Definition; epidemiology; and etiology of obesity in children and adolescents

Author
William J Klish, MD
Section Editors
Kathleen J Motil, MD, PhD
John L Kirkland, MD
Deputy Editor
Alison G Hoppin, MD
Disclosures

All topics are updated as new evidence becomes available and our peer review process
is complete.
Literature review current through: ene 2012. |This topic last updated: ene 31,
2012.
INTRODUCTION — Obesity has become one of the most important public health
problems in the United States [1-3]. As the prevalence of obesity increases, so does
the prevalence of the comorbidities associated with obesity [4]. For this reason it is
imperative that health care providers identify overweight and obese children so that
counseling and treatment can be provided.

The definition, epidemiology, and etiology of obesity in children and adolescents will be
presented here. Comorbidities of obesity in children and adolescents and the clinical
evaluation of the obese child or adolescent are discussed separately. (See
"Comorbidities and complications of obesity in children and adolescents" and "Clinical
evaluation of the obese child and adolescent".)

DEFINITIONS — "Overweight" technically refers to an excess of body weight,

whereas "obesity" refers to an excess of fat. However, the methods used to directly
measure body fat are not available in daily practice. For this reason, obesity is often
assessed by means of indirect estimates of body fat (ie, anthropometrics) [5].

The body mass index (BMI) is the accepted standard measure of overweight and
obesity for children two years of age and older [6]. Body mass index provides a
guideline for weight in relation to height and is equal to the body weight divided by the
height squared (table 1). Other measures of childhood obesity, including weight-for-
height (which is particularly useful for the child younger than two years) and measures
of regional fat distribution (eg, waist circumference and waist-to-hip ratio) are
discussed separately. (See "Measurement of growth in children", section on 'Obesity'.)

Adults with a BMI between 25 and 30 are considered overweight; those with a BMI
≥30 are considered to be obese. Unlike adults, children grow in height as well as
weight. Thus, the norms for BMI in children vary with age and sex. In 2000, the
National Center for Health Care Statistics and the Centers for Disease Control (CDC)
published BMI reference standards for children between the ages of 2 and 20 years
(graph 1A-B). BMI percentiles also can be determined using a calculator for boys
(calculator 1) and for girls (calculator 2). As children approach adulthood, the 85th and
95th percentile BMI for age and sex are approximately 25 and 30, the thresholds for
overweight and obesity in adults, respectively [7].

A growing consensus supports the following definitions for children between 2 and 20
years of age (table 2):

 Underweight — BMI <5th percentile for age and sex.

 Normal weight — BMI between the 5th and 85th percentile for age and sex.
 Overweight — BMI between the 85th and 95th percentile for age and sex.
 Obese — BMI ≥95th percentile for age and sex.
 Severe obesity — BMI ≥120 percent of the 95th percentile values, OR a
BMI≥35. This corresponds to approximately the 99th percentile, or BMI z-
score ≥2.33 [8].

The definition of severe obesity in children and adolescents is not fully standardized.
The above definition has been proposed because it is clinically practical, and the CDC
growth standards are not sufficiently precise to use percentile curves at the extremes.
At age 18 a BMI ≥120 percent of the 95th percentile corresponds approximately to the
BMI threshold of 35 kg/m2, which defines Class II obesity in adults. (See "Screening
for and clinical evaluation of obesity in adults".) Children with severe obesity represent
approximately four percent of children and adolescents in the United States, with the
highest prevalence in Black and Mexican-American youth [9,10]. This group has
significantly more cardiovascular risk factors and a greater risk for having obesity in
adulthood [9]. Therefore, this threshold appears to define a group with medically
significant obesity in children and adolescents. Adolescents with this severe degree of
obesity should be treated with tertiary care intervention with a multidisciplinary
pediatric weight management team, which may include consideration for weight loss
surgery [11]. (See "Surgical management of severe obesity in adolescents".)

The term "morbid obesity" is sometimes used to identify individuals with obesity-
related comorbidities. However, this term is often inappropriately used as a synonym
for severe obesity, and it also may have pejorative connotations to patients, so its use
is discouraged. (See "Comorbidities and complications of obesity in children and
adolescents".)

In the discussion that follows, the term "obesity" refers to children with BMI >95
percentile for age and sex and "overweight" refers to children with BMI between the
85th and 95th percentile for age and sex, unless otherwise noted.
EPIDEMIOLOGY

Prevalence

Currently, almost one third of children and adolescents in the United States are either
overweight or obese [12]. The population is distributed into higher weight categories
with advancing age, as shown below:

 Overweight or obese (BMI≥85 percentile)

26.7 percent of preschool children (2 to 5 years)

32.6 percent of school-aged children (6 to 11 years)
33.6 percent of adolescents (12 to 19 years)

 Obese (BMI≥95 percentile)

12.1 percent of preschool children

18.0 percent of school-aged children
18.4 percent of adolescents

 Severe obesity (defined as a BMI≥97 percentile for these data)

9.7 percent of preschool children

13.0 percent of school-aged children
13.0 percent of adolescents

Childhood obesity is more common among American Indian, non-Hispanic blacks, and
Mexican Americans than in non-Hispanic whites [4,12-15]. Having an obese parent
increases the risk of obesity by two- to threefold. Obesity is also more prevalent
among low-income populations. As an example, 14.6 percent of low-income preschool
aged children were obese in 2008, as compared with 12.4 percent in this age group in
the general population [14]. In the same study, the prevalence of obesity among the
low-income preschool-aged population increased from 1998 to 2003, but plateaued
between 2003 and 2008.

The prevalence of childhood overweight and obesity is also increasing in most other
developed countries worldwide (figure 1). It is difficult to directly compare prevalence
rates between countries because of differences in definitions and dates of
measurements. Use of the International Obesity Task Force (IOTF) standards typically
results in lower prevalence estimates than other standards [16,17]. However, studies
using comparable statistics show that rates are particularly high (greater than 30
percent) in most countries in North and South America, as well as in Great Britain,
Greece, Italy, Malta, Portugal, and Spain [18]. There are somewhat lower rates in the
Nordic countries, and the central portion of Western Europe. In Russia and most of the
countries of Eastern Europe the prevalence of overweight is lower (less than 10
percent), but increasing. In China, the prevalence of overweight among children is
approximately 1/3 of that in the US, but a greater proportion of pre-school-aged
children are affected [17].

Thus, across a wide range of developed and developing countries, and using a variety
of measures, studies show increasing prevalence of obesity in children. Small studies
show reversal of the trend in a few populations since 2000, including children in
Scotland [19], and El Paso, Texas [20]. The reasons for the apparent improvement are
not addressed in these studies.

The increased prevalence of childhood obesity has resulted in an increased prevalence

of the comorbidities associated with obesity [4]. As an example, the prevalence of
conditions such as sleep apnea and gall bladder disease in US children and adolescents
tripled between 1979 and 1981, and 1997 and 1999 [21]. Comorbidities of childhood
obesity are discussed separately. (See "Comorbidities and complications of obesity in
children and adolescents".)

Trends — The prevalence of obesity among school-aged children (6 to 11 years) and

adolescents (12 to 19 years) in the United States dramatically increased between
1976-1980 and 2009-2010 (from 6.5 to 18.0 percent in children, and from 5.0 to 18.4
percent in adolescents) (graph 2) [1,12]. The prevalence of obesity also doubled for
preschool-aged children (2 to 5 years) from 5 percent in 1976-1980 to 12.1 percent in
2009-2010. Among infants and toddlers, the prevalence of high weight for recumbent
length was 9.7 percent in 2009-2010.

Of note, the increase in obesity prevalence reached a plateau around 2000; the
percentage of children and adolescents in each weight category remained
approximately stable between 2000 and 2010 [12]. On subgroup analysis, a slight
increasing trend in obesity prevalence and BMI was seen overall among adolescent
boys, but not among girls. However, these data do not reflect the variability in trends
for other subgroups, which vary among ethnic groups, age groups, and region. As an
example, childhood obesity in California declined overall between 2001 and 2008, but
continued to increase for Black and American Indian girls [22]. In New York City,
obesity rates among children aged 5 to 14 years decreased from 21.9 percent in 2006-
2007 to 20.7 percent in 2010-2011 [23]. Obesity also decreased to variable degrees in
most ethnic and socioeconomic subgroups analyzed. Similar plateaus in the prevalence
of childhood obesity are reported in population studies from Australia [24] and France
[25], and a decreasing trend among school-aged children in Switzerland [26].

Persistence into adulthood — It is difficult to predict which overweight children will

become obese adults. The likelihood of persistence of childhood obesity into adulthood
is related to age [27-29], parental obesity [30-32], and severity of obesity [33,34].

In longitudinal studies, approximately 25 percent of obese preschool children remain

obese as adults [35], compared to approximately 50 percent of obese 6-year olds, and
80 percent of obese 10- to 14-year olds who had one obese parent [30]. In a large
international cohort, 82 percent of individuals who were obese as children (age 11.4 ±
4.0 years) remained obese as adults at follow up approximately 23 years later [36].
These statistics must be interpreted with caution since the dietary habits and activity
levels of today's children may differ from those of the children in the studies, thereby
altering the risk of obesity in adulthood [7]. As a general rule, a sedentary obese child
who does not alter his or her caloric intake and lifestyle is unlikely to be of normal
weight as an adult.

The severity of obesity during adolescence is an important predictor of whether the

obesity is likely to persist into adulthood. In a large population study in the United
States about 75 percent of adolescents with severe obesity (BMI >120% of the 95 th
percentile) remained severely obese as adults (BMI >40) [37]. Meanwhile, 8 percent of
adolescents with more moderate obesity developed severe obesity as adults. In
younger age groups, the severity of obesity also predicts the risk of persistent obesity,
but to a lesser extent than in adolescents [9].

Whether gender affects the risk that obesity will persist into adulthood varies markedly
among studies in different populations. In one study from 1980, approximately 80
percent of obese adolescent girls remained obese into adulthood, whereas
approximately 30 percent of obese adolescent males did so [38]. This was presumably
related to changes in body composition that occur at puberty, when body fat decreases
in boys and increases in girls [39]. However, in later cohorts, obesity in adolescent
boys is more likely to be persistent, and the risk of persistent obesity is similar to that
of girls [40-42]. In one study from Australia, both boys and girls who were obese
during adolescence had a 50 to 60 percent chance of remaining obese as a young adult
[42]. However, among adolescents who were overweight but not obese during
adolescence, boys were more likely than girls to remain obese during young adulthood
(15 versus 12 percent).
The natural history of obesity and risk factors for persistence into adulthood is
discussed in greater detail separately. (See "Etiology and natural history of obesity",
section on 'Age at which overweight develops'.)

ETIOLOGY — The etiology and pathogenesis of obesity are discussed in greater detail
separately. (See "Etiology and natural history of obesity" and "Pathogenesis of
obesity".)

Environmental factors — Almost all obesity in children is strongly influenced by

environmental factors, caused by either a sedentary lifestyle or a caloric intake that is
greater than needs. The contributions of specific environmental influences are the
subject of considerable discussion and research. Increasing trends in glycemic index of
foods, sugar-containing beverages, portion sizes for prepared foods, fast food service,
diminishing family presence at meals, decreasing structured physical activity,
increasing use of computer-oriented play activity, and elements of the built
environment (eg, availability of sidewalks and playgrounds) have all been considered
as causal influences on the rise in obesity [43]. In particular, a number of well-
designed studies have shown associations between intake of sugar-containing
beverages or low physical activity and obesity or metabolic abnormalities [44-50].
Causal associations seem likely but are difficult to establish with certainty.

Television — Television viewing is perhaps the best established environmental

influence on the development of obesity during childhood. The amount of time spent in
watching television is directly related to the prevalence of obesity in children and
adolescents [51-55]. The effects may persist into adulthood. In two longitudinal cohort
studies, television viewing at ≥5 years was independently associated with increased
BMI at age 26 to 30 years [56,57]. Other studies suggest that the association between
television viewing and obesity is considerably weaker [58-60]. There are several
proposed mechanisms for this association [61-63]:

 Displacement of physical activity

 Depression of metabolic rate
 Adverse effects on diet quality

One study provides evidence that the effects of television on obesity are mediated
primarily by changes in energy intake. In a randomized trial, reducing television
viewing and computer use among overweight four to seven year-old children was
effective in reducing both BMI and energy intake during the two year intervention,
without apparent changes in physical activity [62]. Similar associations between
television viewing and energy intake have been shown in studies of older or non-
overweight youth [64].

Video games — The use of electronic games also has been associated with obesity
during childhood [65]. In the few studies that analyze the influences separately, the
association with obesity is somewhat weaker for electronic games than for television
[65,66], perhaps because the games do not include food advertising.

A few video games have been specifically designed to provide nutritional education and
encourage healthy habits [67,68]. Others require interactive physical activity by the
player [69]. Activity-enhancing games ("exergames") generally cause a small increase
in energy expenditure during playing time [70-73]. Two studies examined some of the
most commonly used games and found that energy expenditure of playing active
games was higher than that of sedentary games, but not as high as playing the
simulated sport itself [72,74]. The energy expenditure depends on the game: in one
study, energy expenditure during six different activity-enhancing games ranged 4.2
metabolic equivalents (mets) for Wii boxing to 7.1 mets for Sportwall, as compared
with 4.9 mets for walking at 3 miles/hour [75]. A small study reported that use of one
of these games had no long-term effect on obesity status, and that use of the game
declined sharply over time [76]. Otherwise, the efficacy of these games to increase
physical activity or treat obesity has not been systematically studied. The long-term
effect of activity-enhancing games probably depends on the intensity and participant
enjoyment of the game, as well as the activities replaced by the gaming.

Sleep — Cross-sectional studies suggest an association between shortened sleep

duration and obesity or insulin resistance, after adjustment for a number of potential
environmental confounders [43,77-80]; the effects are more marked in children at the
upper end of the weight range [81]. Similar findings have been seen in adult
populations [82]. Several longitudinal studies also showed associations after
adjustment for confounders, suggesting that the association may be causal [83-86].
By contrast, another large longitudinal study using self-report measures of sleep
patterns failed to find an association between short sleep duration during early
childhood and subsequent development of childhood obesity [87]. The mechanism for
the possible association has not been established, but may include alterations in serum
leptin and ghrelin levels, both of which have been implicated in the regulation of
appetite, or perhaps a longer opportunity to ingest food. (See "Etiology and natural
history of obesity", section on 'Sleep deprivation'.)

Medications — A number of drugs can cause weight gain, including psychoactive

drugs (particularly olanzapine and risperidone), antiepileptic drugs, and glucocorticoids
(table 3). Brief courses of glucocorticoids (eg, several days for an exacerbation of
asthma) are unlikely to have long-term effects on body weight unless they are
prescribed frequently. Weight gain and hyperlipidemia induced by olanzapine may be
particularly severe in adolescents as compared to adults [88]. Medication-induced
weight gain is discussed separately. (See "Etiology and natural history of obesity",
section on 'Drug-induced weight gain' and "Pharmacotherapy for schizophrenia: Side
effect management", section on 'Metabolic side effects'.)

Virus — Preliminary evidence suggests the possibility that obesity can be triggered or
exacerbated by exposure to a virus. Adenovirus 36 increases body fat in several animal
models [89]. Human studies, including a small study in twins, have shown an
association between adenovirus 36 antibodies and obesity status in adults [90]. A
multicenter study in children and adolescents also showed an association between
adenovirus 36 antibodies and the prevalence and severity of obesity [91]. Possible
explanations for the observations in humans include a true causal association,
vulnerability to adenovirus infection or persistence among individuals with obesity, or
the presence of unmeasured confounders. (See "Etiology and natural history of
obesity", section on 'Viral agents and obesity' and "Epidemiology and clinical
manifestations of adenovirus infection", section on 'Possible association with obesity'.)

Genetic factors — Genetic factors play a permissive role and interact with
environmental factors to produce obesity. Studies suggest that heritable factors are
responsible for 30 to 50 percent of the variation in adiposity [92], but most of the
genetic polymorphisms responsible have not yet been isolated. Thus, genetic
contributions to common obesity likely exist, but most of the molecular mechanisms
for these factors have yet to be determined. (See "Pathogenesis of obesity", section on
'Common obesity'.)

A variety of specific syndromes and single-gene defects which are linked to obesity in
childhood have been identified (table 4). These are rare causes of obesity, accounting
for less than one percent of childhood obesity in tertiary care centers [1,93,94]. In
addition to being overweight, children with genetic syndromes associated with obesity
typically have characteristic findings on physical examination. These include
dysmorphic features, short stature, developmental delay or intellectual disability
(mental retardation), retinal changes, or deafness. (See "Clinical features, diagnosis,
and treatment of Prader-Willi syndrome".)

For most of the syndromes, including Prader-Willi syndrome, the genetic cause has
been sufficiently isolated to permit specific testing, but the exact mechanism through
which they cause obesity is not understood or is attributable to multiple genes (table
5). Other disorders are attributable to a mutation in a single gene involved in
regulation of body weight, although the mutations also may have effects on
pigmentation (POMC) and the reproductive system (table 6). Several of these affect
the melanocortin pathway in the central nervous system. The most common single
gene defect currently identified in populations with severe obesity are mutations in the
melanocortin 4 receptor, but this is still rare, accounting for only about four to six
percent of severe obesity [95,96].

Endocrine disease — Endocrine causes of obesity are identified in less than 1 percent
of children and adolescents with obesity [94]. The disorders include hypothyroidism,
cortisol excess (eg, the use of corticosteroid medication, Cushing syndrome), growth
hormone deficiency, and acquired hypothalamic lesions (eg, infection, vascular
malformation, neoplasm, trauma) (table 4) [93,97,98]. Most children with these
problems have short stature and/or hypogonadism (figure 2) [94]. These disorders are
discussed in detail separately. (See "Acquired hypothyroidism in childhood and
adolescence" and "Epidemiology and clinical manifestations of Cushing's syndrome"
and "Diagnosis of growth hormone deficiency in children".)

Metabolic programming — There is increasing evidence that environmental and

nutritional influences during critical periods in development can have permanent
effects on an individual's predisposition to obesity and metabolic disease. The precise
mediators and mechanisms for these effects have not been established, but are the
subject of ongoing investigations [99].

Nutrition during gestation and early life — Maternal nutrition or endocrine profile
during gestation is probably an important determinant of metabolic programming, as
illustrated by the following studies:

 Individuals born small for gestational age (SGA) or large for gestational age
(LGA) have higher rates of insulin resistance during childhood and young
adulthood, even after controlling for obesity status [100-102]. Similarly,
many population-based studies confirm an association between birthweight
(reflecting fetal nutrition) and later diabetes, heart disease, insulin
resistance, and obesity [103,104].
 Studies of a cohort of individuals exposed to the Dutch famine in 1944 to 1945,
and controlled studies of over- and under-feeding in animals, support the
notion that there are causal associations between nutritional exposures
during gestation and later obesity and metabolic disease [105,106].
  A mother’s prepregnancy weight and weight gain during pregnancy are
important predictors of the child’s birthweight, even after accounting for
genetic and other prenatal environmental factors [102,107,108].
 In a study from Sweden, maternal diabetes mellitus during gestation was
associated with an increased BMI in adult male offspring, independent of
maternal BMI in early pregnancy [109].
 Children born to women who have had gastric bypass surgery appear to have a
lower prevalence of obesity than those born before gastric bypass,
suggesting that reversal of maternal obesity had beneficial permanent effects
on the metabolic profile of the offspring [110].

Infancy and early childhood are probably also critical periods for metabolic
programming. Studies in a variety of populations have shown consistent associations
between rates of weight gain during infancy or early childhood and subsequent obesity
or metabolic syndrome during early childhood [111-113], adolescence or adulthood
[114] (for systematic reviews, see references [115-118]), or with intermediate
outcomes such as adiposity and blood pressure in early childhood [119-121]. Similarly,
a preponderance of evidence suggests that breastfeeding has a modest protective
effect on the development of obesity. In conjunction with the evidence supporting
metabolic programming, these observations suggest that early intervention might be
an important tool in preventing obesity. (See "Infant benefits of breastfeeding", section
on 'Obesity'.)

Controlled trials of early nutritional interventions with long-term outcomes are still
lacking. Nonetheless, there is ample circumstantial evidence to support clinical efforts
to optimize nutrition during gestation, infancy, and early childhood. Appropriate goals
are to optimize glycemic control in pregnant women and target moderate rates of
weight gain in infants and young children. Nutritional goals are less clear for low-
birthweight infants, for whom catch-up growth is associated with improved
neurodevelopmental outcomes, but also with increased risks for metabolic disease
[122-124]. Increasing the protein component of feeding (eg, a maximum protein
content of 3.6 g/100 kcals) appears to normalize serum IGF-1 concentrations [122].
This strategy has been proposed to achieve improved neurodevelopmental and
metabolic outcomes for these infants, but it is not yet tested.

Other maternal endocrine factors — Other markers of the maternal endocrine

milieu are also associated with childhood obesity, although the mechanisms for the
association are not established. In a study of 6009 children and their mothers, younger
age of the mother at menarche was an independent predictor of the child's obesity
status, after adjustment for the maternal obesity status as well as socioeconomic
factors [125]. The children whose mothers had earlier menarche also had more rapid
growth during the first two years of life, whereas birthweight and growth after two
years were similar. The results of this study do not distinguish between mechanisms of
metabolic programming versus genetic mechanisms for the transgenerational obesity
transmission observed in this study. Environmental mechanisms are less likely because
of the adjustment for maternal BMI and socioeconomic factors, but these cannot be
excluded.

A large longitudinal study failed to demonstrate intergenerational acceleration

mechanisms (maternal-child transmission) from maternal weight status during
pregnancy [126]. Among 4654 parent-child pairs, the father-offspring and mother-
offspring associations for BMI were equally strong. Parental height and weight were
self-reported during the pregnancy, and the child's BMI was measured at
approximately 7.5 years of age. This study did not detect any effects of maternal
obesity transmitted to the child through the intrauterine environment. Thus, if
metabolic programming is a mechanism for intergenerational transmission of obesity,
the effect is either subtle, or the mediators are more complex than maternal BMI. It is
also possible that the study systematically under-estimated parental BMIs because
measurements were self-reported.

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education

materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces
are written in plain language, at the 5th to 6th grade reading level, and they answer the
four or five key questions a patient might have about a given condition. These articles
are best for patients who want a general overview and who prefer short, easy-to-read
materials. Beyond the Basics patient education pieces are longer, more sophisticated,
and more detailed. These articles are written at the 10th to 12th grade reading level
and are best for patients who want in-depth information and are comfortable with
some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage
you to print or e-mail these topics to your patients. (You can also locate patient
education articles on a variety of subjects by searching on “patient info” and the
keyword(s) of interest.)

 Basics topic (see "Patient information: My child is overweight (The Basics)")

SUMMARY
 The body mass index (BMI) is the accepted standard measure of overweight
and obesity for children two years of age and older. Body mass index
provides a guideline for weight in relation to height and is equal to the body
weight divided by the height squared (table 1). Reference standards vary by
age and sex (graph 1A-B). (See 'Definitions' above.)
 For children between 2 and 20 years of age, the following weight categories are
used (table 2): (See 'Definitions' above.)

 Overweight: BMI between the 85th and 95th percentile for age and sex.
 Obesity: BMI ≥95th percentile for age and sex.
 Severe obesity: BMI ≥120 percent of the 95th percentile values, OR a BMI
≥35. This corresponds to approximately the 99th percentile, or BMI z-score
≥2.33

 Currently, almost one third of children and adolescents in the United States are
either overweight or obese. The prevalence of obesity among children and
adolescents tripled from the late 1970s to 2000, and subsequently plateaued
for most age and weight categories (graph 2). The prevalence of childhood
overweight and obesity is also increasing in most other developed countries
worldwide (figure 1). (See 'Prevalence' above.)
 The likelihood of persistence of childhood obesity into adulthood is related to
age, parental obesity, and severity of obesity. Obesity is somewhat more
likely to persist in girls than in boys. (See 'Persistence into adulthood' above.)
 A variety of environmental factors probably contribute to the development of
obesity in children, including increasing trends in glycemic index of foods,
sugar-containing beverages, larger portion sizes for prepared foods, fast food
service, diminishing family presence at meals, decreasing structured physical
activity, shortened sleep duration, and changes in elements of the built
environment (eg, availability of sidewalks and playgrounds). Television
viewing is one of the best established environmental influences on the
development of obesity during childhood. For a few children, medications (eg,
psychoactive drugs) have an important causal role (See 'Environmental
factors' above and 'Medications' above.)
 Genetic factors play a permissive role and interact with environmental factors to
produce obesity. Studies suggest that heritable factors are responsible for 30
to 50 percent of the variation in adiposity, but most of the genetic
polymorphisms responsible have not yet been isolated. A few specific
 syndromes and single-gene defects which are linked to obesity in childhood
have been identified (table 4). (See 'Genetic factors' above.)
 Endocrine causes of obesity are identified in less than 1 percent of children and
adolescents with obesity. The disorders include hypothyroidism, cortisol
excess (eg, the use of corticosteroid medication, Cushing syndrome), growth
hormone deficiency, and acquired hypothalamic lesions (eg, infection,
vascular malformation, neoplasm, trauma). Most children with these
problems have short stature and/or hypogonadism. (See 'Endocrine disease'
above.)
 There is increasing evidence to support a role for “metabolic programming” in
the development of obesity. Metabolic programming refers to the concept
that environmental and nutritional influences during critical periods in
development, particularly during gestation, can have permanent effects on an
individual's predisposition to obesity and metabolic disease. (See 'Metabolic
programming' above.)

Clinical evaluation of the obese child and adolescent

Author
William J Klish, MD
Section Editors
Kathleen J Motil, MD, PhD
John L Kirkland, MD
Deputy Editor
Alison G Hoppin, MD
Disclosures

All topics are updated as new evidence becomes available and our peer review process
is complete.
Literature review current through: ene 2012. |This topic last updated: nov 11,
2011.
INTRODUCTION — Obesity has become one of the most important public health
problems in the United States (figure 1) [1-3]. As the prevalence of obesity increases,
so does the prevalence of the comorbidities associated with obesity [4]. For this
reason, it is imperative that health care providers identify overweight and obese
children so that counseling and treatment can be provided.

Recommendations, guidelines, and consensus statements for the evaluation and

treatment of obese children and adolescents have been published by the North
American Society for Pediatric Gastroenterology, Hepatology, and Nutrition
(NASPGHAN) [5], an Expert Committee convened by the Maternal and Child Health
Bureau (MCHB) of the Health Resources and Services Administration (HRSA) [6,7], the
American Heart Association [8], the United States Preventive Services Task Force
(USPSTF) [9] and an international Obesity Consensus Working Group [10].
The clinical evaluation of the overweight child and adolescent will be presented here.
Other aspects of clinical management of obesity in children are discussed in separate
topic reviews:

 (See "Management of childhood obesity in the primary care setting".)

 (See "Definition; epidemiology; and etiology of obesity in children and
adolescents".)
 (See "Comorbidities and complications of obesity in children and adolescents".)

DEFINITIONS — In the discussion that follows, the term "obesity" refers to children
with body mass index (BMI) >95 percentile for age and sex and the term "overweight"
refers to children with body mass index (BMI) between the 85th and 95th percentile
for age and sex (graph 1A-B), unless otherwise noted. BMI percentiles also can be
determined using a calculator for boys (calculator 1) and for girls (calculator 2).
Calculation of body mass index and definitions of obesity are discussed in detail
separately. (See "Definition; epidemiology; and etiology of obesity in children and
adolescents", section on 'Definitions'.)

DIAGNOSIS — The BMI is the most effective tool for the assessment of overweight
and obesity in children. It correlates with adiposity [11-13] and complications of
childhood overweight [14-17]. The BMI provides a guideline for weight in relation to
height; it is equal to the body weight (in kilograms) divided by the height (in meters)
squared (table 1). Health care providers typically underestimate weight status on
casual examination, so it is important to use measured heights and weights to
determine and track obesity status [18].

All children older than two years should have their height and weight measured and
BMI calculated at least yearly [6,7,19,20]. These measurements should be plotted on
an appropriate growth curve (graph 1A-B). (See "The pediatric physical examination:
General principles and standard measurements", section on 'Standard
measurements'.)

The BMI percentile and trend of percentile for age and sex determines whether the
child is underweight (<5th percentile), of normal weight (between 5th and 85th
percentile), overweight (BMI ≥85th percentile and <95th percentile), or obese (≥95th
percentile). BMI percentiles also can be determined using a calculator for boys
(calculator 1) and for girls (calculator 2). Management strategies vary accordingly
(figure 2). (See "Management of childhood obesity in the primary care setting".)
  If the BMI is below the 85th percentile (graph 1A-B) but has increased more
than three to four units (kg/m2) per year and begins to cross percentile lines,
particularly if the child is older than four years, the family should be warned
that the child is at risk of becoming overweight and be provided with simple
tips for maintaining a healthy weight (table 2A-B and table 3 and table 4 and
table 5 and table 6A-C) [6,20,21].
 If the BMI is ≥85th percentile (graph 1A-B) but less than the 95th percentile,
the child is overweight by definition (table 7). He or she should be screened
for comorbidities of obesity and given counseling to optimize lifestyle habits
with a goal of slowing the rate of weight gain [6].
 If the BMI is ≥95th percentile (graph 1A-B), the child is obese by definition
(table 7). He or she has a significant likelihood of obesity in adulthood [22-
25]. He or she should be carefully evaluated for comorbidities of obesity [6].
(See 'Evaluation' below and "Comorbidities and complications of obesity in
children and adolescents".)

The age of the child and growth patterns of the family must be taken into
consideration when evaluating trends in BMI percentile [4]. The influence of maternal
nutrition and intrauterine environment are reflected primarily in the growth parameters
at the time of birth, whereas genetic factors have a later influence. Thus, the weight
percentile of some children whose birthweight percentile is less than what would be
expected based upon family growth patterns may increase over time. However, less
than 5 percent of children cross two major percentiles lines upward on the growth
charts of the CDC after four years of age [26], and children who do so are at risk of
overweight [4]. (See "Normal growth patterns in infants and prepubertal children",
section on 'Determinants of normal growth'.)

EVALUATION — The evaluation of the overweight or obese child should identify

treatable causes and comorbidities [4,7]. The evaluation should include a complete
history and physical examination. Laboratory and radiologic studies also may be
obtained as indicated by the history and examination.

History — The history should include the age of onset of overweight and information
about the child's eating and exercise habits. The age of onset is helpful in
distinguishing overfeeding from genetic causes of overweight since syndromic
overweight often has onset before two years of age (table 8A-B). Information from the
dietary and activity history may identify potential areas for intervention [4].

The dietary history should include [4,27]:

  Identification of the caretakers who feed the child
 Identification of foods high in calories and low in nutritional value that can be
reduced, eliminated, or replaced (eg, juice, soda)
 Assessment of eating patterns (eg, timing, content, and location of meals and
snacks); a child or adolescent who feels unable to control consumption of
large amounts of food may have an eating disorder [28] (see "Eating
disorders: Epidemiology, pathogenesis, and overview of clinical features")

The activity history should include [4]:

 Identification of barriers to walking or riding a bike to school

 Evaluation of time spent in play
 Evaluation of school recess and physical education (frequency, duration, and
intensity)
 Assessment of after-school and weekend activities
 Assessment of screen-time (television, videotapes and DVDs, and video games)

Review of systems — The review of systems should probe for evidence of

comorbidities or underlying etiologies (table 9) [4,7].

Family history — The risk of comorbidities of obesity is strongly influenced by the

family history of such morbidities, whether or not the affected family member is
overweight [29]. Obesity in one or both parents is an important predictor for whether a
child's obesity will persist into adulthood [30-33]. Thus, the family history should
include information about obesity in first-degree relatives (parents and siblings) [4]. It
also should include information about common comorbidities of obesity, such as
cardiovascular disease, hypertension, diabetes, liver or gall bladder disease, and
respiratory insufficiency in first- and second-degree relatives (grandparents, uncles,
aunts, half-siblings, nephews and nieces).

Psychosocial history — The psychosocial history should include information related

to:

 Depression (eg, sleep disturbance, hopelessness, sadness, appetite changes)

(see "Depression in adolescents: Epidemiology, clinical manifestations, and
diagnosis")
 Information about school and social issues (eg, does the child have friends? Is
he or she a target for teasing?)
  Tobacco use, since cigarette smoking increases the long-term cardiovascular
risk [7,34-37] (see "Comorbidities and complications of obesity in children
and adolescents")

Examination — As with the history, the examination of the overweight child or

adolescent should evaluate the presence of comorbidities and underlying etiologies.

General appearance — Important aspects of assessment of general appearance

include assessment for dysmorphic features, which may suggest a genetic syndrome
(table 8A-B), assessment of affect, and assessment of fat distribution [4,7].

The distribution of the excess fat may help to distinguish the etiology of obesity. The
excess fat in obesity from overeating or overfeeding usually is distributed in the trunk
and periphery. In contrast, the "buffalo type" distribution of body fat (concentrated in
the interscapular area, face, neck, and trunk) is suggestive of endocrine causes of
obesity, such as Cushing syndrome and hypothyroidism. (See "Definition;
epidemiology; and etiology of obesity in children and adolescents" and "Epidemiology
and clinical manifestations of Cushing's syndrome" and "Clinical manifestations of
hypothyroidism".)

Abdominal obesity (also called central, visceral, android, or male-type obesity) is

associated with certain comorbidities, including the metabolic syndrome, polycystic
ovary syndrome, and insulin resistance. Measurement of the waist circumference, in
conjunction with calculation of the BMI, may help to identify patients at risk for these
comorbidities. Abdominal obesity and measurement of the waist circumference are
discussed separately. (See "Measurement of growth in children", section on 'Obesity'
and "Comorbidities and complications of obesity in children and adolescents" and
"Clinical features and diagnosis of polycystic ovary syndrome in adolescents".)

Blood pressure — A careful blood pressure should be obtained with a proper sized
cuff. The bladder of the cuff should cover at least 80 percent of the arm circumference
(the width of the bladder will be about 40 percent of the arm circumference) (figure 3)
[38]. In many children and adolescents with obesity, this will require use of "adult" or
"large adult" sized cuffs. Hypertension increases the long-term cardiovascular risk in
overweight or obese children [7]. In addition, hypertension may be a sign of Cushing
syndrome [4]. (See "Epidemiology and clinical manifestations of Cushing's syndrome".)

Hypertension is defined as a blood pressure greater than the 95th percentile for
gender, age and height on three separate occasions (table 10A-B and table 11A-B).
Age- and height-specific blood pressure percentiles also may be determined using
calculators for boys (calculator 3) or for girls (calculator 4). (See "Definition and
diagnosis of hypertension in children and adolescents".)

Stature — Assessment of stature and height velocity is useful in distinguishing

exogenous obesity from obesity that is secondary to genetic or endocrine abnormalities
[39]. Exogenous obesity drives linear height, so most obese children are tall for their
age. By contrast, most endocrine and genetic causes of obesity are associated with
short stature (figure 4). Growth velocity may be slowed in children with endocrine
causes of obesity, and children with Prader-Willi syndrome are often short for their
genetic potential and/or fail to have a pubertal growth spurt. (See "Clinical features,
diagnosis, and treatment of Prader-Willi syndrome".)

Head, eyes, throat — Examination of the head, eyes, and throat may provide clues to
the etiology of obesity and/or comorbidities [7].

 Microcephaly is a feature of Cohen syndrome.

 Blurred disc margins (picture 1) may indicate pseudotumor cerebri. (See
"Idiopathic intracranial hypertension (pseudotumor cerebri): Clinical features
and diagnosis".)
 Clumps of pigment in the peripheral retina may indicate retinitis pigmentosa,
which occurs in Bardet-Biedl syndrome.
 Enlarged tonsils may indicate obstructive sleep apnea. (See "Evaluation of
suspected obstructive sleep apnea in children".)
 Erosion of the tooth enamel may indicate self-induced vomiting in patients with
an eating disorder. (See "Eating disorders: Epidemiology, pathogenesis, and
overview of clinical features".)

Skin and hair — Examination of the skin and hair is particularly useful in evaluating
signs of endocrine etiologies or complications [7]:

 Dry, coarse, or brittle hair may be present in hypothyroidism. (See "Clinical

manifestations of hypothyroidism".)
 Striae and ecchymoses are manifestations of Cushing syndrome; however,
striae may also just be the result of rapid accumulation of subcutaneous fat.
 Acanthosis nigricans (picture 2) may signify type 2 diabetes or insulin
resistance [40-42].
 Hirsutism may be present in polycystic ovarian syndrome (PCOS) and Cushing
syndrome.
Abdomen — Abdominal tenderness may be a sign of gallbladder disease [7].
Hepatomegaly may be a clue to nonalcoholic fatty liver disease [7].

Musculoskeletal — The musculoskeletal examination may provide evidence of

underlying etiology or comorbidity of childhood overweight:

 Nonpitting edema may indicate hypothyroidism. (See "Clinical manifestations of

hypothyroidism".)
 Postaxial polydactyly (an extra digit next to the fifth digit) may be present in
Bardet-Biedl syndrome [43], and small hands and feet may be present in
Prader-Willi syndrome (table 8A-B) [4]. (See "Clinical features, diagnosis,
and treatment of Prader-Willi syndrome".)
 The musculoskeletal examination may provide evidence of slipped capital
femoral epiphysis (limited range of motion at the hip, gait abnormality) or
Blount disease (bowing of the lower legs). (See "Slipped capital femoral
epiphysis" and "Comorbidities and complications of obesity in children and
adolescents", section on 'Tibia vara (Blount disease)'.)
 Dorsal finger callousness may be a clue to self-induced vomiting in patients with
an eating disorder [4]. (See "Eating disorders: Epidemiology, pathogenesis,
and overview of clinical features".)

Genitourinary — The genitourinary examination and evaluation of pubertal stage may

provide evidence of genetic or endocrine causes of obesity [4]. Evaluation of pubertal
stage is discussed separately. (See "Normal puberty", section on 'Tanner stages' and
"Clinical features, diagnosis, and treatment of Prader-Willi syndrome", section on
'Hypogonadism'.)

 Undescended testicles, small penis, and scrotal hypoplasia may indicate Prader-
Willi syndrome.
 Microorchidism may suggest Prader-Willi or Bardet-Biedl syndrome [43].
 Delayed puberty may occur in Cushing syndrome, Prader-Willi syndrome, and
Bardet-Biedl syndrome.

Development — Most of the syndromic causes of overweight in children are

associated with cognitive or developmental delay (table 8A-B). Prader-Willi syndrome
is also associated with marked hypotonia during infancy and delayed development of
gross motor skills. (See "Clinical features, diagnosis, and treatment of Prader-Willi
syndrome", section on 'Behavior characteristics'.)
Laboratory studies — The laboratory evaluation for overweight and obesity in
children is not standardized. Some experts suggest that a basic panel of tests (ie,
fasting glucose, insulin, lipid panel [total cholesterol, triglycerides, and HDL-
cholesterol], and serum ALT) be performed in children with BMI >85th or >95th
percentile to evaluate the presence of common comorbidities [7,8,10,39]. Others
suggest that a screening laboratory evaluation should only be undertaken if it is likely
to alter the course of treatment [4]. Abnormal laboratory tests can be an added
stimulus for weight loss. The rationale for these laboratory tests is discussed in detail
separately. (See "Comorbidities and complications of obesity in children and
adolescents".)

Screening for diabetes should be performed in children over 10 years of age who are
overweight or obese AND have two or more additional risk factors, which include a
family history of type 2 diabetes in a first- or second-degree relative, high-risk
ethnicity, acanthosis nigricans, or PCOS (table 12). For these patients, screening
should consist of fasting plasma glucose (FPG) or an oral glucose tolerance test
(OGTT). Hemoglobin A1C can be used as an alternative to fasting glucose for patients
who are not fasting. Patients with intermediate or conflicting results for any of these
tests should undergo repeat testing and be monitored for future development of
diabetes. Definitive diagnosis of diabetes mellitus requires meeting diagnostic criteria
on at least two separate occasions. (See "Epidemiology, presentation, and diagnosis of
type 2 diabetes mellitus in children and adolescents", section on 'Screening'.)

Vitamin D deficiency appears to be common among children and adolescents with

obesity. In studies from Spain and New York, Vitamin D deficiency was present in
about half of children and adults with severe obesity, and was associated with higher
BMI and features of the metabolic syndrome [44-46]. However, there is currently
inadequate evidence to determine whether the risk of vitamin D deficiency is
sufficiently high in all populations of children with obesity to justify routine screening.
Vitamin D levels depend on several factors, including sunlight exposure, skin color, the
use of sun screen, and diet quality. (See "Vitamin D insufficiency and deficiency in
children and adolescents", section on 'Causes of vitamin D deficiency'.)

Interpretation of results

 A fasting glucose of 100 to 125 mg/dL (5.55 mmol/L) is considered to be

prediabetic, and a level of ≥126 mg per dL (7.0 mmol/L) is consistent with
the diagnosis of diabetes. Children with an elevated fasting glucose should
have a confirmatory oral glucose tolerance test (OGTT) or be referred to an
endocrinologist for further evaluation. (See "Epidemiology, presentation, and
 diagnosis of type 2 diabetes mellitus in children and adolescents", section on
'Screening'.)
 Measurement of hemoglobin A1C is a useful marker of chronic blood glucose
concentration. Because of improved assay standardization and validation
against other diagnostic methods, it can now be used for screening or to
make a diagnosis of diabetes mellitus, provided that an assay that is certified
by the National Glycohemoglobin Standardization Program is used. Patients
with hemoglobin A1C 5.7-6.4 percent are considered prediabetic, and those
with hemoglobin A1c>6.5 percent probably have diabetes. Both groups of
patients should be have a confirmatory OGTT or be referred to an
endocrinologist for further evaluation. (See "Epidemiology, presentation, and
diagnosis of type 2 diabetes mellitus in children and adolescents", section on
'Hemoglobin A1C'.)
 A fasting serum insulin of >17 microU/mL (122 pmol/L) is considered to be
elevated and may be associated with insulin resistance. However, it should be
noted that fasting insulin levels are an unreliable measure of insulin
sensitivity, because they are poorly correlated with whole body insulin
sensitivity as measured by the euglycemic hyperinsulinemic clamp. Moreover,
currently there is no recommended pharmacologic treatment for isolated
insulin resistance. Therefore, a consensus conference concluded that
screening for insulin resistance is not justified in the clinical setting for
children, including those with obesity, because the results would not change
clinical decision-making [47]. Nevertheless, in our practice we find that the
results of fasting insulin levels can be helpful in explaining clinical risks to
patients and families. For patients with elevated fasting insulin levels we
explain that there is a significant likelihood that type 2 diabetes mellitus will
eventually develop if weight loss is not achieved. (See "Pathogenesis of type
2 diabetes mellitus".)
 Fasting total cholesterol of >200 mg/dL (5.18 mmol/L) or a LDL cholesterol of
>130 mg/dL (3.38 mmol/L) is consistent with hyperlipidemia. Obese children
with hyperlipidemia should be monitored and perhaps treated, since
hyperlipidemia increases the risk of atherosclerosis as the obese child grows
older. Fasting serum triglycerides of >150 mg/dL (1.70 mmol/L) in
adolescents with obesity are considered to be elevated and an early sign of
the "metabolic syndrome". (See "Comorbidities and complications of obesity
in children and adolescents" and "Management of the child at-risk for
atherosclerosis".)
  Liver function tests should be obtained because nonalcoholic fatty liver disease
(NAFLD) is typically asymptomatic [5]. Obese children with an elevation of
ALT greater than two times the norm that persists for greater than three
months should be evaluated for the presence of NAFLD and other chronic
liver diseases (eg, viral hepatitis, autoimmune hepatitis, Wilson disease,
alpha-1 antitrypsin deficiency) [5]. (See "Comorbidities and complications of
obesity in children and adolescents", section on 'Nonalcoholic fatty liver
disease'.)
 If screening for vitamin D deficiency is undertaken, levels are measured as
serum 25(OH) vitamin D. The reference range varies by region, but levels
<20 ng/mL are generally considered deficient. In populations of children with
obesity, vitamin D deficiency was not generally associated with overt clinical
symptoms [44,45]. However, if deficiency is found, vitamin D
supplementation should be initiated to avoid long-term consequences. (See
"Treatment of vitamin D deficiency in adults".)

Additional testing may be necessary if there are findings consistent with

hypothyroidism, PCOS, Cushing syndrome, and sleep apnea [5,8,10,48]. Endocrine
causes of obesity are unlikely if the growth velocity is normal during childhood or early
adolescence [39]. (see 'Stature' above).

Radiographic evaluation — The radiographic evaluation of overweight or obese

children is directed by findings on the history and physical examination.

 Plain radiographs of the lower extremities should be obtained if there are

clinical findings consistent with slipped capital femoral epiphysis (hip or knee
pain, limited range of motion, abnormal gait) or Blount disease (bowed tibia).
(See "Slipped capital femoral epiphysis".)
 Abdominal ultrasonography may be indicated in children with findings
consistent with gallstones (eg, abdominal pain, abnormal transaminases) [7].
 Abdominal ultrasonography also may be used to confirm the presence of fatty
liver. However, the severity of liver involvement does not correlate with
radiographic findings. (See "Comorbidities and complications of obesity in
children and adolescents", section on 'Nonalcoholic fatty liver disease'.)

INDICATIONS FOR REFERRAL — Children who have comorbidities of obesity that

require rapid weight loss warrant referral to pediatric obesity centers for appropriate
dietary, pharmacologic, and/or surgical therapy [7]. These comorbidities include:
  Pseudotumor cerebri (should also be referred to a pediatric neurologist) (see
"Idiopathic intracranial hypertension (pseudotumor cerebri): Clinical features
and diagnosis")
 Sleep apnea (should also be referred to a pediatric pulmonologist) (see
"Evaluation of suspected obstructive sleep apnea in children")
 Obesity hypoventilation syndrome (should also be referred to a pediatric
pulmonologist) (see "Clinical manifestations and diagnosis of obesity
hypoventilation syndrome")
 Slipped capital femoral epiphysis or Blount disease (should also be referred to a
pediatric orthopedist) (see "Slipped capital femoral epiphysis")

Other children who may merit referral to a pediatric obesity center include obese
children younger than two years, and children with severe obesity (eg, BMI >40
kg/m2, or >120 percent of the 95th percentile), even if they have no comorbidities [6].
Severely overweight children may benefit from referral to a pediatric obesity specialist
for more intensive therapy than can usually be provided by the primary care provider.

Children with type 2 diabetes or polycystic ovary syndrome should be referred to a

pediatric endocrinologist, and those with nonalcoholic fatty liver disease or
cholelithiasis should usually be referred to a pediatric gastroenterologist.

Finally, certain overweight or obese children require referral to mental health

specialists. These include:

 Overweight children who are depressed should be referred for psychologic

evaluation and treatment, since weight loss therapy may be ineffective
without concurrent psychologic care [6].
 Overweight children with findings suggestive of an eating disorder (eg, inability
to control consumption of large amounts of food, self-induced vomiting or
laxative use to avoid weight gain, dorsal finger lesions) should be evaluated
by a therapist with experience in eating disorders; such children require
psychologic treatment and should not participate in weight control programs
without the concurrence of a therapist [6].

RESOURCES — Resources related to overweight in children and adolescents for health

care providers and families include:

 The Maternal and Child Health Library Knowledge Path

(mchlibrary.info/KnowledgePaths/kp_overweight.html)
 The American Academy of Pediatrics (aap.org/obesity)
  The Centers for Disease Control and Prevention
(cdc.gov/nccdphp/dnpa/obesity)
 Produce for Better Health Foundation (5aday.com)
 The Child Care Nutrition Resource System (nal.usda.gov/childcare/)
 Weight Control Information Network (win.niddk.nih.gov/index.htm)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education

materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces
are written in plain language, at the 5th to 6th grade reading level, and they answer the
four or five key questions a patient might have about a given condition. These articles
are best for patients who want a general overview and who prefer short, easy-to-read
materials. Beyond the Basics patient education pieces are longer, more sophisticated,
and more detailed. These articles are written at the 10th to 12th grade reading level
and are best for patients who want in-depth information and are comfortable with
some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage
you to print or e-mail these topics to your patients. (You can also locate patient
education articles on a variety of subjects by searching on “patient info” and the
keyword(s) of interest.)

 Basics topic (see "Patient information: My child is overweight (The Basics)")

SUMMARY AND RECOMMENDATIONS

 Obesity has become one of the most important public health problems in the
United States (figure 1) (see "Definition; epidemiology; and etiology of
obesity in children and adolescents", section on 'Epidemiology').
 The body mass index (BMI) is the accepted standard measure of obesity and
overweight in children. It is equal to the body weight (in kilograms) divided
by the height (in meters) squared (table 1). "Obesity" is defined by a body
mass index (BMI) ≥95th percentile for age and sex (graph 1A-B), and
"overweight" is defined by a BMI between the 85th and 94th percentile for
age and sex (graph 1A-B). (See 'Diagnosis' above.)
 Height and weight should be measured and BMI calculated at least yearly in
children older than two years. Those who are overweight or obese should
undergo evaluation to identify treatable causes and comorbidities. (See
"Definition; epidemiology; and etiology of obesity in children and
 adolescents", section on 'Etiology' and "Comorbidities and complications of
obesity in children and adolescents".)
 The evaluation should include a complete history and physical examination,
with particular attention to the signs and symptoms of comorbidities and
genetic and endocrinologic causes of overweight. (See 'History' above and
'Examination' above.)
 Overweight children and adolescents should undergo basic screening for
dyslipidemia and nonalcoholic fatty liver disease, with a lipid panel and
measurement of alanine aminotransferase. The clinical utility of routine
screening for insulin resistance or vitamin D deficiency in this population has
not been established. However, in our practice we use the results of these
tests as a basis for discussion about the potential consequences of obesity.
Overweight or obese children greater than 10 years of age who have two or
more risk factors including a family history of type 2 diabetes, high-risk
ethnicity, acanthosis nigricans, or PCOS should be screened for type 2
diabetes, using a fasting blood glucose and/or hemoglobin A1C (table 12).
Evaluation for other comorbidities should be performed as indicated by the
history and physical examination. (See 'Laboratory studies' above and
'Radiographic evaluation' above.)
 Overweight and obese children and adolescents with pseudotumor cerebri,
sleep apnea, obesity hypoventilation syndrome, liver disease, diabetes
mellitus, PCOS, slipped capital femoral epiphysis, and tibia vara (Blount
disease) should be referred to a pediatric obesity specialist for weight control.
We also recommend referral for obese children younger than two years. (See
'Indications for referral' above.)
 Overweight and obese children and adolescents with symptoms and signs of
depression should be referred for psychologic evaluation and treatment. (See
"Depression in adolescents: Epidemiology, clinical manifestations, and
diagnosis".)
 Overweight and obese children and adolescents with symptoms and signs of an
eating disorder should be evaluated by a therapist with experience in eating
disorders. We recommend that such patients should not participate in weight
control programs without the concurrence of their therapist. (See "Eating
disorders: Epidemiology, pathogenesis, and overview of clinical features" and
"Eating disorders: Treatment and outcome".)

Management of childhood obesity in the primary care setting

Author
Joseph A Skelton, MD, MS
Section Editors
William J Klish, MD
Martin I Lorin, MD
Kathleen J Motil, MD, PhD
Deputy Editor
Alison G Hoppin, MD
Disclosures

All topics are updated as new evidence becomes available and our peer review process
is complete.
Literature review current through: ene 2012. |This topic last updated: ene 25,
2012.
INTRODUCTION — A variety of mechanisms participate in weight regulation and the
development of obesity in children, including genetics, developmental influences
(“metabolic programming”, or epigenetics), and environmental factors. The relative
importance of each of these mechanisms is the subject of ongoing research and
probably varies considerably between individuals and populations. (See "Definition;
epidemiology; and etiology of obesity in children and adolescents", section on 'Etiology'
and "Etiology and natural history of obesity".)

Among these potential mechanisms, only environmental factors are potentially

modifiable during infancy and childhood. Therefore, prevention and treatment of
overweight and obesity in children in the primary care setting focuses on modifying
behaviors that lead to excessive energy intake and insufficient energy expenditure.

This topic review will address interventions to prevent and treat childhood obesity in
the primary care setting. The definitions, epidemiology, and comorbidities of childhood
obesity are discussed in separate topic reviews. Surgical treatment of severe obesity in
adolescents also is discussed separately. (See "Definition; epidemiology; and etiology
of obesity in children and adolescents" and "Comorbidities and complications of obesity
in children and adolescents" and "Surgical management of severe obesity in
adolescents".)

GUIDING PRINCIPLES — Few long-term randomized trials are available to determine

which techniques to prevent or treat obesity are effective. Moreover, several of the
techniques that have been addressed in clinical trials may not be practical for use in a
clinical setting. Nonetheless, a number of approaches are recommended by expert
consensus, based on clinical experience, inferences drawn from observing obesity-
associated behaviors, and short-term evidence-based trials [1-6].

Recommendations for primary care providers center on:

 Universal measurement of body mass index (BMI) and plotting of results on a

BMI chart to track changes over time
  Routine assessment of all children for obesity-related risk factors, to allow for
early intervention
 Routine brief clinical interventions by the primary care provider, which include:

 Messages of obesity-focused education to all children and families

 Family-centered communication and interventions, rather than those focused on
the child alone
 Emphasis on long-term changes in behaviors that are related to obesity risk,
rather than relying on diets and exercise prescriptions, which tend to set
short-term goals

 Implementing a staged approach to weight management, to address obesity at

different ages and levels of severity

Staged approach to weight management — The American Academy of Pediatrics

(AAP) suggests a staged approach to pediatric weight management [4]. A child’s initial
stage of treatment is determined by multiple factors, including the child’s age, BMI
percentile, and previous weight management history in other stages of treatment. The
recommendations stipulate that all patients should receive counseling on obesity
prevention, regardless of their BMI percentile, at each well-child visit. Additional
intervention to address overweight or obesity is divided into stages representing
escalating degrees of supervision, counseling, and intervention. An overview of care
location, providers, and nutrition goals are provided in the table (table 1).

 Stage 1: Prevention plus

 Stage 2: Structured weight management
 Stage 3: Comprehensive multidisciplinary evaluation
 Stage 4: Tertiary care intervention

At each treatment stage the clinician confronts a variety of barriers, which may include
a lack of clinician time, knowledge and treatment skills, or lack of support services or
of funding to get support services. Nonetheless, we feel that involvement of the
primary care provider in the first two stages of management is both practical and
important. (See 'Brief clinical intervention' below.)

The third and fourth stages of the child’s treatment typically require an intensive
degree of care that is unlikely to be achieved in a primary care setting. For example, in
adolescents with severe and refractory obesity, and particularly those with
comorbidities, treatment options include weight loss surgery. (See "Surgical
management of severe obesity in adolescents".)

RAISING THE SENSITIVE ISSUE OF WEIGHT — In most modern cultures of the

developed world there is some bias against individuals with obesity, which presumes
that obesity is a character flaw. Despite progress in understanding the complex origins
of obesity, which include genetic, epigenetic, cultural, and environmental factors, this
bias remains widespread. The bias is also present within the medical community [7],
where it often causes providers to take a blaming approach to individuals with obesity.
Such an approach is rarely effective and often jeopardizes the therapeutic alliance [8].
Individuals with obesity have often absorbed the bias themselves, leading to self-
criticism, low self-esteem, and hopelessness; these feelings are often barriers to
behavior change.

Because of this widespread cultural bias, many families with obesity are sensitive
about discussing the issue. To form a therapeutic alliance and engage the family in
addressing weight-related behaviors, the provider must carefully avoid a blaming
approach. For these reasons, it is important for providers to understand and
acknowledge the role of genetics and epigenetics in the development of obesity, even
though their assessment and intervention will emphasize modifiable environmental
factors. The genetic and epigenetic mechanisms help to explain why families of similar
education and capabilities may have very different predispositions to obesity and
different success in weight management. This perspective helps the provider take a
supportive rather than blaming approach and also reduces provider frustration in the
challenging endeavor of weight management.

In our practice, we initiate the discussion of weight management by acknowledging

that some individuals gain weight more easily than others, or are “easy gainers.” We
then move on to say that such people may have to “work extra-hard” to keep a
healthy body weight. This dual message avoids blaming a patient or family with
obesity, while still strongly encouraging them to invest in lifestyle change. We
generally use the words “unhealthy weight” or “weight problem” because these terms
are perceived by parents as more motivating and less stigmatizing than the terms
“obese” or “fat” [9].

We also choose terms which focus on health and function, rather than appearance. For
children who are already overweight or obese, we discuss the goal of “growing into a
healthy body weight,” and being “strong and fast.” At least in the initial encounter, we
try to avoid discussing an “ideal weight” for the child, both because this is a moving
target for a growing child, but also because choosing a target ideal weight is often
unrealistic and leads to discouragement, which tends to reduce patient adherence and
chances of success. Appropriate weight goals are discussed later in this topic review.
(See 'Weight' below.)

THEORETICAL BACKGROUND — Recommendations for weight management

counseling draw from framework of behavioral psychology that is outlined here. These
ideas have been translated into a format that can be used for a brief clinical
intervention, as described below. (See 'Brief clinical intervention' below.)

Behavioral strategies — Simply providing patients with education on obesity related

health risks, nutrition, and physical activity is insufficient to induce behavior change.
Instead, nutrition and physical activity should be thought of as habitual behaviors. The
best-established counseling techniques used for pediatric obesity treatment use a
behavioral change model, which includes the following elements [10-15]:

 Self-monitoring of target behaviors (logs of food, activity, or other behaviors,

recorded by patient or family). This process allows the patient and family to
recognize which behaviors may be contributing to their weight gain. Clinician
feedback throughout the self-monitoring process is essential to behavior
change. A patient’s food log may also identify other contributors to eating
behaviors, such as meal-time environment, boredom, and level of hunger, all
of which can be valuable in the evaluation of stimulus control.
 Stimulus control to reduce environmental cues that contribute to unhealthy
behaviors. This includes reducing access to unhealthy behaviors (eg,
removing some categories of food from the house or removing a television
from the bedroom) and also efforts to establish new, healthier daily routines
(such as making fruits and vegetables more accessible).
 Goal-setting for healthy behaviors rather than weight goals. Goal-setting is
widely used for prompting behavior change. However, the process can be
detrimental if goals are not realistic and maintainable. Appropriate goals are
identified by the acronym “SMART,” where goals should be should Specific,
Measurable, Attainable, Realistic, and Timely.
 Contracting for selected nutrition or activity goals. Contracting is the explicit
agreement to give a reward for the achievement of a specific goal. This helps
children focus on specific behaviors and provides structure and incentives to
their goal-setting process.
 Positive reinforcement of target behaviors. Positive reinforcement can be in
the form of praise for healthy behaviors or in the form of rewards for
achieving specific goals. The reward should be negotiated by the parent and
 the child, ideally facilitated by the provider to ensure that the rewards are
appropriate. Rewards should be small activities or privileges that the child
can participate in frequently, rather than monetary incentives or toys; food
should not be used as a reward.

Family-centered communication — Effective approaches to behavior change are

usually patient- or family-centered. In practice, this means that the process of
behavior change should be collaborative rather than prescriptive. As an example,
instead of dictating goals to the family, the clinician should engage the family in a
conversation to select specific behaviors to change [1]. The child should be directly
involved in decision-making, as appropriate to his or her age. This process ensures
that the family and patient have confidence that they can change a behavior and are
invested in the process, which greatly enhances the chance of success.

Motivational interviewing — Motivational interviewing is a patient-centered

counseling technique that is increasingly used for obesity treatment [1,4,16,17].
Initially used for the treatment of addictions, this technique addresses a patient’s
ambivalence to change and focuses on his or her values as a means to resolve that
ambivalence [18]. The clinician employs reflective listening to encourage patients to
identify their own reasons for making a behavior change, as well as their own
solutions. The tone of motivational interviewing is nonjudgmental, empathetic, and
encouraging [16,18]. Clinicians should help the family focus on specific and achievable
behavioral goals; which usually means selecting a few specific behaviors related to
weight management and overall health, and not goals for weight loss itself.

A clinical study of motivational interviewing used to treat overweight children in a

primary care setting demonstrates that the technique is well accepted by parents and
is effective for many patients [16].

Formal assessment of a patient and family’s motivation and self-efficacy has been
successfully applied to a variety of health-related behaviors. Several approaches can
be used to evaluate a patient’s readiness to change (or stage of change) [19],
including global assessment through interviewing questions, or use of a numerical or
visual analog scale (eg, “on a scale of 1 to 10, how ready are you to consider making
this change [to diet or exercise]”). This assessment may help a patient and clinician to
recognize ambivalence, which is an important step in changing behaviors.

Scare tactics — The clinician has a role in educating the family about the health risks
associated with obesity. As an example, families may not accurately assess their child’s
weight status. In this case, it is appropriate to provide information to the family, such
as reviewing the growth chart to show that the child’s weight is in an unhealthy range
and describing some of the health implications of overweight and obesity. However,
the use of scare tactics (ie, conversation that emphasizes specific dire long-term risks)
is not recommended. Scare tactics may garner short-term attention but are rarely
effective in achieving long-term change, perhaps because most people do not think
probabilistically or respond consistently to risk-based thinking [20]. Instead, it is more
effective to focus the discussion on health consequences that are less dire but more
certain, such as persistence of obesity into adulthood, reduced mobility or athletic
ability, and any personalized health concerns experienced by the patient and family.

CLINICAL ASSESSMENT

Body mass index — Measurement of body mass index (BMI) percentile for age and
gender is the most practical tool for clinicians to identify and track overweight and
obesity [21-23]. A rapid increase in weight-for-height or BMI is an important predictor
of future obesity even in children who are currently within a healthy weight category.
(See "Definition; epidemiology; and etiology of obesity in children and adolescents",
section on 'Definitions'.)

BMI percentiles are grouped into the following categories:

 Overweight: BMI 85th to 95th percentile

 Obese: BMI ≥95th percentile
 Severe obesity: BMI ≥99th percentile, or >120 percent of the 95th percentile

Other methods for measuring adiposity include dual energy x-ray absorptiometry
(DEXA) and air displacement plethysmography. These techniques measure fat mass
directly but are too cumbersome and expensive for clinical use. Use of waist
circumference and skin-fold thickness measurements is limited by discrepancies in
normative data, and they are not accurate indices of body fat in many patients [24-
26]. (See "Measurement of body composition in children".)

Assessment of comorbidities — Assessment of the patient for comorbidities and for

genetic causes of obesity is detailed in separate topic reviews. (See "Comorbidities and
complications of obesity in children and adolescents" and "Clinical evaluation of the
obese child and adolescent".)

Following are several tables highlighting key aspects of evaluating the obese child:

 Physical findings to note in children with obesity (table 2).

  Medical screening by weight category. Periodic measurement of a fasting lipid
profile, hemoglobin A1c or fasting glucose level, and aminotransferase levels
are suggested for children and adolescents with obesity (BMI ≥95th
percentile), as outlined in the table (table 3).
 Review of systems (table 4).
 Additional assessment may be required in selected children with symptoms or
signs of weight-related comorbidities (table 5). (See "Comorbidities and
complications of obesity in children and adolescents".)

Nutritional assessment — A detailed assessment of caloric intake is often impractical

in the primary care setting, has low accuracy, and is not usually necessary to support a
brief counseling intervention. Therefore, we suggest a problem-oriented approach,
using a brief dietary assessment to identify a few key dietary habits which are most
likely to be associated with obesity.

Child’s eating habits — The child’s eating habits can be briefly assessed by asking
about the following issues:

 Frequency of eating meals in restaurants (fast food, take-out, or service)

 Intake of calorie containing beverages (including juice and soft drinks)
 Frequency and portion size of energy-dense foods (such as cookies and other
baked goods, chips, or ice cream)
 Servings of vegetables and fruits, and which of these are regularly offered and
accepted. One serving equals one whole fruit, or ½ cup of vegetables.
 Number of meals each day, and frequency of skipping meals
 Typical snacking patterns (timing and foods consumed)
 School lunch (purchased or brought from home)

The rationale and suggested approaches for making these inquiries are listed in the
table (table 6) [1,2,4,24]. Although this assessment is not comprehensive, it generally
yields one or more appropriate targets for intervention, and is an efficient way to
initiate counseling to improve nutritional status when the time available for clinical
interaction is limited.

Family factors — Obesity in a child’s parents is an important predictor of the child’s

risk of persistent obesity; if both parents are obese the child’s risk of being obese as
an adult is increased 6- to 15-fold as compared with a child without obese parents
[27,28]. This is probably primarily due to genetic factors, but shared social and
nutritional factors may also play a role. Therefore, we record the biological parents’
heights and weights, and calculate their body mass index (BMI).
It is important to understand the child’s nutrition patterns within the context of the
family’s nutritional patterns and finances. In our clinical evaluations, we evaluate the
following aspects of family eating habits:

 Shopping habits, including coupon use, meal planning, use of grocery list, food
label reading, and purchasing habits for dairy (full-fat versus reduced-fat
milk) and grains (white, wheat, whole grain)
 Frequency of family meals and who is present at meals
 Whether foods are served “family style” (self-serve) or served by parent or
caregiver before bringing to the table
 Caregiver duties and communication regarding food, (eg, who does shopping
and who does cooking, whether food selection is discussed among family
members, and whether meals are eaten together as a family)
 How child spends time after school, and who supervises this time
 Work schedules of parents or other caregivers
 Meal location (eg, at dining table, in bedroom, or on couch/ in living room) and
emotional climate (especially arguments about food)
 Whether television or other media is used during meals

Other considerations — In some cases, economic or cultural factors may limit a

family’s ability or willingness to make changes in diet or physical activity. These
obstacles can be addressed using a family-centered approach to counseling, in which
families decide when to begin the change process and the intensity with which they are
willing to pursue weight management. To initiate the discussion, the following factors
should be assessed in selected patients:

 Economic challenges – Ask about food insecurity (do you sometimes run out of
money for food?), about the family’s living conditions (whether there is a
working stove and/or refrigerator), and the availability of income assistance
such as food stamps.
 Cultural factors – Ask the parents and child what they think of the child’s
weight. Misperception of the child’s weight status, such as a cultural
preference for overweight in children, may affect a family’s ability to
effectively address the problem. Conversely, excessive anxiety about the
child’s weight status also can interfere with effective management. To
address this issue, it is important to explore reasons for the anxiety in the
parent or child. Reasons for excessive anxiety may include an overestimate of
 the child’s risk for future obesity or a personal history of disordered eating in
the parent.

Activity assessment — Similar to the nutritional assessment of children, we suggest

that clinicians do a brief qualitative assessment of physical activity and sedentary
activity patterns in children. Both sedentary and active behaviors (structured and
unstructured) should be assessed as outlined in the table (table 7).

In addition, assessment of the following social and environmental factors often helps to
identify barriers to activity and opportunities for increasing physical activity [24]:

 Home – Television in bedroom [29]; family physical activity routines [30];

access to and frequency of free play [31,32]; access to and frequency of
organized sports [31-33].
 School – Physical education classes [33,34]; affordability of activities [31,32];
safety concerns [31,32].
 Lifestyle activity – Current habits that require walking or use of stairs [33].

Combining a clinical assessment of baseline activity levels with the barriers to

increased activity will provide the clinician with a basis with which to begin behavior
changes. Reducing sedentary activities is a particularly important target for
intervention and should be incorporated into any clinical approaches to obesity
treatment and prevention. (See 'Sedentary activity' below.)

GOALS FOR WEIGHT MANAGEMENT

Weight — We find that discussion of specific weight loss targets with the patient and
family is not usually helpful and sometimes causes the patient to become discouraged
and to withdraw from weight control efforts. Instead of weight loss goals, we prefer to
emphasize behavior goals for specific dietary habits and activities during discussions
with the patient and family. Nonetheless, it is appropriate for the provider to keep
weight targets in mind to ensure that a patient’s weight trend is safe and realistic.

Weight loss goals are a function of a patient’s age and degree of overweight or obesity
[1,4,6].

 For children and adolescents who are overweight or mildly obese, the goal of
maintaining current body weight is appropriate, because this will lead to a
decrease in BMI as the child grows taller. If the child is in a phase of rapid
linear growth, merely slowing weight gain is more realistic and often
improves weight status.
  At higher degrees of obesity (BMI substantially above the 95th percentile, ie, at
the 99th percentile), gradual weight loss is safe and appropriate, depending
on the child’s age and degree of obesity.

 For children between two and eleven years old with obesity and comorbidities, a
weight loss of up to one pound per month is safe and beneficial but may be
difficult to achieve.
 For adolescents with obesity and comorbidities, it is safe to lose up to two
pounds per week, although a weight loss of one to two pounds per month
usually is more realistic.
Diet — Few clinical trials have evaluated structured dietary interventions for children
with obesity, and even fewer have attempted to control for complex nutritional
behaviors and the family/home environment.

As an example, a few clinical trials have shown modest efficacy of a glycemic index
diet, which focuses on reducing those specific types of carbohydrates which produce a
strong glycemic response (rise in blood sugar) [35,36]. However, overall findings from
glycemic index research have been mixed [37,38]. In addition, this type of highly
structured diet is usually difficult to implement and sustain, so the results of a clinical
study may not be generalizable to general clinical practice or to long-term outcomes.
Diets that are low in carbohydrates are usually relatively low in glycemic index and
may be simpler to teach and easier for patients to sustain [39-41]. Unfortunately,
these and other highly structured diets have poor adherence rates over long periods of
time.

Some semi-structured approaches to pediatric obesity treatment have been promising,

particularly those which focus on encouraging families to select food groups of higher
nutrient quality and lower energy-density [10,42]. As an example, clinicians can use
the “traffic light” format as a teaching tool, grouping foods based on their nutrient
quality and calorie density, and then explaining which foods should be eaten most
often (red: eat rarely; yellow: eat less often; green: eat more often).

For these reasons, we and others suggest that health care providers use semi-
structured dietary goals, seeking long-term improvement in the quality and quantity of
the fats and carbohydrates their patients consume. These goals are most likely to be
achieved by focusing on eating behaviors, rather than by prescribing a specific
structured diet. The following table outlines our approaches to common diet-related
problems encountered in children (table 8). A dietitian can be helpful in providing this
type of counseling, particularly if the emphasis and content are coordinated with and
consistent with that of the primary care provider.
Activity — Counseling to improve physical activity should focus on reducing sedentary
activities as well as increasing physical activity [6].

Sedentary activity — Substantial evidence supports the importance of reducing

sedentary activity as a means of preventing and treating obesity in children [3,43,44].
Indeed, reducing sedentary activity may be more effective than increasing structured
physical activity, perhaps because reducing sedentary activity has the secondary
benefit of reducing caloric intake [44].

In developed countries, sedentary activity is usually in the form of “screen time”:

television, video games, internet and computer, and other media. According to national
surveys in the United States, children between 8 and 18 years spend an average of 7-
1/2 hours daily viewing media, and those under age six spent nearly two hours daily
[45]. Television viewing is perhaps the best established environmental influence on the
development of obesity during childhood. The association between obesity and use of
other media is somewhat weaker. (See "Definition; epidemiology; and etiology of
obesity in children and adolescents", section on 'Environmental factors'.)

We and others recommend that television viewing and other “screen time” (other than
homework) to be limited to less than two hours a day, and that children under age two
avoid television altogether [4,46]. Because many children initially will be viewing
substantially more than this target, the first step should be in decreasing their present
amount. School-wide campaigns and messages [43], and behavioral interventions
using reinforcement and reward strategies have been effective in reducing television
use [44].

Behavioral treatment strategies, detailed above, such as self-monitoring, can be

useful. Children and families should first monitor their present amount of media use
and then set goals to decrease it. We ask families to set firm and consistent media
limits for all family members including parents. Consistent with the policy statement
from the American Academy of Pediatrics, we strongly encourage families to include all
or most of the following components in their media rules [46]:

 No television set in child’s bedroom

 No television viewing during meals
 Maximum time for television and media viewing (eg, no more than two hours
daily, or some strategy that approximates that limit)
 No media viewing for children under two years of age

Substituting healthier behaviors and entertainment is helpful in achieving these goals.

For younger children, this is through the use of “active games” such as tag, hula
hooping, and obstacle courses. Quiet non-media activities such as reading aloud or
playing board games are also acceptable substitutes, because they avoid television
advertising and establish patterns of family interaction that may ultimately lead to
active play. Taking activity breaks during television commercials both reduces
exposure to advertising and establishes specific windows for activity; the average hour
of television features approximately 20 minutes of commercials, and half of the
advertisements are for food.

Strategies to reduce media use for older children are more variable and are best
addressed through a combination of self-monitoring, establishment of family media
limits, and negotiation to identify substitute activities. Because both media and
homework are often accessed through the computer, it can be difficult for a parent to
monitor a child’s actual media use. For these and other reasons, engagement of the
child in the behavior-change process is essential, using the behavioral strategies
outlined above. (See 'Theoretical background' above.)

Physical activity — Despite some discrepancy in study outcomes of increasing

physical activity as a means to lower BMI, increasing child and family levels of physical
activity is a key focus in obesity treatment. It is generally recommended that children
and adolescents participate in 60 minutes or more of physical activity a day
[34,47,48]; the American Academy of Pediatrics recommends at least 30 minutes of
structured physical activity during the school day [47].

As with nutrition goals, strategies for increasing physical activity are individualized.
Clinicians should take into account the developmental stage of the child, family
schedule, and personal preferences for activity, while being mindful of sedentary
activity. Clinicians can support the change process by consistently advising patients
and families to be physically active, suggesting options and encouraging goal-setting.
In addition, clinicians can provide support for physical activity in the community by
forming partnerships with local fitness centers and schools.

To increase physical activity in children it is often helpful to consider a variety of

options. Structured physical activity (organized sports or performance arts) may be
team-based or individual, and competitive or non-competitive. Less structured activity
includes recreational sports with peers or family, self-directed physical training, and
lifestyle activity. Although these categories overlap, consideration of each allows for an
expanded and diverse view of a child and family’s opportunities for physical activity.

For preschool-aged children, most physical activity will be unstructured; outdoor play
is particularly helpful [49]. Providers can encourage physical activity in this age group
by “prescribing” playground time and providing a list of local resources (playgrounds or
other opportunities for active play), in addition to discouraging sedentary time
(television use). The provider can also encourage parents to consider physical activity
levels when they make choices among options for daycare and after-school programs.

For older children, we prefer to encourage structured physical activity when possible
(ie, participation in team or individual sports, or supervised exercise sessions). Patients
are more likely to participate consistently in these activities because they are
accountable to a coach or leader. However, whether a child is willing to engage in
structured activities varies, particularly for adolescents. Some adolescents will enjoy
engaging in sports or fitness centers, while others may not, due to lack of self-
confidence or self-esteem. Directly engaging adolescents in choosing activities to
replace sedentary time is helpful. We have found that a persistent trial-and-error
approach often results in discovering activities that they enjoy.

Non-competitive active games and lifestyle activities may be more appealing to some
children, particularly those with more severe obesity. These activities provide
moderate levels of physical activity, while also replacing sedentary time. Activities to
consider are: a walking program (boosted by use of a pedometer or walking partner),
trial memberships at local gyms (we find local fitness centers often provide trial
memberships or passes to interested children and families), home fitness videos, and
non-traditional sports such as yoga, tai chi, fencing, and martial arts.

If a patient chooses to focus his or her activity on walking, we encourage them to use
a pedometer to measure the number of steps that they take; we find that this
improves motivation (at least initially) and accountability, and allows the patient to
track progress. Typical goals in adults are to walk more than 10,000 steps a day to
improve health. Step counts vary markedly among children, so the best strategy is to
measure current step counts using the pedometer and then set a goal of increasing the
number of steps by approximately 10 percent. Subsequent goal setting can be
modified based on the patient’s weight and other activities. A variety of devices are
available (relative performance and costs are reviewed at www.pedometers.com).

Electronic gaming systems are now programmed to increase physical activity through
interactive control devices, such as the Nintendo Wii™. There is scant research but
great interest in these systems to increase the physical activity level in children, and
some have been used in school-based approaches. In general, the activity levels
achieved while playing these games is modest, but certainly higher than sedentary
activities. (See "Definition; epidemiology; and etiology of obesity in children and
adolescents", section on 'Video games'.)
BRIEF CLINICAL INTERVENTION — For patients in the initial stages of obesity
treatment (stages 1 and 2 above), we suggest that the provider of primary care
perform a brief clinical intervention, using the behavioral strategies, nutritional goals,
and exercise goals outlined above and summarized in the table (table 1).

Problems identified during the brief assessment are addressed in a brief problem-
solving discussion. Even if many problems of lifestyle habits are identified, we suggest
limiting the counseling to two or three problems that the family can agree are solvable.
Other issues can be addressed in future sessions as needed. Consistent with the
theoretical principles outlined above, the intervention should be focused on modifying
lifestyle habits of the entire family, rather than focused exclusively on the identified
child [10,15]. In addition, the intervention should be tailored to the family’s level of
readiness (stage of change), and the tone of the interview should be nonjudgmental,
empathetic, and encouraging [16]. A guide to using these concepts for a brief clinical
intervention is available from the Maine Youth Overweight Collaborative [50]. (See
'Theoretical background' above.)

The counseling session can be very brief (eg, three to five minutes) and use preprinted
handouts. This brief format recognizes the time constraints that are usually present in
the primary care setting, particularly because a substantial fraction of patients in most
practices are overweight or obese and will require this intervention. Additional contact
time is valuable if time permits or if an allied health care provider (eg, dietitian or
nurse) is available to provide counseling. (See 'Intensity of intervention' below.)

These brief counseling sessions are repeated at each subsequent follow-up visit. To
provide continuity and reinforce the message, the provider should review the same
concerns at a follow-up session. If progress has been made the provider should praise
the family and encourage additional work; if no progress has been made the provider
should engage in further problem-solving and/or work with the family to identify other
goals that seem more achievable.

Intensity of intervention — Most available data suggests that substantial hours of

provider contact are necessary to improve a child’s weight status. As an example, a
systematic review concluded that behavioral interventions of moderate or high
intensity (defined as 26 to 75 hours or >75 hours of provider contact, respectively) are
effective in achieving short-term (up to 12 months) weight improvements in children
[5]. Unfortunately, interventions at this level of intensity are usually impractical for use
in a primary care setting, unless ample services from dietitians or other specialized
counselors are readily available and funded.
Low-intensity interventions (less than 25 hours of provider contact) are feasible in a
primary care setting and are recommended by expert panels, although the evidence
base to support this recommendation currently is weak. Clinical trials suggest that
these low-intensity interventions for treatment of childhood obesity have weak or
inconsistent effects [5,51]. A few studies targeted obesity in very young children, and
these also reported limited success on BMI outcomes [52,53]. As an example, a
randomized trial tested a low-intensity intervention in young children (two to five years
of age), and reported reduction in television viewing but not in BMI percentiles [53].
However, it is likely that low-intensity interventions may have important effects on
obesity and health behaviors in individual patients, even if they have little or no
measureable effect on the study population as a whole. Moreover, the effects of low-
intensity interventions on obesity prevention have not been adequately studied.

Therefore, we and others recommend that providers of primary care consistently

provide counseling to all patients to support healthy eating and activity behaviors in an
effort to prevent obesity [2]. In addition, we recommend that providers engage in brief
counseling interventions for their patients with obesity or those with significant risk
factors for developing obesity (such as parental obesity) [1].

For patients who do not respond to a brief clinical intervention or for those with severe
obesity, higher-intensity approaches are needed. These interventions are implemented
in stages, and usually require referral to a dietitian or behavioral counselor, and/or to
specialized weight management programs or tertiary care centers. (See 'Staged
approach to weight management' above.)

Example and materials — Several groups have developed messaging to support this
type of brief clinical intervention as outlined above. Materials to support patient
education and practice process improvement are available at each of the following
websites:

 5210 Let’s go (Maine Youth Overweight Collaborative)

 National Initiative for Children’s Healthcare quality (NICHQ)
 Eat smart move more (North Carolina)
 American Academy of Pediatrics

These models have much in common and have not been directly compared. It is
reasonable for providers to select materials with messaging that is best suited to their
community.

The Maine Youth Overweight Collaborative provides an example of a coordinated

intervention that has been implemented in primary care practices across the state of
Maine, using common approaches and messaging. The Healthcare toolkit includes
extensive materials for patient education and improvement of practice processes, and
is available to download free of charge or can be ordered in hard copy from the
website. Some of these materials are translated into different languages. The program
specifically seeks to evaluate outcomes using both qualitative and quantitative data.
Initial analyses suggest substantial increases in clinician support for several obesity-
related interventions [54]. The office-based initiative is closely integrated with
initiatives in schools, after-school programs, and communities, and supported by
community partners.

OFFICE SYSTEMS — Establishing an effective and efficient approach to prevention

and management of obesity in the primary care setting generally requires systems-
level changes, including practice processes, training of staff and colleagues, and
negotiating insurance reimbursement [2,24].

Practice processes — The most important practice process is to establish regular,

accurate, and efficient BMI screening for all patients during routine health
management visits. Despite the introduction of BMI growth charts over 10 years ago,
many practices are not routinely calculating and plotting BMI [55,56]. The first step is
to identify the personnel responsible for measuring heights and weights, calculating
BMI, and plotting the result. The calculation and plotting can be accomplished
automatically by many electronic health records. Important steps in the practice
improvement process are to set a clear goal for BMI charting and to objectively assess
the practice’s performance against the goal through chart audits.

It is also important to integrate the results of BMI screening into the clinical and
educational process of the health maintenance visit. Discussion of overweight and
obesity with a patient and family can be challenging but is usually well accepted if
approached in a nonjudgmental way (see 'Raising the sensitive issue of weight' above).
The following office systems may facilitate a positive and efficient discussion:

 Educational materials – Having educational materials readily available in the

office improves efficiency and communication. In our practice we have
posters with health related messages on the wall of each clinic room
alongside related educational handouts. The clinician can use the poster as a
talking point in screening health behaviors. The proximity of the materials
and displayed poster allow for efficient and focused discussion and also allows
families to select materials that capture their interest.
  Community resources – To assist families in developing an action plan, the
practice can collect and distribute information about resources in the local
community, including options for physical activity, active after-school
programs, nutrition counseling services, and sources of healthy food (eg,
local sources of fresh produce).

Equipment — Whenever possible, practices should have appropriate equipment to

provide medical care to patients with obesity. This includes a wide range of blood
pressure cuffs (including a “large adult” size) to ensure accurate measurements, and
high-capacity scales (ideally up to 500 or 1000 lbs). In addition, it is helpful to have
office furniture that is appropriate for large patients and their families, including sturdy
armless chairs and lower examination tables.

PREVENTION — Preventing obesity in children should be a focus of every child

healthcare provider. Prevention efforts are best focused on key behaviors associated
with the development of obesity (table 6 and table 7), although other factors including
genetics undoubtedly also contribute to the risk for obesity.

There is modest evidence to suggest that modification of the following factors may
help to prevent the development of obesity. These issues are discussed in more detail
in the linked topic reviews.

Maternal factors:

 Maternal weight prior to conception and weight gain during pregnancy. (See
"Weight gain and loss in pregnancy" and "Definition; epidemiology; and
etiology of obesity in children and adolescents", section on 'Metabolic
programming'.)
 Breastfeeding – Breastfeeding probably has a weak protective effect on the
development of obesity. To maximize the effect, exclusive breastfeeding until
6 months of age and maintenance of breastfeeding until 12 months are
recommended. (See "Infant benefits of breastfeeding" and "Infant benefits of
breastfeeding", section on 'Obesity'.)

Psychosocial factors:

 Establishing a healthy feeding relationship early in life (avoiding overly

restrictive and overly permissive feeding patterns). (See "Introducing solid
foods and vitamin and mineral supplementation during infancy", section on
'Feeding environment'.)
  Encouraging a family to eat meals together. (See "Dietary recommendations for
toddlers, preschool, and school-age children", section on 'Eating
environment'.)

Dietary goals: (See "Dietary recommendations for toddlers, preschool, and school-
age children", section on 'Dietary guidelines'.)

 Limiting consumption of sugar-sweetened beverages, including juice

 Encouraging a diet with ample servings of vegetables and fruits
 Limiting eating at restaurants, particularly fast-food restaurants
 Limiting portion size (which for young children often is less than a “serving size”
as listed on a food label)

Activity goals: (See "Overview of physical activity and strength training in children
and adolescents", section on 'Physical activity'.)

 Encouraging moderate to vigorous physical activity for one or more hours daily.
 Limiting television and other screen time – no screen time for children under
two years of age; less than two hours daily after age two [2].

SUMMARY AND RECOMMENDATIONS — Obesity during childhood is associated with

long-term health consequences and is influenced by genetic, epigenetic factors, and
environmental factors. Among these, only environmental factors are modifiable during
childhood, so these are the focus of clinical interventions.

We suggest the following practices among providers of primary care to children. These
suggestions are based primarily on expert opinion; some are supported by clinical
studies, usually with short-term outcomes.

 Universal measurement of body mass index (BMI) and plotting of results on a

BMI chart to track changes over time. (See 'Body mass index' above.)
 Routine assessment of all children for obesity-related risk factors, to allow for
early intervention. This includes recording the obesity status (BMI) of the
biological parents and assessing key nutritional and physical activity habits
(table 6 and table 7). (See 'Nutritional assessment' above and 'Activity
assessment' above.)
 For children with obesity, weight-related comorbidities should be assessed
through a focused review of systems (table 4), physical examination (table
2), and laboratory screening (table 3). (See 'Assessment of comorbidities'
above.)
 For all children and their families, routine health care should include obesity-
focused education. Key goals to address are the common diet-related
problems encountered in children (table 8), set firm limits on television and
other media early in the child’s life, and establish habits of frequent physical
activity. (See 'Diet' above and 'Sedentary activity' above and 'Physical
activity' above.)
 For children who are overweight or obese, we suggest a series of clinical
counseling interventions in the primary care setting (Grade 2C). Each
session can be brief (3 to 15 minutes); this brief format is most practical for
the primary care setting and is supported by limited clinical evidence.
Additional contact time is valuable if time permits or if an allied health care
provider (eg, dietitian or registered nurse) is available to provide counseling.
(See 'Brief clinical intervention' above and 'Intensity of intervention' above.)
 Educational materials are available from a variety of sources to facilitate the
counseling. These materials have much in common and have not been
directly compared; it is reasonable for providers to select materials with
messaging that is best suited to their community. Options include a
Healthcare toolkit and an outline for a brief clinical intervention, which is
based on the principles of motivational interviewing. (See 'Example and
materials' above.)
 For patients who do not respond to a brief clinical intervention or for those with
severe obesity, higher-intensity approaches are needed. These interventions
are implemented in stages (table 1) and usually require referral to specialized
weight management programs or tertiary care centers. (See 'Staged
approach to weight management' above.)
 To establish a therapeutic relationship and enhance effectiveness, the
communication and interventions should be supportive rather than blaming,
and family-centered, rather than focused on the child alone. Long-term
changes in behaviors that are related to obesity risk should be emphasized,
rather than diets and exercise prescriptions, which tend to set short-term
goals. (See 'Theoretical background' above.)
 To be effective in managing populations with obesity, primary care offices
should develop an efficient office system for calculating and tracking BMI at
each visit and have a wide range of blood pressure cuffs (including a "large
adult" size) and high-capacity scales (ideally up to 500 or 1000 lbs). It is also
helpful to have office furniture that is appropriate for large patients and their
families, including sturdy armless chairs and low examination tables. (See
'Office systems' above.)