Diet and Physical Activity
Meal Frequency and Childhood Obesity
Andre ´ M. Toschke,* Helmut Ku ¨chenhoff,† Berthold Koletzko,‡ and Ru ¨diger von Kries*
BERTHOLD KOLETZKO, AND RU¨DIGER VON KRIES.
Meal frequency and childhood obesity. Obes Res. 2005;13:
Objective: Previous studies have demonstrated an inverse
association between meal frequency and the prevalence of
obesity in adulthood. The aim of this study was to assess the
relationship between meal frequency and childhood obesity.
Research Methods and Procedures: Stature and weight of
4370 German children ages 5 to 6 years were determined in
six Bavarian (Germany) public health offices during the
obligatory school entry health examination in 2001/2002.
An extensive questionnaire on risk factors for obesity was
answered by their parents. Obesity was defined according to
sex- and age-specific BMI cut-off points proposed by the
International Obesity Task Force. The main exposure was
daily meal frequency.
Results: The prevalence of obesity decreased by number of
daily meals: three or fewer meals, 4.2% [95% confidence
interval (CI), 2.8 to 6.1]; four meals, 2.8% (95% CI, 2.1 to
3.7); and 5 or more meals, 1.7% (95% CI, 1.2 to 2.4). These
effects could not be explained by confounding due to a wide
range of constitutional, sociodemographic, and lifestyle fac-
tors. The adjusted odds ratios for obesity were 0.73 (95%
CI, 0.44 to 1.21) for four meals and 0.51 (95% CI, 0.29 to
0.89) for five or more meals. Additional analyses pointed to
a higher energy intake in nibblers compared with gorgers.
Discussion: A protective effect of an increased daily meal
frequency on obesity in children was observed and appeared
to be independent of other risk factors for childhood obesity.
A modulation of the response of hormones such as insulin
might be instrumental.
M., HELMUT KU¨CHENHOFF,
Key words: epidemiology, diet, prevention and control,
energy metabolism, feeding behavior
Overweight and obesity are the most common nutritional
disorders in industrialized countries, and they continue to
increase in their prevalences (1–3). Childhood obesity pre-
dicts obesity in adulthood (4–6) and later cardiovascular
disease (7–9). Effective prevention strategies against child-
hood obesity are needed because therapeutic interventions
are expensive and tend to have poor long-term results
Previous studies have demonstrated an inverse associa-
tion between meal frequency and the prevalence of over-
weight and obesity in adults (11–14). In other studies, the
failure to find a significant protective effect among adults
might possibly have been due to lack of sample power
(15–17). Differences in the ascertainment and definitions of
meals across studies, e.g., total number of meals per day
(11–14) vs. predefined meal periods (15–17), might have
further contributed to the heterogeneity of the findings.
To our knowledge, the association between meal fre-
quency and body weight in children has been addressed in
only three studies, one study including 226 children 6 to 16
years (18) and two recent studies including 1562 children
(19) and 1584 children (20). The failure to detect a signif-
icant association between meal frequency and the preva-
lence of overweight in these studies was possibly due to
lack of sample power (18–20). Thus, the predictive poten-
tial of an increased habitual meal frequency in younger
children remains unknown.
To assess the impact of increased meal frequency on
overweight and obesity among apparently healthy preschool
children, we analyzed data from a cross-sectional survey
performed as part of the 2001/2002 Bavarian school entry
health examination with an extensive questionnaire on a
wide range of sociodemographic and lifestyle factors pos-
sibly related to overweight and obesity in children.
Research Methods and Procedures
Study Population and Data Sources
Children in the year before school entry have to attend a
mandatory school entry health examination in local public
Received for review October 12, 2004.
Accepted in final form August 4, 2005.
The costs of publication of this article were defrayed, in part, by the payment of page
charges. This article must, therefore, be hereby marked “advertisement” in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
*Division of Pediatric Epidemiology at the Institute of Social Pediatrics and Adolescent
Medicine, †Department of Statistics, and ‡Dr. von Hauner Children’s Hospital, Ludwig-
Maximilians-University Munich, Munich, Germany.
Address correspondence to Andre ´ M. Toschke, Institute of Social Pediatrics and Adolescent
Medicine, Division of Pediatric Epidemiology, Ludwig-Maximilians-University Munich,
Heiglhofstrasse 63, D-81377 Munich, Germany.
Copyright © 2005 NAASO
1932OBESITY RESEARCH Vol. 13 No. 11 November 2005
health offices. The purpose of this compulsory examination
is to assess deficits that might influence school performance
(e.g., impaired visual faculty) but can easily be corrected
(e.g., prescription of glasses). Most of the children are at age
5 or 6 when examined. Parents of 8741 children were
invited to participate in a voluntary self-completion ques-
tionnaire study as part of their child’s obligatory school
entry examination in six Bavarian (Germany) communities
from September 2001 to August 2002. Questionnaires were
mailed together with the invitations for the school entry
health examination. Approximately 80% (n ? 7026) com-
pleted questionnaires were returned. Data on a number of
sociodemographic and potential risk factors for childhood
obesity were linked with children’s stature and weight mea-
sures. The study was approved by the Bavarian State Office
for Data Protection.
The analysis was confined to children with German na-
tionality (477 exclusions), at least 5 years but ?7 years of
age (88 exclusions). Further inclusion criteria were full
information on anthropometric measures (293 exclusions),
meal frequency (168 exclusions), and potential confounding
factors (1630 exclusions). After exclusions, data for 2070
girls and 2300 boys (total n ? 4370) were available for
Stature and weight were measured in light clothing and
without shoes by trained nurses of respective public health
offices. Stadiometers and balances are periodically cali-
brated by respective gauging offices. Overweight and obe-
sity were defined according to sex- and age-specific BMI
cut-off points proposed by the International Obesity Task
Force (21), which are equivalent to the widely used cut-off
points of 25 and 30 kg/m2for adult overweight and obesity.
The parental questionnaire was self-administered. The
question on child’s daily frequency of meals was: How
many meals per day does your child consume? Possible
answers were 1/2/3/4/5/?5 and do not know. To define a
meal, examples were given (i.e., breakfast, lunch, tea, din-
ner). These examples represent meals, which are conven-
tionally served on a plate.
The following variables were a priori considered as po-
tential confounding factors for the association between main
meal frequency and childhood obesity due to their reported
or possible associations with childhood overweight/obesity:
parental education, highest level attained by either parent,
ordinal in five levels (self-reported by parents) (22); paren-
tal obesity, metric self-reporting, height in centimeters and
weight in 0.1-kg steps (23,24); watching television or play-
ing video games, daily hours at school entry (self-reported
by parents) (25,26); physical activity at school entry, re-
ported by parents according to the Child Behavior Checklist
in four categories (27); breastfeeding, in categories of none,
up to 1 month, and ?1 month (28); eating snacks while
watching television, ordinal, frequency five weekday cate-
gories (self-reported by parents) (29); having main meals
alone, ordinal, frequency five weekday categories (self-
reported by parents); child’s consumption of instant food,
frequency five weekday categories (self-reported by par-
ents); and smoking in pregnancy (30–32). Additional anal-
yses considered information from a food frequency ques-
tionnaire based on a standard food frequency questionnaire
for children, which was used in a national study (33).
Nutrient intake was calculated based on a standard German
nutrient table (34).
The prevalence of overweight/obesity and 95% exact
confidence limits associated with meal frequency were cal-
culated based on the binomial distribution (35). Dose-re-
sponse effects were estimated with the Cochran-Armitage
test for trend (p ? 0.05). Crude and adjusted odds ratios
(ORs)1and their respective 95% confidence limits for ma-
ternal smoking and overweight/obesity were calculated us-
ing logistic regression analysis. All covariates associated
both with meal frequency and overweight/obesity (p ? 0.2)
(36) were considered as potential confounders. Multicol-
linearity of respective covariates was identified by a vari-
ance inflation factor ? 2.5 (37). A number of possible
interactions (parental education, child’s sex, parental obe-
sity, breastfeeding) with meal frequency and their influence
on offspring’s obesity were considered. All potential con-
founding and independent risk factors (p ? 0.05) were
included in multiple logistic regression analysis. The poten-
tial confounders had been dichotomized for the sake of
better comprehensibility. To assess residual confounding as
a result of dichotomization, covariates were additionally
modeled in their ordinal or continuous forms or were coded
using binary dummy variables (breastfeeding) as appropri-
ate (38). Forward selection was used to generate the final
logistic regression model and was based on the deviance as
goodness of fit measure. To avoid confusion, we did not
present so-called pseudo-R2values, which in logistic mod-
els do not describe the variance explained by the model,
unlike the R2values in linear models.
All calculations were carried out with the software pack-
age SAS version 8.2 (SAS Institute Inc., Cary, NC).
The number of children consuming four meals per day
was 1896 children (43.4%), whereas 1706 children (39.0%)
consumed five daily meals. The proportion of children
consuming more than five daily meals was only 2.9%,
1Nonstandard abbreviations: OR, odds ratio; CI, confidence interval.
Meal Frequency and Childhood Obesity, Toschke et al.
OBESITY RESEARCH Vol. 13 No. 11 November 20051933
whereas 14.7% of the children had only three daily meals at
The prevalence of overweight and obesity decreased by
number of meals per day (Table 1). A dose-response effect
with respect to the numbers of meals consumed could be
To identify potential confounders, we assessed the asso-
ciations between meal frequency and other explanatory
variables and between explanatory variables and over-
weight/obesity (Table 2). Frequent daily meals were asso-
ciated with a high educational level, decreased prevalence
of having main meals alone, daily television watching no
Number of daily meals and prevalence (95% CI) of overweight and obesity among German preschool
Meal frequency per day % Overweight (95% CI)*% Obesity (95% CI)*
Three or fewer (n ? 641)
Four (n ? 1896)
Five or more (n ? 1833)
Overall (n ? 4370)
15.0 (12.3 to 18.0)
10.9 (9.5 to 12.4)
8.1 (6.9 to 9.4)
10.3 (9.4 to 11.3)
4.2 (2.8 to 6.1)
2.8 (2.1 to 3.7)
1.7 (1.2 to 2.4)
2.6 (2.1 to 3.1)
* Cochran-Armitage Trend Test, p ? 0.001.
Description of study sample (n ? 4370); prevalences of overweight and obesity with respect to
Covariate (no. of exposed
subjects to covariate)
Overweight (%)Obesity (%)
Parental education (?10 school years)*
(n ? 3207)
Parental obesity (either parent BMI
? 30 kg/m2)* (n ? 594)
Male sex (n ? 2300)
Having at least 1 main meal/wk alone
(n ? 185)
Daily watching television ? 1 hour*
(n ? 1571)
Child’s consumption of instant food at
least once a week (n ? 982)
Breastfed ? 1 month* (n ? 2637)
Little physical activity at school entry*
(n ? 953)
Having siblings (n ? 3568)
Birth order: first child (n ? 1850)
Smoking in pregnancy* (n ? 905)
Regular snacking while watching
television* (n ? 1686)
* Significantly (p ? 0.05) associated with outcome (overweight and obesity).
Meal Frequency and Childhood Obesity, Toschke et al.
1934OBESITY RESEARCH Vol. 13 No. 11 November 2005
more than 1 hour, no regular snacking in front of the
television, having siblings, non-smoking in pregnancy, and
breastfeeding more than 1 month (data not shown).
Obesity at school entry was most strongly associated with
parental obesity, followed by little physical activity, watch-
ing television, smoking in pregnancy, and snacking in front
of the television. Breastfeeding was inversely related to
obesity, followed by high parental education (Table 2).
None of the interaction terms (parental education, child’s
sex, parental obesity, or breastfeeding with meal frequency)
was significantly associated with overweight/obesity.
A higher meal frequency reduced the crude OR for
child’s overweight and obesity with a clear dose-response
effect (Table 3). Adjustment for parental education, parental
obesity, watching television, breastfeeding, little physical
activity, smoking in pregnancy, and snacking in front of the
television could not explain the effect of meal frequency on
overweight/obesity (Table 3).
The adjusted ORs for four daily meals/five or more daily
meals with ordinal or continuous covariates were 0.75 [95%
confidence interval (CI), 0.57 to 0.99]/0.58 (95% CI, 0.44 to
0.77) for overweight and 0.76 (95% CI, 0.46 to 1.24)/0.52
(95% CI, 0.30 to 0.90) for obesity, which were similar to the
figures based on dichotomous covariates presented in Table
3. The highest variance inflation factor (1.3 for snacking in
front of the television) was clearly lower than 2.5, suggest-
ing absence of multicollinearity.
Coding meal frequency as a metric variable in the same
final regression model resulted in an OR of 0.80 (95% CI,
0.70 to 0.91) for overweight and 0.68 (95% CI, 0.53 to 0.86)
for obesity per additional daily meal. Stratification by sex
did not change the results (data not shown).
Additional analyses regarded the food intake related to
the frequency of meals. Increasing number of daily meals
was associated with a higher daily caloric (64.8 kcal/addi-
tional meal), fat (3.4 g/per additional meal), protein (0.6
g/per additional meal), and carbohydrate (9.7 g/per addi-
tional meal) intake. Although no differences among intake
of bread rolls and lemonade were observed, nibblers con-
sumed fewer potatoes (boiled, fried, chips), pastries, candy
bars, chocolate, and cola, but more pasta, fresh salads, fresh
fruits, and fresh vegetables (Cochran-Armitage trend test,
p ? 0.05).
The primary objective of the study was to assess the
potential role of an established strategy for obese children to
prevent adulthood obesity. An increased meal frequency
was inversely related to the prevalence of childhood over-
with both meal frequency and childhood overweight/obesity)
Results of multiple regression analysis including main exposure and confirmed confounders (associated
OR for meal frequency
Crude AdjustedCrude Adjusted
Four daily meals*
Five or more daily meals*
Parental education (?10
Parental obesity (either parent
BMI ? 30 kg/m2)
Daily watching television
? 1 hour
Breastfed ? 1 month
Little physical activity at
Smoking in pregnancy
Regular snacking while
0.70 (0.54 to 0.90)
0.50 (0.38 to 0.66)
0.73 (0.56 to 0.96)
0.56 (0.42 to 0.75)
0.67 (0.42 to 1.07)
0.39 (0.23 to 0.66)
0.73 (0.44 to 1.21)
0.51 (0.29 to 0.89)
0.65 (0.53 to 0.80)0.85 (0.68 to 1.06) 0.51 (0.34 to 0.75) 0.82 (0.54 to 1.25)
3.06 (2.44 to 3.83) 2.65 (2.09 to 3.35)5.72 (3.90 to 8.39)4.40 (2.93 to 6.60)
2.04 (1.68 to 2.48)
0.66 (0.54 to 0.80)
1.63 (1.31 to 2.03)
0.85 (0.69 to 1.06)
3.45 (2.33 to 5.11)
0.47 (0.32 to 0.68)
2.34 (1.50 to 3.65)
0.73 (0.48 to 1.11)
2.24 (1.82 to 2.76)
1.70 (1.37 to 2.12)
2.05 (1.66 to 2.55)
1.38 (1.10 to 1.74)
3.75 (2.57 to 5.47)
2.65 (1.81 to 3.90)
3.11 (2.10 to 4.63)
1.86 (1.22 to 2.84)
1.34 (1.10 to 1.64)0.94 (0.75 to 1.17)1.74 (1.19 to 2.53)0.93 (0.61 to 1.43)
* Three or fewer daily meals as reference group.
Meal Frequency and Childhood Obesity, Toschke et al.
OBESITY RESEARCH Vol. 13 No. 11 November 2005 1935
weight and obesity, suggesting that frequent meals might be
protective. This finding requires confirmation in future
Our results on meal frequency and overweight/obesity are
consistent with similar studies among adults (11–14,18,39).
The findings of two recent studies in children, which failed
to detect a significant association between meal frequency
and childhood obesity (19,20), are not contradictory to our
observations. Different findings may be explained by dif-
ferences in children’s age, ascertainment/definitions of
meals, definition of overweight/obesity, and lack of power
in these studies. However, the direction and size of the
effect in the Bogalusa Heart Study enrolling 1584 children
(20) was similar to our study: OR of 0.8 for childhood
overweight and eating at least three meals compared with
eating fewer than three meals. This OR fell short of being
significant due to lack of sample power. A sample size of at
least ?2800 children would have been necessary to detect
an effect of this size and in this setting (assumptions:
estimated overweight prevalence ? 11%, ? ? 5%, power ?
80%). Similarly, in the U.S. study, enrolling 1562 children
and yielding a pooled OR of 0.9 (19), lack of power may
explain the failure to detect a significant association. The
study of 226 children, which reported no point estimate of
the effect for children ?10 years of age, was even more
Regarding other risk/protective factors for obesity, our
results also fit into known patterns; as in other studies, poor
educational level (22,40), parental obesity (23,24), and
watching television (25,26) were associated with childhood
obesity, whereas physical activity (27,41) and breastfeeding
were protective (28,42–44).
A wide range of potential confounding factors could not
explain the association between the number of meals and
the prevalence of overweight and obesity. Thus, increasing
meal frequency could possibly be a target for early preven-
tion of overweight/obesity in children. Several methodolog-
ical constraints of the data have to be considered.
This is a cross-sectional study. As with all cross-sectional
studies, reverse causality may be an important issue (45).
Similar effects, however, have been found in randomized
trials among young college students and adults (18,39) and
in prospective experimental animal studies (46). A single
dietary recall might not be adequate to characterize the
usual eating patterns or meal frequency of an individual, but
it is sufficient for characterizing the eating patterns or meal
frequency of large groups of children (19).
Reporting bias is another important issue. Underreporting
of food intake (47,48), especially snacks (48–50) or carbo-
hydrates (49), among obese participants could result in a
spurious association between meal frequency and obesity.
Underreporting or socially accepted answers cannot be
ruled out for our data. Because parents of gorgers reported
higher consumptions of foods with an unhealthy image,
such as fried potatoes and chips, however, accurate answers
regarding food selection habits may be assumed.
Analysis of cases with complete information only could
result in selection bias. To control for a possible selection
bias, crude ORs of meal frequency and childhood over-
weight/obesity were compared between the entire sample
with information on anthropometric measures and meal
frequency (n ? 6255) and the restricted sample with infor-
mation on anthropometric measures and all covariates. The
crude ORs for overweight were similar for the entire sample
with 0.77 (95% CI, 0.63 to 0.93) for four daily mails and
0.51 (95% CI, 0.41 to 0.63) for five or more daily meals
compared with the restricted sample with 0.70 (0.54 to 0.90)
for four daily meals and 0.50 (95% CI, 0.38 to 0.66) for five
or more daily meals. Therefore, a selection bias due to
complete case analysis seems to be unlikely.
Additionally, the return rate of the questionnaires was
high with 80.4%. This is well above the 66% in other
nationwide surveys (51), suggesting that valid analyses are
A surprising finding was the higher nutrient and energy
intake in nibblers. This appears paradoxical at first sight but
is consistent with the results of other studies observing a
lower caloric and fat intake among adults with lower meal
frequencies (11,12,14). This finding indirectly supports the
validity of the self-reported question on meal frequency.
Possible Biological Mechanisms
An increased daily overall thermogenesis after consump-
tion of more meals could be a potential explanation. How-
ever, there is an ongoing controversy regarding the role of
this mechanism because studies on thermic food effects did
not observe different thermogenesis between nibbling vs.
gorging regimens (45). In addition, no difference in total
energy expenditure was observed between gorging and nib-
bling obese adults in a chamber calorimeter study (52).
Another mechanism potentially affecting the role of en-
ergy expenditure could be differing levels of physical ac-
tivity between nibblers and gorgers. An association between
meal frequency and habitual levels of physical activity
seems to be unlikely because parents’ reports on general
physical activity at school entry were not associated with
daily meal frequency in our data.
A further possible mechanism might be the association
between the number of meals consumed and insulin metab-
olism. In a randomized crossover study among healthy men,
lower postprandial insulin concentrations accompanied by
lower serum lipid levels were observed after the nibbling
regimen (39,53). Additionally, an animal study reported
markedly higher postprandial triacylglycerol and choles-
terol ester levels in gorging animals compared with nibbling
ones (46). Insulin is known to stimulate lipogenesis in
Meal Frequency and Childhood Obesity, Toschke et al.
1936 OBESITY RESEARCH Vol. 13 No. 11 November 2005
arterial tissues and enhances the growth and proliferation of
arterial smooth muscle cells (39,53). The increased triglyc-
eride synthesis in adipose tissue triggered by higher post-
prandial glucose and insulin levels (39,53) might contribute
to the higher prevalence of obesity among gorgers. This
concept is further supported by a study among boxers. In a
food restriction trial, one-half of the boxers consumed two
daily meals (600 kcal each), and the other one-half con-
sumed six daily meals (200 kcal each) (54). Both groups
showed a similar amount of weight loss, but boxers eating
600 kcal twice daily lost more muscle and less fat than those
eating 200 kcal six times daily (54). Although infrequent
consumption of huge meals appears to favor fat deposition,
frequent consumption of small servings adding up to the
same total caloric intake does not. Infrequent consumption
patterns might have been useful in times with limited food
resources, as pointed out by Fabry (11), whereas during
times of abundance of food, such eating habits might con-
stitute a risk factor for obesity.
The observed association of an increased daily meal
frequency and childhood overweight/obesity underlines the
importance of food intake patterns in childhood. Skipping
meals might not be an appropriate approach for reducing the
risk of obesity in children. Prospective preventive trials to
confirm the protective potential of frequent meals for child-
hood overweight and obesity might be worthwhile.
This work was partially supported by Bayerisches Sta-
atsministerium fu ¨r Umwelt, Gesundheit und Verbraucher-
schutz, Germany. Partial financial support was also ob-
tained from the Commission of the European Communities,
specific Research and Technology Development program
“Quality of Life and Management of Living Resources,”
QLK1-2001-00389 “Childhood Obesity—Programming by
Infant Nutrition (CHOPIN)”, and “Early Nutrition Program-
ming for Adult Health,” FOOD-CT-2005-007036. This re-
port does not necessarily reflect the views of the Commis-
sion and in no way anticipates its future policy in this area.
1. Freedman DS, Srinivasan SR, Valdez RA, Williamson DF,
Berenson GS. Secular increases in relative weight and adi-
posity among children over two decades: the Bogalusa Heart
Study. Pediatrics. 1997;99:420–6.
2. Ebbeling CB, Pawlak DB, Ludwig DS. Childhood obesity:
public-health crisis, common sense cure. Lancet. 2002;360:
3. Koletzko B, Girardet JP, Klish W, Tabacco O. Obesity in
children and adolescents worldwide: current views and future
directions: Working Group Report of the First World Con-
gress of Pediatric Gastroenterology, Hepatology, and Nutri-
tion. J Pediatr Gastroenterol Nutr. 2002;35(Suppl 2):S205–
4. Charney E, Goodman HC, McBride M, Lyon B, Pratt R.
Childhood antecedents of adult obesity: do chubby infants
become obese adults? N Engl J Med. 1976;295:6–9.
5. Stark O, Atkins E, Wolff OH, Douglas JW. Longitudinal
study of obesity in the National Survey of Health and Devel-
opment. BMJ. 1981;283:13–7.
6. Abraham S, Collins G, Nordsieck M. Relationship of child-
hood weight status to morbidity in adults. HSMHA Health
7. Power C, Lake JK, Cole TJ. Measurement and long-term
health risks of child and adolescent fatness. Int J Obes Relat
Metab Disord. 1997;21:507–26.
8. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS.
The relation of overweight to cardiovascular risk factors
among children and adolescents: the Bogalusa Heart Study.
9. Ravelli AC, van der Meulen JH, Osmond C, Barker DJ,
Bleker OP. Infant feeding and adult glucose tolerance, lipid
profile, blood pressure, and obesity. Arch Dis Child. 2000;82:
10. Feldman W. Periodic health examination, 1994 update: I.
Obesity in childhood: Canadian Task Force on the Periodic
Health Examination. Can Med Assoc J. 1994;150:871–9.
11. Fabry P, Hejl Z, Fodor J, Braun T, Zvolankova K. The
frequency of meals: its relation to overweight, hypercholes-
terolaemia, and decreased glucose-tolerance. Lancet. 1964;18:
12. Edelstein SL, Barrett-Connor EL, Wingard DL, Cohn BA.
Increased meal frequency associated with decreased choles-
terol concentrations; Rancho Bernardo CA, 1984–1987. Am J
Clin Nutr. 1992;55:664–9.
13. Kant AK, Schatzkin A, Graubard BI, Ballard-Barbash R.
Frequency of eating occasions and weight change in the
NHANES I Epidemiologic Follow-up Study. Int J Obes Relat
Metab Disord. 1995;19:468–74.
14. Metzner HL, Lamphiear DE, Wheeler NC, Larkin FA. The
relationship between frequency of eating and adiposity in
adult men and women in the Tecumseh Community Health
Study. Am J Clin Nutr. 1977;30:712–5.
15. Summerbell CD, Moody RC, Shanks J, Stock MJ, Geissler
C. Relationship between feeding pattern and body mass index
in 220 free-living people in four age groups. Eur J Clin Nutr.
16. Andersson I, Rossner S. Meal patterns in obese and normal
weight men: the “Gustaf” study. Eur J Clin Nutr. 1996;50:
17. Dreon DM, Frey-Hewitt B, Ellsworth N, Williams PT,
Terry RB, Wood PD. Dietary fat:carbohydrate ratio and
obesity in middle-aged men. Am J Clin Nutr. 1988;47:995–
18. Fabry P, Hejda S, Cerny K, Osancova K, Pechar J. Effect
of meal frequency in schoolchildren: changes in weight-height
proportion and skinfold thickness. Am J Clin Nutr. 1966;18:
19. Nicklas TA, Yang SJ, Baranowski T, Zakeri I, Berenson
G. Eating patterns and obesity in children: The Bogalusa
Heart Study. Am J Prev Med. 2003;25:9–16.
Meal Frequency and Childhood Obesity, Toschke et al.
OBESITY RESEARCH Vol. 13 No. 11 November 20051937
20. Nicklas TA, Morales M, Linares A, et al. Children’s meal
patterns have changed over a 21-year period: the Bogalusa
Heart Study. J Am Diet Assoc. 2004;104:753–61.
21. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a
standard definition for child overweight and obesity world-
wide: international survey. BMJ. 2000;320:1240–3.
22. Rasmussen F, Johansson M. The relation of weight, length
and ponderal index at birth to body mass index and overweight
among 18-year-old males in Sweden. Eur J Epidemiol. 1998;
23. Locard E, Mamelle N, Billette A, Miginiac M, Munoz F,
Rey S. Risk factors of obesity in a five year old population:
parental versus environmental factors. Int J Obes Relat Metab
24. Stunkard AJ, Sorensen TI, Hanis C, et al. An adoption
study of human obesity. N Engl J Med. 1986;314:193–8.
25. Dietz WH Jr, Gortmaker SL. Do we fatten our children at
the television set? Obesity and television viewing in children
and adolescents. Pediatrics. 1985;75:807–12.
26. Robinson TN. Reducing children’s television viewing to pre-
vent obesity: a randomized controlled trial. JAMA. 1999;282:
27. Moore LL, Nguyen US, Rothman KJ, Cupples LA, Ellison
RC. Preschool physical activity level and change in body
fatness in young children: The Framingham Children’s Study.
Am J Epidemiol. 1995;142:982–8.
28. von Kries R, Koletzko B, Sauerwald T, et al. Breast feeding
and obesity: cross sectional study. BMJ. 1999;319:147–50.
29. O Brien T, Walley PB, Anderson-Smith S, Drabman RS.
Naturalistic observation of the snack-selecting behavior of
obese and non-obese children. Addict Behav. 1982;7:75–7.
30. Toschke AM, Montgomery SM, Pfeiffer U, von Kries R.
Early intrauterine exposure to tobacco-inhaled products and
obesity. Am J Epidemiol. 2003;158:1068–74.
31. von Kries R, Toschke AM, Koletzko B, Slikker W Jr.
Maternal smoking during pregnancy and childhood obesity.
Am J Epidemiol. 2002;156:954–61.
32. Montgomery SM, Ekbom A. Smoking during pregnancy and
diabetes mellitus in a British longitudinal birth cohort. BMJ.
33. Kurth BM. Der bundesweite Kinder- und Jugendgesundheits-
34. Elmadfa I, Aign W, Muskat E, Fritzsche D. Die gro?e GU
Na ¨hrwert-Kalorien-Tabelle. Mu ¨nchen, Germany: Gra ¨fe und
Unzer Verlag; 2003.
35. Clopper CJ, Pearson ES. The use of confidence or fiducial
limits illustrated in the case of the binomial. Biometrika.
36. Altman DG, De Stavola BL. Practical problems in fitting a
proportional hazards model to data with updated measure-
ments of the covariates. Stat Med. 1994;13:301–41.
37. Allison PD. Logistic Regression Using the SAS System: The-
ory and Application. New York, NY: John Wiley & Sons;
38. Brenner H, Blettner M. Controlling for continuous con-
founders in epidemiologic research. Epidemiology. 1997;8:
39. Jenkins DJ, Khan A, Jenkins AL, et al. Effect of nibbling
versus gorging on cardiovascular risk factors: serum uric acid
and blood lipids. Metabolism. 1995;44:549–55.
40. Troiano RP, Flegal KM. Overweight children and adoles-
cents: description, epidemiology, and demographics. Pediat-
41. Pate RR, Dowda M, Ross JG. Associations between physical
activity and physical fitness in American children. Am J Dis
42. Armstrong J, Reilly JJ. Breastfeeding and lowering the risk
of childhood obesity. Lancet. 2002;359:2003–4.
43. Gillman MW, Rifas-Shiman SL, Camargo CA Jr, et al.
Risk of overweight among adolescents who were breastfed as
infants. JAMA. 2001;285:2461–7.
44. Toschke AM, Vignerova J, Lhotska L, Osancova K, Ko-
letzko B, Von Kries R. Overweight and obesity in 6- to
14-year-old Czech children in 1991: protective effect of
breast-feeding. J Pediatr. 2002;141:764–9.
45. Bellisle F, McDevitt R, Prentice AM. Meal frequency and
energy balance. Br J Nutr. 1997;77(Suppl 1):S57–70.
46. Juhel C, Pafumi Y, Senft M, Lafont H, Lairon D. Chron-
ically gorging v. nibbling fat and cholesterol increases post-
prandial lipaemia and atheroma deposition in the New Zealand
white rabbit. Br J Nutr. 2000;83:549–9.
47. Prentice AM, Black AE, Coward WA, et al. High levels of
energy expenditure in obese women. Br Med J (Clin Res Ed).
48. Lichtman SW, Pisarska K, Berman ER, et al. Discrepancy
between self-reported and actual caloric intake and exercise in
obese subjects. N Engl J Med. 1992;327:1893–8.
49. Heitmann BL, Lissner L. Dietary underreporting by obese
individuals: is it specific or non-specific? BMJ. 1995;311:
50. Poppitt SD, Swann D, Black AE, Prentice AM. Assessment
of selective under-reporting of food intake by both obese and
non-obese women in a metabolic facility. Int J Obes Relat
Metab Disord. 1998;22:303–11.
51. Centers for Disease Control and Prevention. National,
state, and urban area vaccination coverage levels among chil-
dren aged 19–35 months, United States, July 1996-June 1997.
52. Taylor MA, Garrow JS. Compared with nibbling, neither
gorging nor a morning fast affect short-term energy balance in
obese patients in a chamber calorimeter. Int J Obes Relat
Metab Disord. 2001;25:519–28.
53. Jenkins DJ, Wolever TM, Vuksan V, et al. Nibbling versus
gorging: metabolic advantages of increased meal frequency.
N Engl J Med. 1989;321:929–34.
54. Iwao S, Mori K, Sato Y. Effects of meal frequency on body
composition during weight control in boxers. Scand J Med Sci
Meal Frequency and Childhood Obesity, Toschke et al.
1938OBESITY RESEARCH Vol. 13 No. 11 November 2005