Prevention and Management of Childhood Obesity and its Psychological and Health Comorbidities

Justin D. Smith, PhD1, Emily Fu, MPH2, Marissa Kobayashi, MHS3

1Department of Psychiatry and Behavioral Sciences, Department of Preventive Medicine, and Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, 750 N. Lake Shore Drive, Illinois, 60611, USA

2Department of Psychiatry and Behavioral Sciences, Northwestern University Feinberg School of Medicine, 750 N. Lake Shore Drive, Chicago, Illinois, 60611, USA

3Department of Public Health Sciences, University of Miami Miller School of Medicine, 1120 NW 14th Street, Suite 1009, Miami, FL 33136. Phone: (305) 972-9961

Abstract

Childhood obesity has become a global pandemic in developed countries, leading to a host of

medical conditions that contribute to increased morbidity and premature death. The causes of

obesity in childhood and adolescence are complex and multifaceted, presenting researchers and

clinicians with myriad challenges in preventing and managing the problem. This chapter reviews

the state-of-the-science for understanding the etiology of childhood obesity, the preventive

interventions and treatment options for overweight and obesity, and the medical complications and

co-occurring psychological conditions that result from excess adiposity, such as hypertension,

non-alcoholic fatty liver disease, and depression. Interventions across the developmental span,

varying risk levels, and service contexts (e.g., community, school, home, and healthcare systems)

are reviewed. Future directions for research are offered with an emphasis on translational issues

for taking evidence-based interventions to scale in a manner that reduce the public health burden

of the childhood obesity pandemic.

Keywords

adiposity; childhood obesity; health psychology; prevention; research translation

1.0 INTRODUCTION

Influenced by genetics, biology, psychosocial factors, and health behaviors, overweight and

obesity (OW/OB) in childhood is a complex public health problem affecting the majority of

developed countries worldwide. Additionally, the key contributors to obesity—poor diet and

physical inactivity—are among the leading causes of preventable youth deaths, chronic

[email protected]

DISCLOSURE STATEMENT Justin D. Smith is co-developer of the Family Check-Up® 4 Health intervention for childhood obesity. The authors are not aware of any other affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this review.

HHS Public Access Author manuscript Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

Published in final edited form as: Annu Rev Clin Psychol. 2020 May 07; 16: 351–378. doi:10.1146/annurev-clinpsy-100219-060201.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

disease, and economic health burden (Friedemann et al 2012, Hamilton et al 2018). Despite

the remarkable need to prevent childhood obesity and to intervene earlier to prevent excess

weight gain in later developmental periods, few interventions have demonstrated long-

lasting effects or been implemented at such a scale to have an appreciable public health

impact (Hales et al 2018).

In this review, we describe the extent and nature of the childhood obesity pandemic, present

conceptual and theoretical models for understanding its etiology, and take a translational-

developmental perspective in reviewing intervention approaches within and across

developmental stages and in the various contexts in which childhood OW/OB interventions

are delivered. We pay particular attention to co-occurring psychological conditions

intertwined with OW/OB for children, adolescents, and their families as they relate to both

development/etiology and to intervention. For this reason, our review begins with

interventions aimed at prevention and moves to management and treatment options for

obesity and its psychological and medical comorbidities. Then, we discuss the state-of-the-

science and expert recommendations for interventions to prevent and manage childhood

OW/OB and what it would take to implement current evidence-based programs at scale.

Last, we end by discussing identified gaps in the literature to inform future directions for

research and the translation of research findings to real-world practice that can curb the

pandemic. For readability, we use the term “interventions for the prevention and

management of childhood OW/OB” to capture an array of approaches referred to by a

variety of monikers in the literature, including primary prevention, prevention of excess

weight gain, weight loss intervention, weight management, and treatment of obesity. More

specific labels are used when needed.

2.0 EPIDEMIOLOGY OF CHILDHOOD OBESITY

Childhood OW/OB is determined by the child’s height and weight to calculate body mass

index (BMI), which is adjusted according to norms based on the child’s age and gender.

BMI between the 85th and 94th percentile is in the “overweight” range, whereas BMI ≥ 95th

percentile for age and gender is in the “obese” range (Centers for Disease Control and

Prevention [CDC] 2018). Rates of obesity among children and adolescents in developed

countries worldwide, collected in 2013, were 12.9% for boys and 13.4% for girls (Ng et al

2014). In the United States (US) from 1999–2016, 18.4% of children ages 2–19 years had

obesity, and 5.2% had severe obesity, defined as BMI ≥120% of the 95th percentile for age

and gender (Skinner et al 2018). The prevalence of obesity has increased between 2011–

2012 and 2015–2016 in children ages 2–5 and 16–19 years (Hales et al 2018). Being in the

obese range during childhood or adolescence makes the youth five times more likely to be

obese in adulthood compared to peers who maintain a healthy weight (Simmonds et al

2016). Compared to obesity, severe obesity is strongly linked with greater cardiometabolic

risk, adult obesity, and premature death (Skinner et al 2015).

OW/OB and its health consequences are disproportionately distributed across the US, with a

higher prevalence among children of disadvantaged racial and socioeconomic backgrounds.

Rates of OW/OB are significantly higher among Non-Hispanic black and Hispanic children

compared to Non-Hispanic White children (e.g., Hales et al 2018). Such disparities are

Smith et al. Page 2

Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

particularly pronounced among severe obesity, where 12.8% of African American children,

and 12.4% of Hispanic children have severe obesity compared to 5.0% of Non-Hispanic

White children (Hales et al 2018). Youth in low socioeconomic households are more likely

to develop OW/OB compared to their counterparts in high socioeconomic households. In

2011–2014, 18.9% of children ages 2–19 living in the lowest income group (≤130% of

Federal Poverty Line) had obesity, whereas 10.9% of children in the highest income group

(>350% Federal Poverty Line) had obesity (Ogden et al 2018). Influences on multiple

socioecological levels put racially diverse children of low socioeconomic status (SES) at

higher risk of developing OW/OB, which is further exacerbated by limited access to health

services that can prevent excess weight gain and its sequelae.

3.0 ETIOLOGY OF CHILDHOOD OBESITY

At the most basic level, childhood OW/OB emerges from consuming more calories than

expended, resulting in excess weight gain and an excess body fat. Caloric imbalance is the

result of, and can be further exacerbated by, a range of obesogenic behaviors. That is,

behaviors that are highly correlated with excess weight gain. The most common obesogenic

behaviors are high consumption of sugar sweetened beverages and low-nutrient, high

saturated fat foods, low levels of physical activity and high levels of sedentary behaviors,

and shortened sleep duration (e.g., Sisson et al 2016). Diet, physical activity, screen time,

and sleep patterns are influenced by a myriad of factors and interactions involving genetics,

interpersonal relationships, environment, and community (e.g., Russell & Russell 2019,

Smith et al 2018d). Children living in the United States commonly consume the “Western

Diet,” known as a diet high in calories, rich in sugars, trans and saturated fats, salt and food

additives, and low in complex carbohydrates, and vitamins. Poor sleep patterns, defined as

short duration and late timing, can contribute to obesity through changing levels of appetite-

regulating hormones, and irregular eating patterns including late night snacking and eating

(Miller et al 2015). Children who experience shortened night time sleep from infancy to

school age are at increased risk of developing OW/OB compared to same-aged children

sleeping average, age-specific hours (e.g., Taveras et al 2014). Research indicates that

children with higher rates of screen time also consume high levels of energy-dense snacks,

beverages, and fast food, and fewer fruits and vegetables, and screen time is hypothesized to

affect food and beverage consumption through distracted eating, reducing feelings of satiety

or fullness, and exposure to advertisements for junk food (sweet and salty, calorically-dense

foods) (Robinson et al 2017). Screen time can also negatively affect children’s sleeping

patterns, and is correlated with sedentary behaviors (e.g., watching television, playing video

games) (Hale & Guan 2015).

3.1 Conceptual Models for Understanding and Addressing Childhood OW/OB

Conceptualizing development of childhood OW/OB requires consideration of interplay of

genetic, biological, psychological, behavioral, interpersonal, and environment factors

(Kumar & Kelly 2017). OW/OB interventions are typically designed to account for these

multilevel factors to assist children in achieving expert recommendations for physical

activity and fruit and vegetable consumption, while limiting sugar sweetened beverages

intake and screen time, and regulating sleep patterns (Kakinami et al 2019). Creating

Smith et al. Page 3

Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

behavioral change requires understanding of the multi-level interactions to identify

opportunities for intervention to prevent excess weight gain long-term. A variety of

conceptual models exist to explain potential interactions and individual influences leading to

obesogenic behaviors and development of childhood OW/OB, and targets for improving

health behaviors and routines. Importantly, basic science and conceptual models can be

translated to develop effective, targeted intervention programs for prevention of excess

weight gain.

3.1.1 Biopsychosocial model—The biopsychosocial model combines biological foundations in child development with environmental and psychosocial influences to

identify and address mechanisms and processes to prevent and manage development of

childhood OW/OB (Russell & Russell 2019). This model features biological factors, such as

genetics, alongside environmental, psychosocial, and behavioral risk factors (e.g., family

disorganization, parenting skills, feeding practices, child appetite, temperament), and the

development of self-regulation. Such an approach can illustrate developmental processes

interacting with biological underpinnings that can be targeted in prevention and management

interventions for OW/OB. Intervening from a biopsychosocial model involves cognitive

behavioral and behavioral therapy to reframe thoughts and replace unhealthy eating

behaviors with new habits.

3.1.2 Ecological systems theory (EST)—EST embeds individual development and change within multiple proximal and distal contexts and emphasizes the need to understand

how an “ecological niche” can contribute to the development of specific characteristics, and

how such niches are embedded in more distal contexts (Davison & Birch 2001). For

example, a child’s ecological niche can be the family or school, which are embedded in

larger social contexts, such as the community and society. Individual child characteristics,

such as gender and age, interact within and between the family and community context

levels, which all influence development of OW/OB. The EST model presents various

predictors of childhood OW/OB through identifying risk factors moderated by

intraindividual child characteristics. The structure of the EST is present in various studies

examining influences of community exposures and children’s individual attributes on weight

outcomes.

3.1.3 The Six C’s Model—The Six-C’s is a developmental ecological model that includes environmental (family, community, country, societal), personal, behavioral, and

hereditary influences, and a system for categorizing environmental influences, all of which

can be adapted to each stage of child development from infancy to adolescence (Harrison et

al 2011). The Six C’s stand for: cell, child, clan, community, country, and culture, which

represent biology/genetics, personal behaviors, family characteristics, factors outside of the

home including peers and school, state and national-level institutions, and culture-specific

norms, respectively. Each C includes factors that contribute to child obesity that occur and

interact simultaneously throughout child development. For example, among preschool age

children, obesity-predisposing genes (cell), excessive media exposure (child), parent dietary

intake (clan), unhealthful peer food choices (community), national economic recession,

Smith et al. Page 4

Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

(country) and oversized portions (culture), are all factors associated with obesity that can

occur simultaneously and interact during this developmental stage.

3.1.3 The developmental cascade model of pediatric obesity—The model described in the Smith et al. (2018b) article offers a longitudinal framework to elucidate the

way cumulative consequences and spreading effects of multiple risk and protective factors,

across and within biopsychosocial spheres and phases of development, can propel children

towards OW/OB outcomes. The cascade model of pediatric obesity (Figure 1) was

developed using a theory-driven model-building approach and a search of the literature to

identify paths and relationships in the model that were empirically based. The model allows

for different pathways and interactions between different combinations of variables and

constructs that contribute to pediatric obesity (equifinality), identifying multi-level risk and

protective factors spanning from the prenatal stage to adolescence stage. The complete

model can, but has yet to, be tested. The model focuses on intra- and inter-individual child

processes and mechanisms (e.g., parenting practices), while acknowledging that individuals

are embedded within the broader ecological systems. St. George et al (in press) then

conducted a systematic review of the intervention literature to elucidate the ways in which

the developmental cascade model of childhood obesity can inform and is informed by

intervention approaches for childhood OW/OB.

3.2 Psychosocial Contributors

3.2.1 Maternal mental and physical health—An emerging body of literature has shown a significant relationship between higher levels of parental stress and youths’ higher

weight status and unhealthy lifestyle behaviors (Tate et al 2015). In a prospective study,

Stout et al (2015) found that fetal exposure to stress, as evidenced by elevated maternal

cortisol and corticotropin-releasing hormone, was related to patterns of increasing BMI over

the first 24 months of life. Children of mothers experiencing psychological distress and

anxiety during pregnancy had higher fat mass, BMI, subcutaneous and visceral fat indices,

liver fat fraction, and risk of obesity at age 10 years compared to those whose mothers did

not (Vehmeijer et al 2019). Early stress can have long-lasting effects, and studies from a

nationally-representative cohort study have shown that postnatal maternal stress during the

first year has a positive longitudinal relationship with the child’s BMI up to age 5 (Leppert

et al 2018), and psychological distress at age 5 was associated with risk of obesity at age 11

in another nationally-representative cohort (Hope et al 2019). Among Hispanic children and

adolescents whose caregivers reported ≥ 3 chronic stressors, Isasi et al (2017) found an

increased likelihood of childhood obesity when compared to those whose parents reported

no chronic stressors. In a systematic review assessing the impact of maternal stress on

children’s weight-related behaviors, O’Connor et al (2017) found mixed evidence for the

relationship specific to dietary intake; however, researchers found consistent evidence for the

detrimental impact on youths’ physical activity and sedentary behavior, which was often

conceptualized as screen time. Understandably, highly stressed parents may have an

increased reliance on convenient fast-food options versus grocery shopping and preparing

fresh and healthy meals for their children and may not have the energy or wherewithal to

support their youths’ physical activity, nor engage in limit-setting behaviors specific to their

children’s screen time.

Smith et al. Page 5

Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

One of the few studies using a longitudinal design did not replicate the relationship between

high parental stress and lower levels of youth physical activity, but the relationship held for

high levels of parental stress and increased fast food consumption (Baskind et al 2019).

Interestingly, this study observed an interaction effect on the relationship of high parental

stress and childhood obesity by only low-income households and among ethnic minority

children, specifically non-Hispanic black children—explaining one of the factors that

contributes to healthy disparities for childhood obesity rates in the US. In another study

using a large, prospective cohort, Shankardass et al (2014) found a significant effect of

parental stress on BMI. The researchers also observed a significantly larger effect among

Hispanics versus the total sample population, further noting that the relationship was weaker

and not statistically significant among non-Hispanic children. Due to the salient role of

caregiver stress on child health behaviors, it seems that interventions for childhood OW/OB

should incorporate stress reduction strategies for parents while simultaneously focusing

efforts on reaching racial/ethnic minority families and the economically disadvantaged.

Maternal mental health, most commonly operationalized as depressive symptoms and

diagnosis, relate to children’s risk for OW/OB. The longitudinal effects of postnatal

maternal depressive symptoms predicted obesity risk in preschool-age children, and

unhealthier lifestyle behaviors, such as high TV viewing time and low levels of physical

activity (Benton et al 2015). Children of mothers with severe depression were more likely to

be obese compared to children of mothers with fewer symptoms (Marshall et al 2018).

Maternal mental health could negatively affect child feeding behaviors such that elevated

depressive symptoms in low-income mothers have been associated with increased use of

feeding to soothe children (Savage & Birch 2017). Few interventions for childhood obesity

to date specifically target caregiver depression, but some protocols provide guidance to

engage caregivers in services to manage depression and related stressors (Smith et al 2018c).

3.2.2 Child mental health—Poor self-regulation and related constructs such as reactivity and impulsivity, are prospective obesogenic risk factors (Bergmeier et al 2014,

Smith et al 2018d). A child’s temperament describes behavioral tendencies in reactivity and

self-regulation. Negative reactivity is characterized by a quick response with intense

negative affect, and is difficult to soothe. Infants and children with negative reactivity are at

high risk of excess weight gain, and developing obesity later on and toddlers with low self-

regulation and inability to control impulses or behavior are at increased risk for obesity and

rapid weight over the subsequent nine years compared to toddlers with higher self-regulation

abilities (Graziano et al 2013). Poorer emotional self-regulation at age 3 is an independent

predictor of obesity at age 11 (Anderson et al 2017). On the other hand, the ability to delay

gratification at age 4 is associated with lower BMI 30 years later (Schlam et al 2013). It is

possible that parents of children with difficult temperament experience challenges effectively

managing children’s behaviors and setting limits, leading to irregular health routines and

increased obesity risk (Bergmeier et al 2014, Smith et al 2018d). Further, parents could

overuse food and feeding to soothe children (Anzman-Frasca et al 2012). Throughout

childhood, emotional regulation deficits and other mental health disorders continue to

predict obesity and weight gain. Emotional regulation in conjunction with stress during

childhood is highly linked to low physical activity, emotional eating, irregular and disrupted

Smith et al. Page 6

Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

sleep, and later development of obesity (Aparicio et al 2016). A longitudinal study

examining emotional psychopathology in preadolescence saw that boys diagnosed with a

social phobia, panic disorder or dysthymia (persistent depressive disorder) had higher waist

circumference and/or BMI, and girls diagnosed with dysthymia had increased waist

circumference at the three-year follow-up (Aparicio et al 2013). In a prospective study,

overweight children who reported binge eating at ages 6–12 years gained 15% more fat mass

over a period of four years compared to overweight children with no binge eating (Tanofsky-

Kraff et al 2006). The predictive role of mental health on physical health conditions and

subsequent comorbidities can be costly and burdensome. Children with obesity-related

health conditions (e.g., type 2 diabetes, metabolic syndrome) and a comorbid psychiatric

diagnosis (e.g., depressive mood disorder, bipolar disorder, attachment disorder) have higher

healthcare utilization and costs per year compared to children without a comorbid

psychiatric diagnosis (Janicke et al 2009a)

There is an association between OW/OB and depression in childhood and adolescence, but

there is mixed evidence of the directionality of this effect among children and adolescents. A

review of high quality studies by Mühlig et al (2016) saw that among nine studies examining

the influence of depression on weight status, six found no significant influence. Of the

studies that reported significant associations, one study saw effects only among female

adolescents, another only for male adolescents, and a third showed effects of adolescent

depressive symptoms on adult obesity at age 53 years only in women. Conversely, OW/OB

status can have significant influences on risk of low self-esteem and depressive symptoms/

diagnosis in adolescence, as discussed later in this paper.

3.2.3 Stigma/bullying—Weight-related stigma, defined as subtly or overtly having discriminatory actions against individuals with obesity, toward children with obesity can

impair quality of life, and contributes to unhealthy behaviors that can worsen obesity such as

social isolation, decreased physical activity, and avoidance of health care services (Pont et al

2017). Unfortunately, stigma is widespread and tolerated in society, furthering the reach of

negative harm. Children with obesity face explicit weight bias and stigma from multiple

environments including from parents, obesity researchers, clinical settings, and school.

Parents not only demonstrate implicit bias against childhood obesity, but also implicit and

explicit biases against children with obesity (Lydecker et al 2018). Even among obesity

researchers and health professionals, significant implicit and explicit anti-fat bias, and

explicit anti-fat attitudes increased between 2001–2013 (Tomiyama et al 2015). Exposure to

stigma and weight bias can have damaging psychosocial effects on children, such that

stigma can mediate the relationship between BMI, depression, and body dissatisfaction

(Stevens et al 2017).

Weight stigma can also initiate bullying and weight related teasing, which can have serious

psychological consequences such as depression among children, further weight gain and

lessen motivation to change. A nationally representative sample of children ages 10–17

years saw that OW/OB adolescents were at higher odds of being a victim of bullying, and

also higher odds of perpetrating bullying and victimizing others (Rupp & McCoy 2019). The

children at higher odds of engaging in bullying, or being bullied were also at significantly

higher odds of having depression, difficulty making friends, and conduct problems

Smith et al. Page 7

Annu Rev Clin Psychol. Author manuscript; available in PMC 2020 May 29.

A u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t A

u th

o r M

a n u scrip

t

compared to OW/OB adolescents who were not bullies or victims of bullying. The

relationship between obesity and bullying needs to be addressed through bullying

engagement, and coping skills for victimization to prevent and manage associated behavioral

and depressive symptoms.

3.2.4 Family functioning and home environment—Evidence suggests a link between general family functioning, parent–child relationships, communication, and use of

positive behavior support strategies and childhood OW/OB (see Smith et al 2017a).

Influence of general parenting styles, as opposed to the more specific feeding styles, have

been extensively studied and linked to children’s diet, physical activity, and weight (Shloim

et al 2015). Children raised with an authoritative (warm and demanding) parenting style had

healthier diet, higher physical activity levels, and lower BMI’s than those raised with the

other styles (Sleddens et al 2011). Parents proactively structuring home environments to

support and positively reinforce healthy dietary and physical activity behaviors also play a

key role in children’s healthy lifestyles (Smith et al 2017b). Children exposed to less

supportive environments consisting of family stress, father absence, maternal depression,

confinement, and unclean home environments at 1 year of age has been associated with high

BMI at age 21 (Bates et al 2018). Taken together, family participation and building parenting

skills can play a salient role in the prevention of childhood OW/OB (Pratt & Skelton 2018,

Wen et al 2011).

4.0 PREVENTION AND MANAGEMENT OF OVERWEIGHT AND OBESITY

This section discusses the state-of-the-science in childhood OW/OB prevention and

management along with salient factors related to their implementation in varied healthcare

delivery systems. The current climate is being shaped by the position of the American

Medical Association. In 2013, the Board voted to classify obesity as a disease that requires

medical attention. This classification aimed to emphasize health risks of obesity, remove

individual blame, and create new implications and opportunities for intervention. This

classification can help to further: 1) a broader public understanding of the obesity condition

and associated stigma; 2) prevention efforts; 3) research for treatment and management; 4)

insurance reimbursement for intervention; and 5) medical education (Kyle et al 2016). In

primary healthcare settings specifically, the US Preventive Services Task Force (USPSTF) </p