shortfall in the consumption of this important nutrient [17].
We examined changes in calcium intake and milk consumption and their associations with other key diet and demographic variables in the U.S. population using the National Health and Nutrition Examination Survey 1999–2002 (NHANES)—the latest publicly available, nationally representative consumption survey—and the Continuing Survey of Food Intake by Individuals 1994–1996, 1998 (CSFII) [18, 19]. In order to allow for data comparison between these consumption surveys, we applied the identical CSFII beverage coding system to the NHANES data. While one must always use caution when comparing results across different surveys, CSFII and NHANES provide very comparable data on dietary intake in the United States.
|
MATERIALS AND METHODS
|
|---|
The NHANES data used in our study were available from the Public-Use Data Files of the Centers for Disease Control and Prevention’s National Center for Health Statistics. The CSFII data used in our study were obtained from the U.S. Department of Agriculture (USDA). The methods used to collect the data have been reported previously [20, 21]. We analyzed calcium intake, fluid milk consumption, and non-milk beverage consumption for individuals aged 6–11, 12–19, 20–39, 40–59, and 60+ years. Males and females were analyzed separately.
In order to determine changes in calcium intake and milk consumption, we compared selected NHANES descriptive results to those obtained from CSFII. Both NHANES and CSFII were designed to provide nationally representative estimates of food and nutrient intakes for the U.S. population. The two surveys used similar 24-hour dietary recall (24 HR) instruments developed by the same team at USDA.
Associations between calcium intake, fluid milk consumption, and key diet and demographic variables were also analyzed. Beverage consumption was calculated from the Individual Foods File in NHANES. Using the USDA coding system [22], we identified the food codes that corresponded to coffee, tea, fruit juice, regular fruit drinks/ades, diet fruit drinks/ades, RCSD, diet carbonated soft drinks (DCSD), fluid milk, and alcohol. Each of these beverages were then aggregated to the individual level and merged with demographic and nutrient data. Total energy was obtained directly from NHANES. Energy from beverages was derived by summing the energy values of all beverages. Energy from non-beverage sources was calculated by subtracting energy from beverages from total energy.
We examined the distribution of RCSD consumption using histograms and boxplots. Boxplots show the interquartile range (the 25th to the 75th percentile) as a box. Lines or "whiskers" extend outward to a maximum of 1.5 times the interquartile range. Beyond that range, single dots represent extreme observations. A light circle represents the median.
Two visual techniques—the sunflower density plot and the linear regression line with associated 95% confidence interval—were used to evaluate the bivariate associations between RCSD or milk consumption and calcium intake for each age-gender category. Sunflower density plots are designed to display data when a traditional scatterplot can not give a visually accurate impression of the data because many observations overlap [23].
Finally, we conducted multiple regression analyses to determine the independent relationships between calcium intake, beverage consumption, and selected demographic variables. We examined the independent effects of consuming fluid milk, coffee, tea, beer, wine, fruit juice, RCSD, DCSD, regular fruit drinks/ades, diet fruit drinks/ades, and calories from non-beverage sources on calcium intake. This model specification allowed us to include all beverage categories while controlling for total calories using the energy decomposition method [24–26].
Demographic variables included age (years), a binary variable for gender (1 = female, 0 = male), and binary variables for race/ethnicity (non-Hispanic African American, Mexican American, other Hispanic, and non-Hispanic other race). Each race/ethnicity binary variable was set equal to 1 for individuals who chose that race/ethnicity category and 0 otherwise. The white category served as the reference category. All explanatory variables were included simultaneously in the estimated models.
We used a standard partition or decomposition model, and all energy was included. The complex survey design of NHANES was corrected with the "svyreg" procedure in Stata 8 for each model [27]. This procedure generated heteroskedastic consistent standard errors, also known as robust standard errors. Diagnostic tests for multicollinearity, non-linearity, and outliers were performed and identified one 19-year-old male that significantly affected the results. This respondent reportedly consumed 3552 g of coffee while mean coffee consumption among males 12–19 y was only 19 g. Because of the unusually large amount of coffee consumption for this individual and its impact on the model estimates, we excluded this observation. Excluding this single observation changed the coefficient estimate for coffee from 0.5 (p < 0.025) to –0.1 (p < 0.069).
The dependent variable in each model was calcium intake (mg/d). The explanatory variables in the model included beverage consumption (g), energy from non-beverage sources (kcal), and demographics. Beverages included fluid milk, fruit juice, coffee, tea, regular fruit drinks/ades, diet fruit drinks/ades, RCSD, DCSD, and alcohol. The fruit juice category included both calcium-fortified and non-fortified juices. Alcohol was omitted from the model for individuals aged 6–11 y because of very low consumption levels.
|
RESULTS
|
|---|
Descriptive Statistics
There were notable gender, age, and race/ethnicity differences in mean energy intake from non-beverage sources, calcium intake, and beverage consumption in NHANES (Table 1). Among males, mean calcium intake peaked at 1125 mg/d in the 12–19 y age category, while the highest mean calcium intake for females was 858 mg/d in the 6–11 y age category. Mean fluid milk consumption was highest in the 6–11 y age category for both males and females (281 and 228 g/d, respectively), but declined steadily in the older age categories regardless of gender. Mean RCSD consumption was highest in the 20–39 y age category at 638 and 430 g/d for males and females, respectively. Mean coffee consumption peaked in the 40–59 y age category at 482 and 374 g/d for males and females, respectively. Mean energy intake from non-beverage sources increased from childhood to young adulthood, but decreased for the 40–59 y and 60+ y age categories. At all ages, males consumed more energy from non-beverage sources than did females.
View this table:
[in this window]
[in a new window]
|
Table 1. Mean Calcium and Beverage Consumption and Demographic Percentages in NHANES 1999–2002 by Gender and Age Category*
|
|
The distribution of RCSD consumption is illustrated using histograms and boxplots for males and females aged 12–19 years (Fig. 1). Mean RCSD consumption for males and females was 1.7 and 1.1 12-oz servings per day, respectively. Each 12-oz serving was equivalent to 370 g of RCSD. The histogram showed that the highest percentage of consumers were in the 0–0.49 servings/day category. Males and females exhibited a second, smaller mode in the 1.0–1.49 servings/day category. Both histograms had a long tail indicating a skewed distribution with a few extreme values. More than 75 percent of females and about 75 percent of males consumed 2 or fewer 12-oz servings of RCSD per day. The 95th percentile of RCSD consumption was 4.7 and 3.3 12-oz servings per day for males and females, respectively. These values are likely to be overestimates of the true long-term 95th percentile since 24 HR data have been shown to overestimate consumption in the higher percentiles compared to long-term consumption patterns [28–30]. However, the 24 HR instrument is also subject to measurement error, which may lead to under- or over-reporting of certain foods.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1. Histograms and Box Plots of RCSD Consumption, Ages 12–19 y. The histograms show the consumption of RCSD for females and males 12–19 y. Female adolescents consume an average of 1.1 12-oz. servings per day and males consume an average of 1.7 12-oz. servings per day.
|
|
Calcium, fluid milk, and RCSD consumption from NHANES were compared to those from CSFII (Fig. 2). Mean RCSD consumption increased significantly in all age-gender categories, except for males 12–19 y. Mean milk consumption did not change in most age-gender categories, except for a significant increase for females 40–59 y. Among males and females 6–11 y, mean milk consumption decreased significantly, yet calcium intake did not change. Calcium intake was also unchanged among males 12–19 y. Calcium intake increased significantly among all other age-gender categories, including adolescent females. Although mean RCSD consumption increased for most age-gender categories, the increase in RCSD consumption did not appear to be at the expense of milk consumption or calcium intake. Mean milk consumption increased or remained unchanged in some of the same age-gender categories that showed an increase in mean RCSD consumption.
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 2. Mean Consumption of calcium, milk, and RCSD in CSFII 1994–96, 1998 and NHANES 1999–2002. The dark gray horizontal bars represent the mean of calcium consumption (mg/d) and milk and RCSD consumption (g/day) in CSFII 1994–96, 1998. The light gray horizontal bars represent the mean of calcium consumption (mg/d) and milk and RCSD consumption (g/day) in NHANES 1999–2002. The error bars represent the 95% confidence interval of each estimate.
|
|
Bivariate Analyses
We examined the bivariate associations between either RCSD or milk consumption and calcium intake for each age-gender category. Two visual techniques—the sunflower density plot and the linear regression line with associated 95% confidence interval—were used to evaluate these relationships (Fig. 3a–d). We restricted our discussion of the bivariate analyses to individuals aged 12–19 y due to space limitations. This age category exhibited the largest shortfall in calcium intake compared to their recommended levels, particularly among females. The charts for all other age-gender categories were similar.
View larger version (44K):
[in this window]
[in a new window]
|
Fig. 3. 3a. Calcium Intake and RCSD Consumption, Males 12–19 y. Sunflower density plots of RCSD consumption and calcium intake among adolescent males. 3b. Calcium Intake and Milk Consumption, Males 12–19 y. Sunflower density plots of milk consumption and calcium intake among adolescent males. 3c. Calcium Intake and RCSD Consumption, Females 12–19 y. Sunflower density plots of RCSD consumption and calcium intake among adolescent females. 3d. Calcium Intake and Milk Consumption, Females 12–19 y. Sunflower density plots of milk consumption and calcium intake among adolescent females. Density-distribution sunflower plots with linear fit regression line, and 95% confidence interval of calcium intake (mg/d) predictions for RCSD and milk consumption (g/day) using the 24-hour dietary recall from NHANES 1999–2002. A single observation is represented by a single point. Multiple observations at the same point on the graph are represented as either a light or dark "flower" with "petals." Light flowers have a clear background, and dark flowers have a gray background. Dark patches with many petals indicate areas of the plot where observations are extremely prevalent or densely clustered. The long-dashed line represents the linear regression (weighted) of RCSD or milk consumption on calcium intake, and the pair of short-dashed lines above and below the regression line represents the 95% confidence interval of the calcium intake prediction.
|
|
The sunflower density plot showed a wide range in calcium intake for both adolescent males and females throughout the range of RCSD consumption (Fig. 3a and 3c). No association between RCSD consumption and calcium intake was observed. The sunflower density plot for milk consumption and calcium intake indicated a strong, positive association for both males and females (Fig. 3b and 3d).
As expected, the bivariate regression results of milk consumption on calcium intake showed a strong, positive relationship. The estimated coefficient was 1.3 (p < 0.01) for both males and females, and the adjusted R2 was 0.48 and for males and 0.46 for females. The figures for the other age-gender categories showed similar bivariate relationships.
A weighted bivariate linear regression between RCSD consumption and calcium intake found no association. The coefficients were 0.07 (p < 0.10) and –0.01 (p < 0.70) for males and females, respectively. The adjusted R2 for both models was 0.00.
Regression Analyses
Like the bivariate analyses, the regression analyses showed that fluid milk consumption had the strongest relationship with calcium intake (Table 2). Each additional gram of fluid milk consumed was associated with a 1.2 mg increase in calcium intake for males and a 1.1 to 1.3 mg increase in calcium intake for females. This model predicts that an additional 8-oz serving of fluid milk (244 g) would increase calcium intake by 293 mg for males and by 268 to 317 mg for females. Since an 8-oz serving of low-fat fluid milk contains 288 mg of calcium, [31] the regression estimates are a very plausible reflection of actual calcium levels in fluid milk. Energy from non-beverage sources had a significant, though smaller, relationship with calcium consumption among males and females. In many of the age-gender categories, fruit juice had a small, but statistically significant, association with calcium intake. This may reflect the increasing number of fruit juices that are fortified with calcium.
View this table:
[in this window]
[in a new window]
|
Table 2. Multivariate Regression Results: The Association of Calcium Consumption with Beverages Intakes in NHANES 1999–2002, MALES
|
|
The multivariate regression models among males aged 6–11, 12–19, 20–39, 40–59, and 60+ years explained 66, 72, 59, 66, and 66 percent of the variance in calcium consumption, respectively (Table 2). Consumption of fluid milk had the strongest and most consistent relationship with calcium consumption. Consumption of fruit juice was positively associated with calcium intake in 3 of these 5 age categories and approached significance in a fourth age category (p < 0.058 for 40–59 y). Energy from non-beverage sources had a significant, positive association with calcium consumption across all of these age categories. For males 40–59 y, DCSD consumption was positively associated and age was negatively associated with calcium intake. Coffee consumption was negatively associated with calcium intake for males 6–11 y, and tea consumption was negatively associated with calcium intake for males 6–11 y and 20–39 y. Being African American was associated with significantly lower calcium consumption among all of the age categories. For example, African-American males aged 20–39 y were predicted to consume 237 mg less calcium than their white counterparts after controlling for all types of beverage consumption and other sources of energy in the diet. Other race/ethnicity variables (i.e. not white or African American) tended to have negative relationships with calcium intake, but these associations were not consistently significant.
Similarly, the multivariate regression models for females aged 6–11, 12–19, 20–39, 40–59, and 60+ years explained 67, 72, 70, 67, and 64 percent of the variance in calcium consumption, respectively (Table 3). Consumption of fluid milk again had the strongest and most consistent relationship with calcium consumption. Consumption of fruit juice was positively associated with calcium intake in 4 of these 5 age categories. RCSD consumption was negatively associated with calcium intake for females 40–59 y, and DCSD consumption was negatively associated with calcium intake for females 6–11 y. Energy from non-beverage sources had a significant, positive association (ranging from 0.3 to 0.4 mg/kcal) with calcium consumption across all of these age categories. As with males, African-American females were predicted to consume significantly less calcium than their white counterparts. African-American females aged 20–39 y were predicted to consume 150 mg less calcium than white females aged 20–39 y. Mexican-American and other Hispanic females aged 20–39 y were also predicted to consume significantly less calcium than their white counterparts, but there were no consistent and significant differences among the other age categories. As before, the lower predicted calcium consumption for these race/ethnicity categories remained after controlling for all types of beverage consumption and other sources of energy in the diet.
View this table:
[in this window]
[in a new window]
|
Table 3. Multivariate Regression Results: The Association of Calcium Consumption with Beverages Intakes in NHANES 1999–2002, FEMALES
|
|
Limitations
CSFII and NHANES are large, well-executed, nationally representative surveys that are widely used in health research. Like any data set, however, they have certain limitations. The 24 HR used has been extensively tested, but it is likely to have some measurement error for the variables used in this analysis. Beverage consumption may be underestimated because respondents fail to remember, or choose not to report, certain eating/drinking occasions. However, the multi-pass system used in both surveys attempts to minimize underreporting.
Calcium intake may be underestimated because of incomplete information on calcium-fortified juices and other calcium-fortified foods, such as cereals and breads. CSFII and NHANES do include food codes for calcium-fortified juices and other foods, but they may be underreported since many juices are coded as "not specified further." In addition, the nutrient database may not be completely up-to-date with the quickly changing food market and fewer foods were fortified during the earlier CSFII survey. It is important for the nutrient database to accurately and quickly reflect changing product formulations, and for the 24 HR instrument to appropriately probe for consumption of fortified products, in order to obtain accurate estimates of calcium and other nutrient consumption. This is likely to be an ongoing concern in nutrition monitoring if trends in fortification continue.
A significant proportion of the population obtains additional calcium through supplements. Thirty-five percent of adults in NHANES reported taking a multivitamin/multimineral in the past month [32]. At the time of this analysis, data on supplement use were not available for the 2001–2002 survey years, so calcium intake from supplements was not included in our models. Because of these limitations, caution must be used in drawing conclusions.
Another limitation is that the cross-sectional nature of the survey prevents the analysis of how changes in beverage consumption are related to changes in calcium intake over time. Finally, while we controlled for the major factors associated with calcium intake, it is possible that other unmeasured variables may affect our results.
|
DISCUSSION
|
|---|
Calcium intake from foods and beverages has increased since CSFII in some critical age-gender categories, including 12–19 y, 20–39 y, and 40–59 y females. This finding is important given the number of earlier studies that have consistently found decreases in milk consumption, particularly among adolescents [4, 8]. However, calcium intake is still below recommended levels for some age categories.
These results show that calcium intake is largely determined by consumption of fluid milk and that there is no direct, negative association between calcium intake and RCSD consumption. Consumption of most non-milk beverages has little or no association with calcium intake, except for fruit juice. In our study, fruit juice had a positive association with calcium intake in most age-gender categories. These results could reflect specific food intake patterns, such as consumption of calcium-fortified orange juice in addition to cereal and milk for breakfast. Studies have shown that eating breakfast has a positive impact on milk intake [4].
Calcium-fortified fruit juices and other foods may be part of the reason for the increased calcium intake observed in the data. Calcium fortification of foods began in the late 1980s, and the number of fortified foods grew significantly during the late 1990s. Few foods were fortified with calcium in 1991. In fact, only four USDA food codes were available for calcium-fortified beverages at that time, and breakfast cereals were not routinely fortified with calcium [17]. Consumer awareness did not increase until the late 1990s. In 1997, only 10% of consumers reported trying to increase their use of fortified foods. By 1999, 66% of consumers reported trying to increase their use of fortified foods [33].
We found a slightly positive association between fruit juice consumption and calcium intake in most age-gender categories. As discussed under limitations, this may be underestimated because of underreporting of calcium-fortified juices. This illustrates the potential role that calcium fortification could play in addressing low calcium intake. The dietary supplement use database for NHANES 2001–2002 was not available at the time of our analysis, but the current use of calcium supplements should also be carefully studied in future research.
These results have important implications for dietary and policy interventions designed to increase calcium consumption. As we have shown previously, African Americans consume fewer milk products and less RCSD, but more regular fruit drinks/ades, than do whites [34]. The data from NHANES demonstrate that, on average, females—especially African Americans and those in the Other Race/Ethnicity category—do not consume recommended levels of calcium or milk products. These results mirror those published by Breifel and Johnson using NHANES 1999–2000. "In 1999–2000, only 30% of the population age 2 and older met the recommendations for milk and other dairy food consumption [35]."
|
CONCLUSION
|
|---|
Dietary guidance and nutrition professionals should continue to encourage the consumption of low-fat dairy products to increase calcium intake. However, we must also acknowledge that after decades of encouragement, some important subpopulations still do not consume enough low-fat dairy products to meet their calcium requirements. Strategies that make low-fat dairy foods, especially flavored or regular fluid milk, more readily available and sold in attractive, "kid-friendly" containers in schools offer more choices and may encourage greater consumption. Augmenting these strategies with carefully chosen calcium-fortification programs and targeted calcium supplementation to the most at-risk individuals may be necessary to reduce the risks of inadequate calcium intake.
|
ACKNOWLEDGMENTS
|
|---|
This project was supported by an unrestricted gift from the American Beverage Association.
|
FOOTNOTES
|
|---|
This project was supported by an unrestricted gift from the American Beverage Association.
Received April 5, 2005.
Accepted December 9, 2005.
|
REFERENCES
|
|---|
- Food and Nutrition Board, Institute of Medicine, National Academy of Sciences: Dietary Reference Intakes: Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein and Amino Acids. Washington, DC: National Academy Press,2002
.
- Fishbein L: Multiple sources of dietary calcium—some aspects of its essentiality. Reg Tox Pharm39
:67
–80,2004
.
- Storey ML, Forshee RA, Anderson PA: Associations of adequate intake of calcium with diet, beverage consumption, and demographic characteristics among children and adolescents. J Am Coll Nutr23
:18
–33,2004
.[Abstract/Free Full Text]
- Bowman S: Beverage choices of young females: Changes and Impact on nutrient intakes. J Am Dietetic Assoc102
:1234
–9,2002
.[Medline]
- Park YK, Meier ER, Bianchi P, Song WO: Trends in children’s consumption of beverages: 1987 to 1998. Fam Econ Nutr Rev14
:69
–79,2002
.
- Department of Health & Human Services: 2005 Dietary Guidelines Advisory Committee Report, U.S. Internet: http://www.healthierus.gov/dietaryguidelines/ (Accessed August 10, 2005.)
- U.S. Department of Health & Human Services: The Report of the Dietary Guidelines Advisory Committee on Dietary Guidelines for Americans, 2000. Internet: http://www.health.gov/dietaryguidelines/dgac/ (Accessed August 10, 2005.)
- Harnack L, Stang J, Story M: Soft drink consumption among US children and adolescents: nutritional consequences. J Am Diet Assoc99
:436
–441,1999
.[Medline]
- Johnson RK, Frary CD, Wang MQ: The nutritional consequences of flavored-milk consumption by school-aged children and adolescents in the United States. J Am Diet Assoc102
:853
–856,2002
.[Medline]
- Moshfegh AJ: Food and nutrient intakes by individuals in the United States, by region 1994–96. U.S. Department of Agriculture. December1999
.
- Johnson RK, Panely C, Wang MQ: The association between noon-time beverage consumption and the diet quality of school age children. J Child Nutr Mgmt2
:95
–100,1998
.
- Barr SI: Associations of social and demographic variables with calcium intakes of high school students. J Am Dietetic Assoc94
:260
–266,1994
.[Medline]
- U.S. Department of Health & Human Services: The Report of the Dietary Guidelines Advisory Committee on Dietary Guidelines for Americans, 2000. Internet: http://www.health.gov/dietaryguidelines/dgac/ (Accessed August 10, 2005.)
- Guthrie JF, Morton JF: Food sources of added sweeteners in the diets of Americans. J Am Dietetic Assoc100
:43
–51,2000
.[Medline]
- Gleason P, Suitor C: Changes in Children’s Diets: 1989–1991 to 1994–1996. USDA Food and Nutrition Service Special Nutrition Programs Report #CN-01-CD2. January2001
.
- Department of Health and Human Services, National Center for Health Statistics: 1999–2002 National Health and Nutrition Examination Survey. May2004
. Internet: http://www.cdc.gov/nchs/about/major/nhanes/nhanes99_00.htm (Accessed August 30, 2005.)
- Berner LA, Clydesdale FM, Douglass JS: Fortification contributed greatly to vitamin and mineral intakes in the United States, 1989–1991. Nutrition131
:2177
–2183,2001
.
- US Department of Agriculture, Agricultural Research Service. 1994–96, 98 Continuing Survey of Food Intakes by Individuals 1998 and 1994–96. CD-ROM, Washington, DC,1998
.
- Center for Disease Control and Prevention: National Center for Health Statistics. 1999–2000 National Health and Nutrition Examination Survey: Dietary Interview (Individual Foods File). May2004
. Internet: http://www.cdc.gov/nchs/about/major/nhanes/nhanes99_00.htm (Accessed October 13, 2005.)
- Center for Disease Control and Prevention. National Center for Health Statistics: 1999–2000 National Health and Nutrition Examination Survey: Dietary Interview (Total Nutrients). May2004
. Internet: http://www.cdc.gov/nchs/about/major/nhanes/nhanes99_00.htm (Accessed October 13, 2005.)
- US Department of Agriculture, Agricultural Research Service: 1994–96 Continuing Survey of Food Intakes by Individuals and 1994–96 Diet and Health Knowledge Survey and related materials. CD-ROM, Washington, DC,1998
.
- U.S. Department of Agriculture, Agricultural Research Service: Continuing Survey of Food Intakes by Individuals 1994–96, 1998. CD-ROM, Washington, DC,2000
.
- Silverman BW: "Density Estimation for Statistics and Data Analysis." London: Chapman and Hall,1986
.
- Willett WC, Stampfer M: Implications of total energy intake for epidemiologic analysis. In Willett WC (ed): "Nutritional Epidemiology," 2nd ed. New York, NY: Oxford University Press, pp273
–30,1998
.
- Kipnis V, Freedman LS, Brown CC, Hartman A, Schatzkin A, Wacholder S: Interpretation of energy adjustment models for nutritional epidemiology. Am J Epidemiol137
:1376
–1380,1993
.[Abstract/Free Full Text]
- Mackerras D: Energy adjustment: the concepts underlying the debate. J Clin Epidemiol49
:957
–962,1996
.[Medline]
- Stata Corporation: Stata statistical software. Release 8.0.2004
.
- Tran NL, Barraj L, Smith K, Javier A, Burke TA: Combining food frequency and survey data to quantify long-term dietary exposure: a methyl mercury case study. Risk Anal24
:19
–30,2004
.[Medline]
- Nusser SM, Carriquiry AL, Fuller WA: A Semiparametric Transformation Approach to Estimating Usual Daily Intake Distributions. Research Agreement No. 58-3198-90-032 and Cooperative Agreement No. 58-3198-2006, HNIS, USDA and Center for Agriculture & Rural Development, Iowa State University,1993
.
- Carriquiry AL, Jensen H, Dodd KW, Nusser SM, Borred LG, Fuller WA: Estimating usual intake distributions. Journal Paper No. J-14654 of the Iowa Agriculture and Home Economics Experiment Station, Ames, Iowa. Project No. 2806,1992
.
- U.S. Department of Agriculture, Agricultural Research Service: USDA National Nutrient Database for Standard Reference, Release 17. Nutrient Data Laboratory Home Page, 2004. http://www.nal.usda.gov/fnic/foodcomp (Accessed 13 October 2005.)
- Radimer K, Bindewald B, Hughes J, Ervin B, Swanson C, Picciano MF: Dietary supplement use by US adults: data from the National Health and Nutrition Examination Survey, 1999–2000. Am J Epidemiol160
:339
–349,2004
.[Abstract/Free Full Text]
- Bishai D, Nalubola R: "The History of Food Fortification in the United States: Its Relevance for Current Fortification Efforts in Developing Countries. Economic Development and Cultural Change." Chicago: The University of Chicago,2002
.
- Forshee RA, Storey ML: Total beverage consumption and beverage choices among children and adolescents. Intl J Food Sci Nutr54
:297
–307,2003
.
- Briefel R, Johnson C: Secular trends in dietary intake in the United States. Ann Rev Nutr24
:401
–431,2004
.[Medline]
This article has been cited by other articles:
|
|
|
|
 
P. K. Newby
Are Dietary Intakes and Eating Behaviors Related to Childhood Obesity? A Comprehensive Review of the Evidence
J. Law Med. Ethics,
March 1, 2007;
35(1):
35 - 60.
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
