Dietary Approaches to the Treatment of Pediatric Overweight

Joanne P. Ikeda, MA, RD,* and Rita A. Mitchell, RD+

• Co-Director, Center for Weight and Health, University of California, Berkeley; and Cooperative Extension Nutrition Education Specialist & Lecturer, Department of Nutritional Sciences and Toxicology, University of California, Berkeley.

+ Cooperative Extension Nutrition Research Associate, Department of Nutritional Sciences and Toxicology, University of California, Berkeley.

Corresponding author for proof and reprints Coauthor address

Joanne P. Ikeda, MA, RD Rita A. Mitchell, RD

Department of Nutritional Sciences & Department of Nutritional Sciences

Toxicology & Toxicology

223 Morgan Hall 209 Morgan Hall

University of California University of California

Berkeley, CA 94720-3104 Berkeley, CA 94720-3104

(510) 642-2790 (510) 642-3080

(510) 642- 4160 (510) 642- 4160

jikeda@socrates.berkeley.edu (email) ritamitc@uclink4.berkeley.edu

1

INTRODUCTION

Initially, the dietary treatment of pediatric overweight may appear to be the

straightforward implementation of a low-fat, restricted calorie diet. This has been

standard practice for over 50 years.4 However, research demonstrates this approach is

fraught with potential problems and has not resulted in any demonstrable success. For

example, a recent multicenter site study using a nutrition intervention with over 1,300

children was deemed a failure when the dropout rate exceeded 90%.31 Indeed, practicing

clinicians who have been at the “front lines” treating this problem have noted that, “most

pediatric obesity interventions are marked by small changes in relative weight or

adiposity and substantial relapse” and note that, “research is needed to improve treatment

outcomes and maintenance of treatment effects.”11 Therefore, practitioners need to

recognize this complex problem will not be solved by the wholesale recommendation of

putting overweight children on calorie-restricted diets.

IMPACT OF CALORIE RESTRICTED DIETS ON GROWTH

Limiting the energy intake of a growing child is a risky undertaking. In the mid-

1980’s Dietz found that mildly restrictive diets were associated with a reduction in linear

growth velocity in obese children.9 He examined the impact of treating 19 obese children

with balanced calorie-deficit diets containing 1.5 to 2.0 g of protein per kilogram of ideal

body weight. Mean duration of the weight reduction period was 9.7 months. Prior to

weight reduction, the mean (+/- SD) Z score for height velocity was 2.32 +/- 2.47 units.

Eleven patients were 2 SDs or more above the mean height velocity for age and sex.

During weight reduction, the mean Z score for height velocity decreased significantly to

2

0.62 +/- 2.37 units. The change in Z score before and during weight reduction was

significantly correlated with the change in weight but not with the change in weight for

height. Based on this data, Dietz concluded that even mildly restrictive diets might be

associated with a reduction in linear growth velocity. He emphasized the importance of

careful monitoring of obese children during restrictive dietary therapy even when the

caloric deficit was prudent. Since then, some researchers have argued that caloric

restriction need not impair growth in height in overweight children and that the

deceleration in height velocity Dietz identified was a natural consequence of earlier

growth spurts in these obese children.13,14 However, many health professionals remain

wary of recommending calorie restriction because of the need for adequate nutrient intake

to support the maintenance of lean tissue and growth and the adverse impact on eating

attitudes and behaviors.10

NUTRIENT AND FOOD INTAKE OF CHILDREN AND ADOLESCENTS

Certainly a risk associated with caloric restriction is inadequate intake of specific

nutrients. A number of children, whose energy intakes are adequate, have inadequate

intakes of iron, zinc, and calcium and vitamins A and E.25 Iron deficiency anemia

remains a common nutrition problem in this country although there has been a slight

increase in iron intake among children, possibly due to an increased intake of iron-

fortified cereals and other grain products. Adolescent females continue to have low iron

intakes. Recent national data indicates their mean iron intake was only 61% of the new

guidelines for adequate intakes (AI)35 for teenage girls with only 28% of girls having

3

adequate iron intakes.41 It is even worse with respect to calcium intake with a mere 13%

of adolescent girls meeting the AI36 for this nutrient. Teenage girls are also at risk for

low intakes of phosphorus, magnesium, zinc, and Vitamins A, C, E and folate.

There is concern that widespread recommendations to limit the energy intake of

overweight children and teenagers could increase the percentages with inadequate

nutrient intakes and might even put them at risk of malnutrition. At this point in time,

there is little doubt that the diets of ALL children, not just overweight children, need to

be improved with respect to nutritional quality.

DIETING AND YOUNG CHILDREN

Dieting,” defined as a means of losing weight, has been a common practice

among adult women. Over the last two decades, this practice has spread to the pediatric

population. A recent national survey found that 31% of 5th grade girls have dieted.27

Despite the prevalence of this practice; there are only a few studies that have explored

what dieting means to children. None of these have examined how dieting impacts the

nutrient intake of children who choose to diet.

Abramovitz and Birch recently explored 5-year-old girls’ ideas about dieting.1

They found that although a significant proportion of girls (55%) were still naive about

dieting and weight loss, a substantial proportion (45%) understood the link between

dieting and the attainment of a thin body shape. Girl’s responses to the question, “What

do people do when they are on a diet?” included descriptions of modified eating

behaviors such as drinking diet shakes, eating more fruits and vegetables, drinking diet

4

sodas, and eating special diet foods. Girls also described restrictive eating behaviors

such as not eating, eating less food, eating less fat, and not eating any snacks. The

researchers investigated family factors that might predict girls’ ideas about dieting and

found that only a family history of overweight and mothers’ current or recent dieting

predicted daughters’ emerging ideas, concepts, and beliefs about dieting. Girls whose

mothers had dieted or were currently dieting were twice as likely to have ideas about

dieting than girls whose mothers did not report these practices. In their suggestions for

practical application of these findings, the authors of this article recommend that health

professionals discourage mothers from using health compromising weight loss strategies

such as restrictive and restrained eating since they are modeling these behaviors for their

impressionable daughters.

A semi-structured interview was used to assess third through sixth graders’

knowledge and beliefs about dieting.34 The authors were struck by the prevalence of

body dissatisfaction among these children; one-half of them wanted to weight less and

over one-third desired a thinner body shape. The concept of dieting did not necessarily

mean caloric restriction to these children. The majority of children defined dieting

behavior as a combination of exercise and altering food choices to avoid “fattening”

foods and eating more healthy foods. Nonetheless, the authors noted that, “Parents have

the responsibility to let children know that dieting behaviors they see in adults are not

appropriate for children.”

A preliminary study carried out by Stanford psychiatrists investigated the timing

of onset for eating disturbances during childhood and tested whether parental or infant

characteristics would predict the emergence of eating problems.37 They followed 216

5

newborn infants and their parents for 5 years using questionnaires, direct measures and

observation to gather data. The risk for emergence of inhibited eating, secretive eating,

overeating, and vomiting in these children increased annually through age 5. Maternal

body dissatisfaction, internalization of the thin-ideal body, dieting, bulimic symptoms,

and maternal and paternal body mass prospectively predicted the emergence of these

childhood eating disturbances.

There is increasing evidence that parents who use controlling approaches to child

feeding may impede energy self-regulation in children and put these children at higher

risk of overweight.21 Johnson and Birch found that coercive child-feeding strategies

commonly used by parents to ostensibly ensure adequate nutritional intake may be both

unnecessary and harmful. They emphasized that it is important for parents to create an

optimal environment for their children’s growth and health by providing a variety of

nutritious food at meals and snacks, but that children should be allowed to determine the

amount of food eaten. Birch’s most recent research with 4 to 6 year old girls found that

parental restriction of young girls access to palatable snack foods such as ice cream,

potato chips, fruit-chew candy, and chocolate bars, increased intake of these foods when

the girls had free access to them.16 In addition, the girls reported felt guilty and ashamed

when they ate these foods because their behavior was at odds with parental expectations.

Birch’s work on parental control of child feeding and potential interference with

energy self-regulation in children was supported by recently published results from the 6-

year Framingham Children’s Study, a prospective study of the determinants of dietary

habits and physical activity patterns throughout childhood.19 Data from this study

showed that children whose parents had high degrees of dietary control had greater

6

increases in body fatness than children whose parents had the lowest levels of dietary

restraint and disinhibition.

DIETING AND ADOLESCENTS

Many studies have identified inappropriate dieting behaviors among teenager.

The most recently published data comes from a study examining weight-related

behaviors among a nationally representative sample of 6,728 teenagers.27 Forty-five

percent of teenage girls and 20% of teenage boys reported dieting at some point in their

lives. Older girls were significantly more likely to diet than younger girls. Dieting was

reported by 31% of 5th graders and increased monotonically to 62% among 12th graders.

The largest increase was between 8th grade girls (40%) and 9th grade girls (53%). Among

boys, associations between diet and grade level were weak and inconsistent.

Reasons for dieting were: to look better (87% of girls, 62% of boys); to improve

health (18% of girls, 27% of boys); because of parent’s suggestion (15% of girls, 14% of

boys); because of a doctor’s or nurse’s suggestion (7% of girls, 14% of boys); and

because of a coach’s or sports instructor’s suggestion (4% of girls, 22% of boys).

Although overweight girls were significantly more likely to report dieting, dieting was

also prevalent among non-overweight girls. About two-thirds of the girls and one-half of

the boys reported that it was “very important” to them to not be overweight.

The prevalence of dieting was highest among white girls and lowest among

black girls. Among boys, there was no association between ethnicity and dieting.

Disordered eating was reported by 13% of all girls and 7% of all boys. The prevalence of

7

disordered eating was highest among Hispanic girls and lowest among black girls.

Among girls, large differences in the prevalence of dieting and disordered eating across

socioeconomic levels were not noted. Among boys, disordered eating was highest among

those with low socioeconomic status.

For both boys and girls, statistically significant relationships were found between

dieting, disordered eating, low self-esteem, high levels of depression, suicidal ideation,

and high levels of stress. Alcohol and drug use were directly and significantly associated

with dieting and disordered eating among girls and boys, as was tobacco use but only

among girls. Associations between frequency of physical activity and dieting and

disordered eating were inconsistent and weak.

This study did not examine the impact of dieting on nutrient intake although it did

provide data to show, irrespective of dieting status, most adolescents were not practicing

health-promoting behaviors. More than half the adolescents had not eaten a vegetable

during the day preceding data collection, and only about 15% of the students had eaten 5

or more servings of fruits and vegetables on that day. Only a third of adolescent girls and

about half of the adolescent boys reported having engaged in some daily exercise during

the previous week.

Earlier, the same investigator examined the dietary intake of 459 adolescents in

relationship to weight control behaviors using cross-sectional data from 4 regions of the

U.S.28 She found differences in nutrient intakes between female adolescent who were

dieting and those who were not, with dieters having lower energy intakes and higher

percentages of energy from protein. Other differences in nutrient intake were not

statistically significant although less healthy patterns among youth trying to control their

8

weight were noted. This author has made a case for distinguishing between positive and

negative attitudes and behaviors associated with dieting by adolescents.29 She proposed

categorizing behaviors as health promoting (exercising; eating fruits, vegetables, and

reduced fat foods; limiting amount of food; avoiding sweets and junk food) or health

compromising (using diet pills, laxatives or water pills; self-induced vomiting; skipping

meals; and fasting).

The relationship between dieting and weight change during adolescence was

examined in a prospective study following 692 ninth grade girls for four years.38

According to the authors, the most striking finding was that elevated dieting and radical

weight-loss efforts predicted greater subsequent growth in relative weight and an

increased risk for the onset of obesity. The authors theorized that weight-reduction

efforts reported by teens may not reflect decreased calorie intake, or that weight reduction

efforts by teens are a marker for a propensity to become obese. They were of the opinion

that the findings of this study could be used to dissuade youth from engaging in

ineffective dieting behaviors, which in turn, may help prevent the onset of both obesity

and eating pathology.

IMPROVING NUTRITIONAL QUALITY OF THE DIET

If calorie restricted diets are to be abandoned as the standard dietary approach to

treating pediatric overweight, what should replace them? There is almost unanimous

agreement that improving the nutritional quality of the diets of ALL children is critically

important in terms of promoting health and reducing the risk of obesity as well as the risk

9

for chronic diseases. Clues as to how to improve dietary quality can be gleaned from food

consumption data gathered through the U.S. Department of Agriculture’s Continuing

Survey of Food Intakes by Individuals.41

Reduce Soft Drink Intake

Data from the 1977-1979 and 1994 surveys indicate there has been a dramatic

increase in the consumption of soft drink among children and adolescents.18 Twelve

percent of preschool age children drank an average of 9 ounces or more of soft drink a

day. This increased to 33% among school-age children and over 50% among

adolescents. Closer examination of the data reveals that 22% of teenagers drank more

than 26 ounces of soft drink a day. It appears that soft drinks are replacing more

nutritious beverages like milk in the diets of children since milk consumption has

declined during the same period.25 Data show that high soft drink consumption leads to

excessive caloric intake. School-age children who consumed soft drinks had a mean

energy intake of 2,018 kcal/day as compared to an energy intake of 1,830 kcal/day for

children who did not drink soft drinks.

Concern has also been raised recently about the association of carbonated

beverage consumption and bone fractures in adolescent girls, especially because of their

proneness to osteoporosis in later life.42,43

There are no nutritional risks associated with eliminating soft drink from the diets

of children. In examining food sources of nutrients for U.S. children soft drink

contributes 4.3% of the energy intake of children ages 2 to 18 years, but did not

contribute significant amounts of any other nutrients. 40 Reducing soft drink consumption

10

has the potential to be nutritionally beneficial, particularly in cases where it has replaced

milk in the diet and adversely affected calcium intake.

Decrease Calories from Fat, Saturated Fat, and Cholesterol by Increasing Fruit,

Vegetable and Whole Grain Consumption

Morton and Gurthrie examined USDA’s Continuing Survey of Food Intakes by

Individuals to assess changes in children’s fat intake and determine the implications for

dietary quality.25 They found 2 to 17 year old children had reduced a percentage of

calories from fat but did not have decreased fat intake since overall caloric intake had

increased. There was increased carbohydrate consumption; however, a considerable

proportion of this increase was due to increase in refined carbohydrates coming from

sweeteners in sodas rather than from complex carbohydrates, as recommended in the US

Dietary Guidelines.

Peterson and Grant studied the impact of specific dietary strategies for reducing

fat intake on the nutrient intake of children.30 They found that American children who

used skim milk in places of higher-fat milks can closely approximate dietary

recommendations while maintaining adequate micronutrient intake. However, replacing

higher fat meats with lean meats had the potential to compromise energy and vitamin E

intakes unless this change was guided and monitored by a health professional. Using fat-

modified versions of cheese, yogurt, salad dressing, cake and pudding made no

significant impact on energy or micronutrient intake and did not achieve dietary

recommendations for either total far or saturated fat.

11

To determine if a low fat diet is associated with increased risk of nutritional

inadequacy in children 2 to 8 years old, Ballew and her colleagues used the USDA

Continuing Survey of Food Intake by Individuals to identify children who had different

levels of fat intake.2 They found that moderate-fat diets were not consistently associated

with an increased proportion of children at risk for nutritional inadequacy, and higher-fat

diets were not consistently protective against inadequacy. They concluded that dietary

fat could be reduced by judicious selection of lower-fat foods without compromising

nutritional adequacy. Risk of nutritional adequacy could be minimized with dietary

guidance of a health care professional.

Munoz and colleagues examined the food intakes of US children and adolescents

compared with recommendations.26 They found mean number of servings per day were

below minimum recommendations for all food groups except the dairy group for youth

ages 2 to 11. Percentages of youth meeting recommendations ranged from approximately

30% for fruit, grain, meat, and dairy to 36% for vegetables. Sixteen percent of youth did

not meet any recommendations, and a paltry 1% met all recommendations. They

concluded that children and teens in the United States follow eating patterns that do not

meet national recommendations.

The USDA Center for Nutrition Policy and Promotion issued a “Report Card on

the Diet Quality of Children” stating, “the dietary quality of children and adolescents

steadily declines as they get older.”23 Using the “Healthy Eating Index” to evaluate the

diets of children, they recommended substantial improvement in the diets of children by

increasing fruits, vegetables, and milk products. This concurs with Munoz’s

recommendations for improvement of the dietary intake of children.26

12

The diets of children in the U.S are not consistent with the recommendations in

the Food Guide Pyramid,20 nor do they conform to the Dietary Guidelines for

Americans.9 Overweight children are more susceptible to the risks of poor dietary intake

as evidenced by the fact that their genetic tendency towards obesity has been expressed in

a poor quality food environment. However, all children will eventually be affected by

this situation in terms of increased risk for chronic disease during their adult years.

GUIDELINES FOR TREATMENT

The publication of articles on the etiology, assessment, prevention, and treatment

of pediatric overweight has increased as various government agencies have focused

increased attention on this problem.22,24,32,33,39 A number of organizations have issued

guidelines or policy statements regarding treatment.3,6,7,17 These are summarized in Table

1, Guidelines for the prevention and treatment of childhood overweight issued by

national groups and organizations. All of these organizations are emphatic about the

need for multi-dimensional approaches targeting families of obese children and focusing

on encouraging and supporting behavioral changes in eating and activity patterns. All

acknowledge the need to promote self-esteem among children as well as help them

establish realistic expectations with respect to body size and shape. And all recognize the

risk of inadvertently promoting disordered eating, body dissatisfaction, and poor body

image.

13