The Relationship of Breakfast and Cereal Consumption to Nutrient Intake and Body Mass Index: The National Heart, Lung, and Blood Institute Growth and Health Study

RESEARCH Current Research

The Relationship of Breakfast and Cereal Consumption to Nutrient Intake and Body Mass Index: The National Heart, Lung, and Blood Institute Growth and Health Study BRUCE A. BARTON, PhD; ALISON L. ELDRIDGE, PhD, RD; DOUGLAS THOMPSON, PhD; SANDRA G. AFFENITO, PhD, RD; RUTH H. STRIEGEL-MOORE, PhD; DEBRA L. FRANKO, PhD; ANN M. ALBERTSON, MS, RD; SUSAN J. CROCKETT, PhD, RD

ABSTRACT Objective To describe changes in breakfast and cereal consumption of girls between ages 9 and 19 years, and to examine the association of breakfast and cereal intake with body mass index (BMI) and consumption of nutrients. Design Data from the National Heart, Lung, and Blood Institute Growth and Health Study, a longitudinal biracial observational cohort study with annual 3-day food records. Subjects/setting The National Heart, Lung, and Blood Institute Growth and Health Study recruited 2,379 girls (1,166 white and 1,213 black), ages 9 and 10 years at baseline, from locations in the Berkeley, CA; Cincinnati, OH; and Washington, DC, areas. Main outcome measures Frequency of consumption of breakfast (including cereal vs other foods) and cereal; BMI; and dietary fat, fiber, calcium, cholesterol, iron, folic acid, vitamin C, and zinc. Statistical analyses Generalized estimating equations methodology was used to examine differences in the frequency of breakfast and cereal eating by age. Generalized

B. Barton is president and chief executive officer and D. Thompson is a senior research scientist, Maryland Medical Research Institute, Baltimore. A. Eldridge is manager, Nutrition Science, A. Albertson is a senior nutrition scientist, and S. Crockett is senior director, General Mills, Bell Institute of Nutrition, Minneapolis, MN. S. Affenito is an assistant professor, Department of Nutrition and Family Studies, Saint Joseph College, West Hartford, CT. R. Striegel-Moore is a professor, Department of Psychology, Wesleyan University, Middletown, CT. D. Franko is a professor, Department of Counseling and Educational Psychology, Northeastern University, Boston, MA. Address correspondence to: Douglas Thompson, PhD, Maryland Medical Research Institute, 600 Wyndhurst Ave, Baltimore, MD 21210. E-mail: dthompson@ mmri.org Copyright © 2005 by the American Dietetic Association. 0002-8223/05/10509-0004$30.00/0 doi: 10.1016/j.jada.2005.06.003

© 2005 by the American Dietetic Association

estimating equations and mixed models were used to examine whether breakfast and cereal consumption were predictive of BMI and nutrient intakes, adjusting for potentially confounding variables. Results Frequency of breakfast and cereal consumption decreased with age. Days eating breakfast were associated with higher calcium and fiber intake in all models, regardless of adjustment variables. After adjusting for energy intake, cereal consumption was related to increased intake of fiber, calcium, iron, folic acid, vitamin C, and zinc, and decreased intake of fat and cholesterol. Days eating cereal was predictive of lower BMI. Conclusions Cereal consumption as part of an overall healthful lifestyle may play a role in maintaining a healthful BMI and adequate nutrient intake among adolescent girls. J Am Diet Assoc. 2005;105:1383-1389.

C

hildren benefit from breakfast in a myriad of ways. Research supports a role for breakfast in improved cognition and learning in children (1,2). In addition to the cognitive benefits, there are nutritional benefits as well. Breakfast consumption has been consistently associated with favorable nutrient intakes and improved diet quality in children and adolescents (3-7). Trends since 1965 indicate that breakfasts of children and adolescents now include more low-fat dairy foods; more whole-grain breads and cereals; and more citrus fruits, other fruits, and juices (6). Children who consume breakfast are much more likely to meet recommended intakes for vitamins and minerals than those who skip breakfast (8-10). The relationship between breakfast and positive health outcomes may be due to the specific foods consumed at breakfast, rather than breakfast per se (11). In children, cereal ranks high among breakfast foods eaten (5,10). Based on a single 24-hour recall, 21% of the 10-year-olds from the Bogalusa Heart Study reported eating ready-toeat cereal (RTEC) sometime during the day, most often at breakfast (12). Over a 14-day period, more than 90% of 4to 12-year-olds reported eating RTEC at least once (10). Breakfasts that include cereal (“cereal breakfasts”) may have a particularly positive effect on overall nutrient intake. Most cereals are fortified with essential nutrients and many cereals provide dietary fiber. Children who eat

Journal of the AMERICAN DIETETIC ASSOCIATION

1383

Table 1. Number and percent of National Heart, Lung, and Blood Institute Growth and Health Study participants by basic demographic characteristics White

Total enrolled Annual family income ($) ⬍10,000 10,000-19,999 20,000-29,999 30,000-39,999 40,000-49,999 50,000-74,999 ⱖ75,000 Unknown Maximum education of household Did not complete high school High school graduate/equivalent Post high school College 1-3 y College graduate Graduate school Unknown Number of girls living with both natural parents

%

No.

%

Total No.

1,166

49.1

1,213

50.9

2,379

88 105 174 185 185 258 115 56

7.5 9.0 14.9 15.9 15.9 22.1 9.9 4.8

317 218 182 153 94 147 24 78

26.1 18.0 15.0 12.6 7.8 12.1 2.0 6.4

405 323 356 338 279 405 139 134

48 188 49 302 198 380 1 786

4.1 16.1 4.2 25.9 17.0 32.6 0.1 67.4

107 275 68 506 121 135 1 508

8.8 22.7 5.6 41.7 10.0 11.1 0.1 41.9

155 463 117 808 319 515 2 1,294

cereal consume significantly less fat and cholesterol (3,10). Less is known about the relationship between breakfast, cereal consumption, and body mass index (BMI) in children. In cross-sectional studies, breakfast skipping is significantly more common among overweight and obese children and adolescents compared with normal-weight youth (13-15). In addition, children who consumed RTEC eight or more times in 2 weeks had significantly lower BMIs than those consuming three or fewer servings in 14 days (10). One longitudinal study examined changes in breakfast consumption in a cohort of females between the ages of 9 and 19 years (7). Girls ate breakfast less often as they grew older. Frequency of breakfast consumption was predictive of BMI in models controlling for study site, ethnicity, and age. However, the effect of breakfast fell below significance when physical activity, total energy intake, and parental education were added to the model, suggesting that breakfast consumption may be a marker for lifestyle and/or socioeconomic factors that are related to BMI. Although these studies suggest a beneficial association between breakfast, cereal, and BMI in children, no longitudinal studies have explored the frequency and health outcomes of both cereal and breakfast consumption. The first objective of the current research, therefore, was to describe the frequency of breakfast and cereal consumption, separately and in combination, among black and white girls over a 10-year period (from childhood through adolescence). A second objective was to examine the effect that cereal breakfast had on nutrient intake for key nutrients including calcium, total fat, fiber, and cholesterol. The third objective was to examine the relationships between breakfast, cereal consumption, and BMI as these girls matured.

1384

September 2005 Volume 105 Number 9

Black

No.

MATERIALS AND METHODS As previously reported, the National Heart, Lung, and Blood Institute (NHLBI) Growth and Health Study is a 10-year longitudinal study of 2,379 girls who were 9 or 10 years old at study entry (Table 1) (16). Girls were recruited from three study sites: University of California at Berkeley; University of Cincinnati/Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; and Westat Inc/Group Health in Rockville, MD. The study protocol was approved by the Institutional Review Boards of all participating sites. All girls who entered the Growth and Health Study had assented and a parent or guardian consented to their participation. Only instruments of relevance to the present report are described here. Procedures and Measures Three-day food records that had been previously validated (17) were collected at visits (years) 1 through 5 and then again at visits 7, 8, and 10. Dietitians instructed girls to record all food and drink and time of intake for 3 consecutive days (2 weekdays and 1 weekend day). Dietitians reviewed completed food records individually with the girls, using standard probes to clarify incomplete responses. Food records were coded and analyzed for nutrient content (18). Nutrient values were updated annually to reflect changes in the nutrient composition of individual foods. BMI was calculated based on measurements of girls’ height and weight by trained research staff. Two BMIbased measures were used. First, BMI-for-age z scores indicated girls’ BMI relative to other girls of the same age. Second, based on Centers for Disease Control and Prevention guidelines (http://www.cdc.gov/nccdphp/dnpa/ bmi/bmi-for-age.htm), “at risk of overweight” was coded

Table 2. Sample size, percentage of girls reporting cereal consumption on 0, 1, 2, or 3 days of the 3-day food diary, and mean number of days that cereal was consumed (out of 3 possible days), by age in the National Heart, Lung, and Blood Institute Growth and Health Study Days of Cereal Consumption (Out of 3 Possible Days) Age (y)

Sample size

9 10 11 12 13 14 15 16 17 18 19

1,015 2,034 1,879 1,815 1,731 877 829 1,456 772 876 963

0 1 2 3 4™™™™™™™™™™™™™™™™™™™™ % ™™™™™™™™™™™™™™™™™™™3 20.0 27.2 32.5 20.3 24.3 29.4 29.0 17.3 31.2 27.7 25.4 15.7 33.2 28.8 24.9 13.1 38.9 28.1 21.1 11.8 41.5 30.1 19.5 8.9 41.6 27.6 20.7 10.0 44.2 28.3 18.5 9.0 45.3 28.4 15.7 10.6 43.6 26.6 20.2 9.6 43.7 26.9 20.1 9.2

as 1 (at or more than the age- and sex-specific 85th percentile of BMI) or 0 (less than the 85th percentile). Girls’ ages were recorded as age at last birthday. Race (black or white) was self-reported. Girls were categorized as being from one- or two-parent households. The highest level of parental education for either parent was categorized as 4 or more years of college vs less than 4 years. Physical activity was assessed using the Habitual Activity Questionnaire (19). A physical activity score was computed by multiplying an estimate of metabolic equivalents for the recorded activities by the weekly frequency, duration, and fraction of the year during which activities were performed. Analysis Breakfast was defined as eating between 5 AM and 10 AM on weekdays or 5 AM to 11 AM on weekends. Variables were constructed to indicate the number of days (out of 3 reported annually) that a girl consumed: (a) breakfast without cereal (days of “noncereal breakfast”); (b) cereal at breakfast time, both ready-to-eat and cooked (days of “cereal breakfast”); (c) cereal at any time of the day, including breakfast time and all other times (“cereal days”); and (d) no cereal at any time during the day (“noncereal days”). For each girl at each age, nutrient intake per day at breakfast time was estimated for cereal breakfasts and noncereal breakfasts, while nutrient intake per day over the entire course of a day was estimated for cereal days and noncereal days. To compare breakfasts/days including cereal with breakfasts/days without cereal, a separate model was created for each nutrient, with type of breakfast/day as the predictor of interest, controlling for potential confounders (age, site, and mean daily energy intake). The analyses were implemented with PROC MIXED in SAS version 9.1 (2002-2003, SAS Institute, Cary, NC), taking into account clustering due to multiple measurement periods (years) within girls. Further analyses examined the effects on BMI of number of days eating breakfast and number days eating cereal. Because there were the repeated measurements

Mean days of cereal consumption 1.5 1.4 1.3 1.2 1.1 1.0 1.0 0.9 0.9 1.0 0.9

for each girl, generalized estimating equations methodology was used (PROC GENMOD in SAS), with appropriate distribution and link functions to model the binary measure (at risk of overweight) and the continuous measure (BMI-for-age z scores). Days eating breakfast, days eating cereal, and age were represented as ordinal variables. Type III Wald ␹2 tests were used to test the significance of predictors in the generalized estimating equations models.

RESULTS Age Trends in Breakfast and Cereal Consumption The number of days of breakfast consumption and the number of days of cereal consumption were computed for each girl at each age (Objective 1). Detailed data on days of breakfast consumption by age in the same cohort of girls used in the present analysis were presented by Affenito and colleagues (7). Days of breakfast consumption decreased significantly as girls grew older [repeatedmeasures Mantel-Haenzel test of nonzero correlation: ␹2(10)⫽1,603.09, P⬍.0001]. Similar to the pattern for breakfast consumption, days of cereal consumption decreased significantly as girls grew older [repeated-measures Mantel-Haenzel test of nonzero correlation, ␹2(10)⫽582.32, P⬍.0001 (Table 2)]. At each year of age, the percentage of days that each girl exhibited the following breakfast patterns was computed: (a) no breakfast; (b) breakfast including cereal; and (c) breakfast not including cereal. The frequency of no breakfast and breakfast with and without cereal is depicted in Figure 1. The percentage of days that girls ate breakfast not including cereal tended to decrease with age [Type III Wald ␹2(10)⫽108.20, P⬍.0001, for the main effect of age in generalized estimating equations]. The patterns for both breakfast skipping and cereal breakfast showed more dramatic shifts, with younger girls much more often eating cereal breakfasts [Wald ␹2(10)⫽654.04, P⬍.0001] and older girls skipping a significantly greater percentage of breakfasts entirely [Wald ␹2(10)⫽958.11, P⬍.0001]. September 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION

1385

Figure 1. Mean and 95% confidence interval of percent of days (out of 3 possible) that girls reported consuming no breakfast, breakfast not including cereal, and breakfast including cereal, by age in the National Heart, Lung, and Blood Institute Growth and Health Study.

Association of Cereal Consumption with Nutrient Intake Food choice at breakfast (specifically, cereal vs other foods consumed at breakfast) is likely to impact the nutritional quality of the meal. To examine this issue, the nutrient contribution of eating that included cereal was compared with the nutrient content of eating that did not include cereal (Objective 2). Nutrient content was analyzed separately for foods eaten at breakfast time (ie, “cereal breakfasts” were compared with “noncereal breakfasts”) and for foods eaten throughout the day, including breakfast as well as other times of the day (ie, “cereal days” were compared with “noncereal days”). To estimate the association of cereal consumption with nutrient intake, adjusted means were computed for eight key nutrients: fat, calcium, fiber, cholesterol, iron, folic acid, vitamin C, and zinc (Table 3). The same findings were present for foods eaten at breakfast time and foods eaten throughout the day. Specifically, compared with eating that did not involve cereal (ie, noncereal breakfasts and noncereal days), eating that involved cereal (ie, cereal breakfasts and cereal days) had more fiber, calcium, iron, folic acid, vitamin C, and zinc (all P⬍.0001), and less fat and cholesterol (all P⬍.0001). Cereal Consumption and Body Mass Index The third objective of this study was to explore the relationship among breakfast, cereal consumption, and BMI. A graphical representation of the relationship between cereal consumption and raw BMI is presented in Figure 2. This graph shows unadjusted means, whereas the models described later control for other factors that may impact BMI in addition to age and cereal consumption. The unadjusted values clearly demonstrate that BMI increased significantly with age. Girls who ate cereal on 3 days out of 3 possible days had lower BMI than girls who did not eat cereal. A similar trend was seen for breakfast consumption, with those consuming breakfast on 3 days having lower BMIs than girls who skipped breakfast on all or most days (data not shown). The next step was to model the relationship between

1386

September 2005 Volume 105 Number 9

breakfast and cereal consumption and BMI. In separate models with the same set of predictors, BMI was represented as continuous (BMI-for-age z scores) and as binary (at risk of overweight vs not). As predictors, all models included the number of days of breakfast and number of days of cereal consumption (out of 3) reported by each girl. Site and age were included as control variables. All models included factors likely to influence BMI: family context (number of parents in the household and parental education, a proxy for socioeconomic status); race (black or white); physical activity score; and total daily energy intake averaged over the 3 days of the food records. Interactions of theoretical interest (ie, age and race by days eating cereal, age and race by days eating breakfast, and days eating cereal by days eating breakfast) were included in preliminary versions of the models, but because they were not predictive of either BMI-based measure, they were not included in the final models. Days eating cereal was predictive of BMI z scores as well as risk of overweight [␹2(3)⫽14.35 and 11.62 for BMI z scores and risk of overweight, respectively, P⬍.01] but days eating breakfast was not predictive of either BMI indicator [␹2(3)⫽4.54 and 4.98, P⬎.17]. Figure 3 shows predicted BMI z scores for 14-year-olds who ate breakfast on 2 days, by number of days eating cereal, controlling for the other variables in the models. Data from 14-year-olds were chosen because it is in the middle of the age range of the study participants, and 2 days of breakfast was chosen because that is the average for 14-year-olds (the effects of age and days of breakfast were additive in this model, so different values would simply shift the predicted values up or down). Post-hoc contrasts showed that for BMI z scores, girls who ate cereal on 3 days had lower scores than did girls who ate cereal 0, 1, or 2 days (P⬍.05), while girls who ate cereal 0, 1, or 2 days did not differ from one another (P⬎.15). In a follow-up model with days eating cereal treated as a continuous variable but all else held constant, the coefficient for cereal was significant and negative (␤⫽⫺.015, P⬍.001), indicating a predicted BMI z score decrease of 0.015 for each additional day eating cereal (Figure 3). For the model of risk of overweight, girls who ate cereal on 1, 2, or 3 days had lower rates of risk than did girls who ate cereal on 0 days (P⬍.05), while girls who ate cereal on 1, 2, or 3 days did not differ from one another (P⬎.14). Compared with girls who ate cereal on 0 days, girls who ate cereal on 1, 2, or 3 days were 0.93, 0.90, and 0.87 times as likely to be at risk of overweight; that is, eating cereal on 1 or more days resulted in a reduction in risk of overweight. For both BMI indicators, there was a significant effect of age [␹2(10)⫽397.83 and 54.19, P⬍.0001] due to differences among individual ages, but there was no systematic trend across the entire age range. Both BMI indicators were higher among girls with less-educated parents [␹2(1)⫽10.07 and 15.91, P⬍.005] and among girls in oneparent households [␹2(1)⫽6.96 and 5.52, P⬍.05]. Greater physical activity was predictive of lower BMI z scores and risk of overweight [␹2(1)⫽5.32 and 9.40, P⬍.05]. Higher energy intake was predictive of greater BMI z scores [␹2(1)⫽12.90, P⬍.0005], but the independent association

Table 3. Mean estimated daily intake of nutrients and nutrients at breakfast time, on days/breakfasts including cereal and days with days/breakfasts not including cereal, adjusted for average daily energy intake, study site, and age in the National Heart, Lung, and Blood Institute Growth and Health Study Breakfasta

Total Daya

Mean Daily Estimated Nutrient Intake . . .

Mean Daily Estimated Nutrient Intake . . .

Nutrient

. . . from cereal breakfasts

. . . from noncereal breakfasts

P value

. . . on cereal days

. . . on noncereal days

P value

Total fat (g) Fiber (g) Calcium (mg) Cholesterol (mg) Iron (mg) Folic acid (mcg) Vitamin C (mg) Zinc (mg)

10.13 3.27 322.55 35.75 8.23 214.65 50.54 3.50

16.25 2.09 183.46 96.71 3.01 71.46 43.53 1.76

⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001

69.07 12.58 875.12 190.19 16.90 346.76 120.53 10.78

76.20 11.13 717.59 246.28 11.47 192.27 109.12 9.21

⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001

a

Means adjusted for site, age, and average daily energy intake.

Figure 2. Mean raw body mass index (BMI) by age and days of cereal consumed in the National Heart, Lung, and Blood Institute Growth and Health Study.

of energy intake with risk of overweight was not significant [␹2(1)⫽0.15, P⫽.70]. DISCUSSION The objective of this research was to investigate the role of breakfast and cereal consumption on BMI as girls matured through adolescence. The longitudinal analysis clearly demonstrated that cereal consumption was predictive of lower BMI in these girls. Previous research has documented several negative behaviors associated with overweight or obesity in youth, including lack of physical activity, unhealthful weight-control measures, unhealthful food choices or poor diet, and skipping breakfast (13,15,20). Although these studies provided valuable insight about factors that may be associated with overweight or obesity, they are unable to predict the effects of positive health behaviors, such as cereal consumption, on weight over time. The model used in this research controlled for physical activity level and demonstrated a

Figure 3. Estimated body mass index (BMI)-for-age z scores for 14-year-old girls who ate breakfast on 2 days, by days of cereal consumption, adjusted for study site, parental education, number of parents, physical activity, and energy intake. The dots show estimates for the final model with days of cereal consumption treated as an ordinal variable. The bars indicate 95% confidence intervals for predicted BMI z scores. The line shows the linear trend of days of cereal consumption in a model where days of cereal consumption was treated as a continuous variable but all other aspects of the model were held constant (coefficient for days eating cereal⫽⫺.015, standard error⫽.005, z⫽3.41, P⬍.001).

strong relationship between cereal consumption and lower BMI. In the Growth and Health Study, breakfast and cereal consumption decreased during the important growth period of adolescence. As girls matured through adolescence, BMI increased as would be expected, but cereal eaters were leaner than girls who did not eat cereal, regardless of age (based on the significant main effect of cereal consumption in the model of BMI, but no significant age-by-cereal interaction). Cereal consumption has previously been associated with reduced risk for overweight/obesity in children ages 4 to 12 years (10) and in 10-year-olds and young adults in the Bogalusa Heart Study (12). The contribution of cereal to BMI may be partly due to calcium-rich foods (eg, milk) that are often consumed together with cereal. Calcium intake is negatively associated with children’s BMIs (21), possibly be-

September 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION

1387

cause calcium may play a role in regulating body fat (22). Results from this study are also consistent with previous research in children and adolescents in which breakfast skipping was more common among older adolescents and much less frequent among younger children (6,13,23). In a model adjusting for demographic characteristics, physical activity, and energy intake, after days of cereal consumption was taken into account, days of breakfast consumption did not make a significant, independent contribution to the prediction of two BMI-based measures (BMI-for-age z scores and risk of overweight). Because cereal is usually (but not always) consumed at breakfast, this finding may be partly an effect of the foods eaten at breakfast—that is, it may be the content of breakfast (eg, cereal vs other foods) that predicts BMI, not breakfast itself. Cho and colleagues (11) found that eating breakfast per se was not related to lower BMI; rather, only breakfasts consisting of certain kinds of foods (eg, cereal) predicted reduced BMI. Compared with adults who did not eat breakfast (skippers), those who ate breakfasts consisting primarily of cereal had significantly lower BMIs, whereas skippers did not differ from adults who ate breakfasts consisting primarily of meat and eggs, dairy products, fruit and vegetables, breads, beverages, or fats and sweets. In the present study, the absence of a breakfast main effect (beyond the cereal effect) may be due to the fact that breakfasts not involving cereal consist primarily of foods that are not predictive of reduced BMI.

It is possible that consistent cereal eating is a marker for consistent intake of nutrient-rich foods and/or a tendency to consistently follow a generally healthful lifestyle. The model of BMI z scores indicated that consistent cereal eating (ie, eating cereal during all 3 days of a food diary) was predictive of lower BMI, while sporadic cereal eating (ie, eating cereal 1 or 2 days out of 3) was no better than eating no cereal at all. It is possible that consistent cereal eating is a marker for consistent intake of nutrient-rich foods and/or a tendency to consistently follow a generally healthful lifestyle. However, eating cereal on 1, 2, or 3 days (compared with 0 days) was associated with reduced risk of overweight. That is, any cereal eating, not necessarily consistent cereal eating, was associated with differences at the very high end of the BMI scale, indicated by risk of overweight. It is possible that different degrees of cereal intake affect different ranges along the BMI scale; even a small dose of cereal affects the high end of the scale, whereas more cereal is needed to show differences in other ranges of BMI. Although these findings are suggestive, the relationship between cereal and BMI merits further study. The Growth and Health Study was not specifically designed to examine this relationship. It is possible that this epidemiologic analysis detected only the largest associations. There may be associations at other consumption levels for

1388

September 2005 Volume 105 Number 9

both BMI z scores and risk of overweight, but this analysis was unable to detect them. Beyond the relationship between cereal and BMI, cereal consumption had positive effects on nutrient intake in girls, resulting in diets significantly higher in calcium and fiber and significantly lower in fat and cholesterol. RTECs made significant contributions to nutrients in the diets of these adolescent girls. This is likely the case, not only because of the nutritional attributes of the cereal itself, but also because cereal is often eaten with other foods (like milk or juice) that positively impact the diet. Another possibility is that cereal displaces less-healthful food choices at breakfast. CONCLUSIONS Cereal consumption may be one component of a healthful lifestyle that helps adolescent girls to maintain adequate nutrient intake and a healthful BMI. Cereal consumption may form part of an overall eating pattern that promotes maintenance of healthful body weights. This research was supported by General Mills, Inc. It was also supported by a grant from the National Heart, Lung, and Blood Institute (NHLBI) (HL/DK71122) and by contracts HC55023-26 and Cooperative Agreements U01HL-48941-44. Participating Growth and Health Study Centers included Children’s Medical Center, Cincinnati, OH (Stephen R. Daniels, MD, principal investigator, John A. Morrison, PhD, co-investigator); Westat, Inc, Rockville, MD (George B. Schreiber, ScD, principal investigator, Ruth Striegel-Moore, PhD, co-investigator); and University of California, Berkeley, CA (Zak I. Sabry, PhD, principal investigator, Patricia B. Crawford, DrPH, RD, co-investigator). Maryland Medical Research Institute, Baltimore, MD (Bruce A. Barton, PhD, principal investigator) served as the data coordinating center. Program Office: NHLBI (Eva Obarzanek, PhD, RD, project officer 1992-present, Gerald H. Payne, MD, project officer 1985-1991). We acknowledge with gratitude the long-term commitment of all Growth and Health Study participants and their families who contributed to this study, and of the Growth and Health Study personnel for their dedication to the project. References 1. Pollitt E, Mathews R. Breakfast and cognition: An integrative summary. Am J Clin Nutr. 1998;67(suppl 4):S804-S813. 2. Wesnes KA, Pincock C, Richardson D, Helm G, Hails S. Breakfast reduces declines in attention and memory over the morning in school children. Appetite. 2003;41:329-331. 3. Morgan KJ, Zabik ME, Leveille GA. The role of breakfast in nutrient intake of 5- to 12-year-old children. Am J Clin Nutr. 1981;34:1418-1427. 4. Ruxton CH, Kirk TR. Breakfast: A review of associations with measures of dietary intake, physiology and biochemistry. Br J Nutr. 1997;78:199-213. 5. Nicklas TA, O’Neil CE, Berenson GS. Nutrient contribution of breakfast, secular trends, and the role of ready-to-eat cereals: A review of data from the Boga-

6.

7.

8. 9. 10.

11.

12.

13.

14.

lusa Heart Study. Am J Clin Nutr. 1998;67(suppl 4):S757-S763. Siega-Riz AM, Popkin BM, Carson T. Trends in breakfast consumption for children in the United States from 1965 to 1991. Am J Clin Nutr. 1998; 67(suppl 4):S748-S756. Affenito SG, Thompson D, Barton BA, Franko DL, Daniels SR, Obarzanek E, Schreiber GB, StriegelMoore R. Breakfast consumption in black and white adolescent girls correlates positively with calcium and fiber intake and negatively with body mass index. J Am Diet Assoc. 2005;105:938-945. Zabik ME. Impact of ready-to-eat cereal consumption on nutrient intake. Cereal Foods World. 1987;32:234239. Nicklas TA, Bao W, Webber LS, Berenson GS. Breakfast consumption affects adequacy of total daily intake in children. J Am Diet Assoc. 1993;93:886-891. Albertson AM, Anderson GH, Crockett SJ, Goebel MT. Ready-to-eat cereal consumption: Its relationship with BMI and nutrient intake of children aged 4 to 12 years. J Am Diet Assoc. 2003;103:1613-1619. Cho S, Dietrich M, Brown CJP, Clark CA, Block G. The effect of breakfast type on total daily energy intake and body mass index: Results from the third National Health and Nutrition Examination Study (NHANES III). J Am Coll Nutr. 2003;22:296-302. Nicklas TA, Myers L, Berenson GS. Total nutrient intake and ready-to-eat cereal consumption of children and young adults in the Bogalusa Heart Study. Nutr Rev. 1995;53(9 Pt 2):S39-S45. Wolfe WS, Campbell CC, Frongillo EA Jr, Haas JD, Melnik TA. Overweight schoolchildren in New York State: Prevalence and characteristics. Am J Public Health. 1994;84:807-813. Ortega RM, Requejo AM, Lopez-Sobaler AM, Quintas ME, Andres P, Redondo MR, Navia B, Lopez-Bonilla MD, Rivas T. Difference in the breakfast habits of

15.

16.

17.

18. 19.

20.

21.

22. 23.

overweight/obese and normal weight schoolchildren. Int J Vitam Nutr Res. 1998;68:125-132. Boutelle K, Neumark-Sztainer D, Story M, Resnick M. Weight control behaviors among obese, overweight, and nonoverweight adolescents. J Pediatr Psychol. 2002;27:531-540. National Heart, Lung, and Blood Institute Growth and Health Study Research Group. Obesity and cardiovascular disease risk factors in black and white girls: The NHLBI Growth and Health Study. Am J Public Health. 1992;82:1613-1620. Crawford PB, Obarzanek E, Morrison J, Sabry ZI. Comparative advantage of 3-day food records over 24-hour recall and 5-day food frequency validated by observation of 9- and 10-year-old girls. J Am Diet Assoc. 1994;94:626-630. Schakel SF, Sievert YA, Buzzard IM. Sources of data for developing and maintaining a nutrient database. J Am Diet Assoc. 1988;88:1268-1271. Kimm SYS, Glynn NW, Kriska AM, Barton BA, Kronsberg SS, Daniels SR, Crawford PB, Sabry ZI, Liu K. Decline in physical activity in black girls and white girls during adolescence. N Engl J Med. 2002; 347:709-715. St-Onge M-P, Keller KL, Heymsfield SB. Changes in childhood food consumption patterns: A cause for concern in light of increasing body weights. Am J Clin Nutr. 2003;78:1068-1073. Skinner JD, Bounds W, Carruth BR, Ziegler P. Longitudinal calcium intake is negatively related to children’s body fat indexes. J Am Diet Assoc. 2003;103: 1626-1631. Zemel MB. Role of calcium and dairy products in energy partitioning and weight management. Am J Clin Nutr. 2004;79(suppl 5):S907-S912. Lytle LA, Siefert S, Greenstein J, McGovern P. How do children’s eating patterns and food choices change over time? Results from a cohort study. Am J Health Promot. 2000;14:222-228.

September 2005 ● Journal of the AMERICAN DIETETIC ASSOCIATION

1389