Treatment of Adolescent Obesity

Home , Adolescent Health, Management of obesity

REvIEwS

Treatment of adolescent obesity

Katharine S. Steinbeck1, Natalie B. Lister1,2, Megan L. Gow2 and Louise A. Baur1,3* Abstract | The increased prevalence of adolescent obesity and associated short-term and long- term complications emphasize the need for effective treatment. In this Review , we aim to describe the evidence for, and elements of, behaviour management and adjunctive therapies and highlight the opportunities and challenges presented by obesity management in adolescence. The broad principles of treatment include management of obesity-associated complications; a developmentally appropriate approach; long-term behaviour modification (dietary change, increased physical activity , decreased sedentary behaviours and improved sleep patterns); long- term weight maintenance strategies; and consideration of the use of pharmacotherapy , more intensive dietary therapies and bariatric surgery. Bariatric surgery should be considered in those with severe obesity and be undertaken by skilled bariatric surgeons affiliated with teams experienced in the medical and psychosocial management of adolescents. Adolescent obesity management strategies are more reliant on active participation than those for childhood obesity and should recognize the emerging autonomy of the patient. The challenges in adolescent obesity relate primarily to the often competing demands of developing autonomy and not yet having attained neurocognitive maturity.

Adolescence is one of the most rapid phases of human evidence for, and elements of, behavioural management, development, associated with marked physical, neuro the use of adjunctive therapies (such as more intensive developmental, psychological and social changes. The dietary therapies, pharmacotherapy and bariatric sur WHO defines an adolescent as a young person in gery) and the management of the metabolic complica the second decade of life (10–19 years)1. Adolescent tions of obesity such as insulin resistance and polycystic obesity has become a globally prevalent health problem ovaries. The opportunities and challenges presented by affecting young people from low-income and middle- the management of obesity specifically in adolescents income countries, as well as those from high-income are highlighted. countries2,3. Figure 1shows the trends in obesity for children and adolescents aged 5–19 years from differ Clinical assessment ent regions over 40 years from 1975 to 2016 (ref.3). A A thorough clinical history and physical examination widening social disparity is recognized in economi of the adolescent is essential to assess current obesity- cally advanced countries, with an increased prevalence associated complications, the risk of future compli of obesity in more socially disadvantaged adolescents cations and whether the patient has any modifiable 1Discipline of Child and compared with adolescents from higher socioeconomic lifestyle practices. It is beyond the scope of this Review Adolescent Health, University groups4. Additionally, growing evidence indicates that to provide further details on clinical assessment, and of Sydney, Sydney, New the prevalence of severe or morbid obesity is increas readers are referred to recent reviews and clinical guide South Wales, Australia. ing in this age group in several countries, including the lines for additional information10–14. Key recommenda 2 Institute of Endocrinology United States and Australia5–7. tions for history-taking and physical examination are and Diabetes, The Children’s Boxes 1,2 Hospital at Westmead, Obesity in adolescence is often complicated by outlined in . Westmead, New South Wales, psychosocial distress and can be associated with a range Australia. of other health problems; the normal neurobiological Factors that affect management 3Weight Management development that occurs in this age group also has a con Neurocognitive development Services, The Children’s siderable effect1. Furthermore, the rate of spontaneous The pace of brain development in adolescence and into Hospital at Westmead, remission of obesity to a healthy weight is low; at least young adulthood is second only to that of early child Westmead, New South Wales, 15 Australia. 90% of adolescents with obesity will have overweight or hood . Pruning of previous synaptic proliferation to 8,9 *e-mail: louise.baur@ obesity in young adulthood . optimize neural networks and increasing myelination health.nsw.gov.au For all these reasons, both the prevention and the of nerve fibres for rapid neural messaging are impor https://doi.org/10.1038/ treatment of adolescent obesity are important. In this tant for the development of mature and efficient neu s41574-018-0002-8 Review, we focus on obesity treatment and detail the ral networks. The maturation of the emotional brain

Stigmatization Key points Weight stigmatization is widespread and leads to • The preferred treatment approach for adolescent patients with obesity includes psychological, social and physical health consequences. management of obesity-associated complications, a developmentally appropriate Discussion of the causes and effects of weight bias and approach, support for long-term behavioural change, long-term weight maintenance stigmatization are beyond the scope of this Review, strategies and consideration of other therapies. and readers are referred to recent reviews for further • Long-term weight maintenance therapy, through face-to-face or electronic support, details22–25. Health-care professionals have an important phone coaching or group programmes, is likely to be required. role in reducing the negative effects of obesity stigma 2 • Bariatric surgery should be considered in adolescents with severe obesity (BMI >40 kg/m , tization. The American Academy of Pediatrics recom or 35 kg/m2 in the presence of complications), and patients should be treated in > mends that clinical care can be improved through role centres experienced in bariatric surgery. modelling, using appropriate language, good clinical • Pharmacotherapy should be used as adjunctive therapy, particularly in the documentation and ensuring a safe and welcoming maintenance of weight loss, and for the treatment of obesity-associated 22 complications. environment . Elements of behavioural health screening and behaviour change counselling, as outlined in more • Clinicians should understand where the adolescent is in puberty development — age does not equal stage — and not support beliefs that young people will grow out of detail in subsequent sections, can identify and address obesity with the onset of puberty. weight-based victimization to improve outcomes. • Clinicians must work within the contexts of both the normal rapid changes in adolescent brain development and the psychosocial tasks of adolescence that are Overview of treatment principles necessary for mature adulthood. Effective obesity treatment in adolescents can be defined in various ways, including a reduction in BMI or weight outcomes, improvement in obesity-associated compli with its associated reward-seeking, hyperexcitability cations, a reduction in markers of risk of future com and proneness to impulse occurs in mid-adolescence plications or a changed weight gain trajectory. In adult (14–16 years). The development of the prefrontal studies, effectiveness is often categorized into ‘success’ or cortex, which is responsible for planning, organiza ‘failure’; however, no standard definition of effectiveness tion, risk versus benefit assessment and the ability to exists26,27, and different criteria have been used between delay immediate gratification for future gain, occurs and within studies to classify weight maintenance or later with final maturation in the mid-20s16 and regain. Calculations can also include initial weight, has important implications for the management of initial weight loss, postintervention weight, postinter adolescent obesity. vention weight change or final weight. In adolescents, Choices based on emotion or impulse, difficulty in dichotomizing success is even more difficult given the consistently following through with action plans, ina need to normalize measures for age and sex in adoles bility to comprehend future health risks, the need to be cents aged <18 years (hence the frequent reporting of like their peers (which makes lifestyle change challeng BMI z-scores) and the variability in the reporting of ing) and the need to be independent (which reduces treatment outcomes. A systematic review and meta- responsiveness to adult directives and advice as a result regression published in 2016 reports how weight loss of prefrontal cortex immaturity16) are all likely to derail in children and adolescents translates to favourable car attempts at lifestyle change. The concept that lifestyle diometabolic outcomes28 (such as blood pressure and change programmes for adolescents need to be formu HDL cholesterol levels), which might be more clinically lated differently from those of adults now has a strong meaningful than weight loss perse. neurocognitive rationale and an emerging clinical trial evidence base17,18. Findings from systematic reviews The 2017 Cochrane Review on diet, physical activity Puberty and behavioural interventions for the treatment of obe Puberty is an early adolescent event that results in dra sity in adolescents aged 12–17 years included 44 rand matic changes in body composition and has long-term omized controlled trials (RCTs) with follow-up periods implications for health and well-being19. Puberty might of 6–24 months17. A range of trials that mostly used have both positive and negative effects on weight man multibehavioural change interventions were included agement. Positives include increasing lean body mass in the review, with no specific programme being rec in males and, for both sexes, if obesity management is ommended over another. A meta-analysis of studies undertaken before the height growth spurt, then the comparing intervention groups versus wait-list or no- caloric deficit induced with weight maintenance in a treatment control groups showed that, at the longest situation where weight would naturally increase with follow-up period, there was a mean reduction in BMI of height increase results in reduced adiposity. Negatives −1.18 kg/m2 (95% CI −1.67 to −0.69), in BMI z-score include increasing adiposity in females as a result of of −0.13 units (95% CI −0.21 to −0.05) and in body gonadal hormones and, in both sexes, the possibility that weight of −3.67 kg (95% CI −5.21 to −2.13). The effect the insulin resistance that occurs during puberty (related on weight outcomes persisted in those trials with longer in part to increased secretion of growth hormone) follow-up periods of 18–24 months, a finding that is might persist after puberty is complete if there are other somewhat at variance with those from studies in adults, drivers such as obesity present20,21. These factors need which show more weight regain25. to be taken into account when managing individuals The authors of the Cochrane Review noted that, with obesity. where measured, there were improvements in percentage

60 60

40 40

20 20 Number of girls with obesity (millions) Number of boys with obesity (millions)

0 0 1980 1990 2000 2010 1980 1990 2000 2010 Year Year

West Africa Central Asia Melanesia Caribbean High-income English-speaking countries Central Africa South Asia Polynesia and Andean Latin America Northwestern Europe Southern Africa Southeast Asia Micronesia Central Latin America Southwestern Europe East Africa East Asia Southern Latin America Central Europe Middle East and High-income Asia Paciﬁc Eastern Europe North Africa

Fig. 1 | Trends in the number of girls (left panel) and boys (right panel) aged 5–19 years with obesity , by region, from 1975 to 2016. There has been a dramatic increase in the number of children and adolescents affected by obesity over more than four decades. Middle Eastern, East Asian, high-income English-speaking and central and southern Latin American countries are especially affected. Adapted with permission from ref.3, CC-BY-4.0.

body fat, waist circumference and total fat mass follow Adapting the findings from such studies to ‘real- ing behavioural change interventions17. Improvements world’ obesity clinics that might be more poorly in quality of life, although not in self-esteem, were also resourced is often difficult25. For example, in comparison noted at the longest follow-up. However, outcomes for with clinical trials, in usual clinical practice it might not dietary intake and physical activity were inconsistent be possible to review patients as frequently and suffi across the included studies. The review reported no sub cient specialist nursing, allied health or medical exper group differences in interventions with or without paren tise might not be available. Further, the patients in usual tal involvement or by intervention type, setting (such as clinical practice might have additional comorbidities or school, clinic or community) or mode of delivery (group be more socially disadvantaged than those who partic versus individual). The overall quality of evidence for the ipate in clinical trials, raising challenges in optimizing primary outcomes of interest was rated as low, except treatment adherence. Nevertheless, the principles of for body weight, which was rated as moderate. Adverse management are well recognized10–14 (Box 3). effects were only infrequently reported and hence could not be reliably assessed. Elements of treatment of obesity The findings of this most recent systematic review Behavioural management add to the evidence base from previous reviews12,18,29,30. Behavioural strategies are routinely used to support The 2017 Cochrane Review did not report outcomes for changes in dietary intake, physical activity levels and cardiometabolic parameters from behavioural change sedentary behaviours and to facilitate long-term main interventions, although they were previously reported in tenance of these changes31,32. Although most RCTs a 2012 systematic review18. This review showed that life of adolescent obesity management include a behav style interventions led to statistically significant improve ioural component, these details are often inadequately ments in levels of LDL cholesterol (−0.30 mmol/l, 95% described in publications17,33. CI −0.45 to −0.15), triglycerides (−0.15 mmol/l, 95% CI A 2015 evidence update of behavioural strategies −0.24 to −0.07), fasting insulin (−55 pmol/l, 95% CI −71 for obesity treatment in children and adolescents to −39) and blood pressure (systolic −3.72 mmHg, 95% identified much stronger evidence (‘well-established CI −4.74 to −2.69; diastolic −1.69 mmHg, 95% CI −3.15 interventions’) in children than in adolescents for both to −0.24) up to 12 months from baseline. parent-only interventions and family-based behavioural

Box 1 | Key elements of history taking in adolescents with obesity general history • Consider prenatal and birth — including history of gestational diabetes and birthweight • Current medications — including glucocorticoids, some antiepileptics (for example, sodium valproate) and antipsychotics (for example, risperidone or olanzapine) Weight history • Onset of obesity and duration of parental and adolescent concerns about the patient's weight • Previous weight management interventions • Previous and current dieting behaviours complications history • Psychological effects of obesity, including bullying, poor self-esteem, anxiety and depression • Sleep routine, presence of snoring and/or possible sleep apnoea (for example, lack of refreshment after sleep, daytime somnolence and witnessed apnoeas) • Exercise tolerance • Specific symptoms, including - Acne and hirsutism - Morning headache and visual disturbance (potential benign intracranial hypertension) - Gastrointestinal: vomiting, abdominal pain, constipation and gastrointestinal reflux - Nocturnal enuresis and daytime dribbling - Hip and knee joint pain - Menstrual history (girls) Family history • Ethnicity (high-risk groups for cardiometabolic complications include Native American, Hispanic, Maori, Pacific Islander, Australian Aborigine, South Asian, East Asian, Mediterranean and Middle Eastern) • Family members with a history of - Obesity - Type 2 diabetes mellitus and gestational diabetes - Hypertension, dyslipidaemia and cardiovascular disease - Obstructive sleep apnoea - Polycystic ovary syndrome - Bariatric surgery - Eating disorders • Home environment, including - Household members - Parental relationship - Parental employment, hours and home supervision lifestyle factors • Diet and diet-related behaviours - Breakfast consumption - Eating patterns, including snacking, sneaking food, fast-food intake and binge eating - Beverage consumption (sodas, juice and other sugary drinks) - Family routines around food and eating • Physical activity - Transport to and from school - Sports participation - After-school and weekend recreation - Family activities • Sedentary behaviours - Screen time per day (TV, video game, cell phone, tablet and computer use) - Number of TVs, computers, tablets, laptops, cell phones and gaming consoles in the house and bedrooms - Pattern of TV or other screen viewing (for example, during meals)

treatments in which parents are specifically targeted in that few studies looked specifically at adolescents. to change their own behaviours and achieve weight Overall, the authors rated the evidence for behavioural loss31. The update included 11 studies involving both weight-loss treatment with family involvement in ado children and adolescents undergoing behavioural lescents as “possibly efficacious” while commenting weight-loss treatment with family involvement during that treatment is optimized if parents are included in which the parent facilitates the lifestyle changes. Nine the treatment programme. of these studies demonstrated improvements in weight Several common behaviour change techniques outcomes compared with the control condition. The are used in obesity treatment31,32,34. Some commonly evidence for family-based interventions in adolescent- used techniques include goal-setting, stimulus control only interventions was, however, far more limited (modifying or restricting environmental influences)

Box 2 | notable findings on physical examination10–14 Dietary interventions Following dietary interventions is effective for mod Skin or subcutaneous tissues est weight loss in adolescents with obesity18. Caloric • Acanthosis nigricans (neck, axillae, skin folds and over joints) restriction is usually incorporated into multicomponent • Keratosis pilaris lifestyle-based interventions for weight management. • Acrochordons (skin tags) Importantly, lifestyle interventions that include a dietary • Hirsutism component can improve cardiometabolic outcomes in • Acne adolescents with obesity even in the absence of weight 18,29,33 • Striae loss or changes in body composition . Various dietary manipulations can achieve caloric • Pseudo-gynaecomastia in males restriction and resultant weight loss in adolescents, • Intertrigo with or without secondary infection most commonly by using the ‘stop/traffic light approach’ • Xanthelasmas (hypercholesterolaemia) but also by altering macronutrient distributions and/or neurological improving carbohydrate quality. No dietary intervention • Papilloedema and/or reduced venous pulsations on fundoscopy (pseudotumour has been shown to be optimal, and adherence to die cerebri) tary protocols is the most effective predictor of weight Head and neck loss35. Dietary interventions with evidence of being • Tonsillar size and obstructed breathing effective for weight management are discussed in the cardiovascular following sections. • Hypertension • Heart rate (cardiorespiratory fitness) Stop/traffic light approach. This calorie-controlled inter respiratory vention categorizes foods according to caloric density and uses these groups to guide intake frequency (Fig. 2). • Exercise intolerance Low-calorie foods are labelled ‘green’ and can be eaten • Wheeze (asthma) freely, moderate-calorie foods are labelled ‘yellow’ and gastrointestinal should be eaten judiciously and high-calorie foods are • Hepatomegaly and hepatic tenderness (nonalcoholic fatty liver disease) labelled ‘red’ and should be eaten rarely36. This approach • Abdominal pain (secondary to gallstones or gastro-oesophageal reflux) can produce effective weight loss (BMI decreases range Musculoskeletal from 0.18 to 2.6 kg/m2)37,38 in adolescents in the short to • Pes planus medium term (6 months to 2 years)18. • Groin pain and painful or waddling gait (slipped capital femoral epiphysis) • Lower limb arthralgia and restriction of joint movement Macronutrient distribution. Altering the macronutrient Endocrine ratio of a calorie-restricted diet is purported to improve dietary adherence, weight-loss outcomes and cardi • Goitre ometabolic measures39. Common diet interventions • Extensive striae, hypertension, dorsocervical fat pad or ‘buffalo hump’ (signs of have different macronutrient distributions (Fig. 3). Cushing syndrome) Conventional interventions are typically referred to as • Pubertal staging ‘low fat’ and aim for <35% daily energy from fat. A low- • Reduced growth velocity carbohydrate diet, aiming for <50 g carbohydrate per day, Psychosocial has often been compared to a low-fat approach; however, • Flat affect and low mood (depression) a 2014 systematic review found no evidence that either • Poor self-esteem and social isolation approach was more effective for long-term weight loss 39 other — evidence of a possible underlying genetic syndrome in adolescents . Even so, low-carbohydrate diets might • Short stature or disproportion provide some benefit for short-term weight loss (BMI 2 39 • Dysmorphism decreases from baseline range from 1.2 to 5.2 kg/m ) and insulin concentrations (the decrease in fasting insu • Developmental delay lin levels from baseline range from 6.0 to 7.5 μU/l)40,41 compared with low-fat interventions. In longer-term and self-monitoring (Box 4). Note that goals need to be studies of >12 months, weight loss was not significantly concrete, developmentally relevant and achievable and different, possibly owing to eventual migration to a more might take considerable time to set and review in clin typical carbohydrate intake35,39. ical sessions. Reward systems support the repetition of Increasing protein from a typical 15% of daily energy a desired behaviour and can include rewards based on in conventional diets to 20–40% of daily energy has improvements in lifestyle behaviours as well as in weight no demonstrated benefit for weight loss in adolescents outcomes. Importantly, behaviour change strategies under isocaloric conditions39. Increasing protein intake should be used by all treating clinicians. is hypothesized to improve satiety and reduce hunger, Further research is needed to develop a stepped thereby resulting in a lower ad libitum energy intake model of care for treatment (especially for adolescents), and greater weight loss42. However, studies in free-living to identify the optimal intensity and frequency of behav adolescents demonstrate that achieving higher dietary ioural treatments for long-term sustainability and to protein targets is difficult43, which might contribute to enhance scalability and implementation into community the lack of effect of increased levels of dietary protein in and clinical settings. adolescents. The unique challenges to dietary compliance

Box 3 | Principles of obesity treatment completion of VLED) outcomes remain unclear. The severe nature of the diet requires intensive monitoring • Assess and manage obesity-associated disorders by health professionals; however, these diets could be • Ensure a developmentally appropriate approach an alternative to pharmacological therapies or surgical • Support long-term behaviour change interventions to treat adolescents with severe obesity. • Dietary change • Increased physical activity Eating behaviours • Decreased sedentary behaviours, including screen An emerging area of research is the role of eating behav behaviours iours in obesity, defined as “self-perceived behaviour and 52 • Improved sleep attitudes towards eating behaviour and food” (ref. ), • Strategies for long-term weight maintenance which could partly explain the different responses to dietary interventions seen in patients. Longitudinal data • Consider more intensive dietary interventions on eating behaviours and weight gain are limited, and • Consider pharmacotherapy few intervention studies directly assess more than one • Consider bariatric surgery eating behaviour53. In adolescents, prescriptive dietary advice consisting of hypocaloric nutrition coaching by a during adolescence are not well defined in the literature trained dietician can normalize eating behaviours in the but include peer influences, reliance on parental support short to medium term54. However, more data are needed (including the provision of meals) and the neurocognitive to prospectively examine the long-term effects of dietary development issues outlined in a previous section. interventions on the eating behaviours of adolescents.

Carbohydrate quality. Glycaemic index (GI) and gly Physical activity caemic load are used to describe carbohydrate qual Adolescent obesity is associated with suboptimal physi ity. Although low-GI diets do not seem to confer any cal activity levels, poor fitness and excessive time in sed weight-loss benefit for adolescents, a 2015 systematic entary behaviours55. Despite the certain contributions to review found that lower-GI diets improved insulin energy deficit, the specific weight-reducing effect of the resistance (HOMA index mean difference of −0.70, 95% physical activity component of a lifestyle intervention CI −1.37 to −0.04, P = 0.04) and triglyceride concentra in adolescents with obesity is not clear. However, phys tions (mean difference −15.14 mg/dl, 95% CI −26.26 to ical activity is still an integral part of obesity treatment −4.00, P = 0.008)44. The major limitation with manipulat owing to the additional health benefits attained through ing GI to improve carbohydrate quality is that lower-GI regular exercise, including improved body composi foods do not always translate to ‘healthy’ options; for tion, psychological profile and functional capacities55–62 example, high-fat discretionary foods such as ice cream (Fig. 4). Physical activity is also associated with improve and potato chips have a low GI. ments in the cardiometabolic profile of an adolescent with obesity, including improved insulin sensitivity, Intermittent fasting. Over the past decade, intermit blood pressure and cholesterol levels55–62. Hence, regu tent energy-restricted diets have gained popularity as lar physical activity has an important role in preventing a weight-loss approach in adults45,46. Limited evidence the development of the metabolic syndrome and T2DM is available for their use in adolescents; however, these in adolescents with obesity, irrespective of a change in diets have been used in clinical practice. Dietary model weight. One systematic review comparing diet-only ling and nutrient analysis show that, over a 7-day period with diet-plus-exercise or an exercise-only interven (3 days restricted to ~2.5–2.9 MJ or 600–700 kcal), inter tion found that, although weight loss was not different mittent energy-restricted meal plans can be designed to between the groups, the addition of exercise training to meet adolescents’ unique micronutrient requirements47. a dietary intervention led to a greater improvement in levels of HDL cholesterol and fasting insulin concentra More intensive dietary therapies tions in the short term (over 6 months) than a diet-only A very-low-energy diet (VLED) is an intensive die programme29. tary regimen, typically consisting of <800 kcal per day, Several national guidelines recommend that all ado with <50 g of carbohydrate and adequate provision of lescents partake in at least 60 minutes of daily moderate- micronutrients often being achieved using meal replace to-vigorous physical activity and that no more than 2 h ments such as shakes and bars. Owing to the difficulty per day is spent using electronic media for entertainment of achieving the strict calorie prescription, this diet is (for example, computer games, TV and Internet)63–66. typically recommended only for short periods of time However, few adolescents achieve these targets55. This (8–12 weeks)48. challenge is exemplified in adolescents with obesity, In adolescents with obesity, a VLED can safely induce who often report negative peer experiences around rapid weight loss in the short term (4–15 kg over 3–12 exercise67 and have reduced exercise tolerance, which weeks) while preserving lean body mass49,50. VLEDs presents a barrier to participation in physical activity also lead to short-term improvements in common met owing to the early onset of fatigue68. Reduced capacity abolic risk factors49,51. Published in 2016, a small pilot to perform basic physical functioning tasks, including study demonstrated that VLED reversed type 2 diabetes walking and climbing stairs, has led to obesity in adoles mellitus (T2DM) in newly diagnosed adolescents with cents being described as a disability69. Exercise capacity obesity50. However, the long-term (>12 months after in adolescents with obesity can be enhanced through

Box 4 | Examples of common behavioural change techniques no trials have been conducted in young people already affected by obesity75,76. Strategies might include an earlier goal-setting — performance goals adolescent bedtime, removal of electronic screen devices • “I will decrease my discretionary screen time from 6 h per day to 4 h over the next from the bedroom and family rules about exposure to month. I will do this by moving my cell phone out of my room at night (I will charge it screens before bedtime. Importantly, targeting sleep has in the family room) and by agreeing to move my laptop to the family room. When I additional benefits in terms of general health, mood, have done this, I will work out some new screen time goals.” school performance and quality of life77. • “I will go to bed an hour earlier on school nights (11 pm instead of midnight). I will do this by charging my cell phone in the family room overnight.” Digital technologies • “I will get an extra half hour of physical activity into my day. I will do this by catching High-quality studies using digital technologies to man the train and bus to and from school instead of asking my parents to drive me. I will age obesity are lacking78. A 2017 meta-analysis of all also ask my friend to go to the school gym with me twice a week after school.” mobile health interventions in youth concluded that goal-setting — outcomes goals these “appear to be a viable behaviour change interven • Specific weight-loss goals (amount of weight loss or reduction in waist circumference tion modality for youth” (ref.70). Digital technologies over a specified time), developed in consultation with the therapist represent a promising area of future clinical research Self-monitoring in adolescent obesity, including apps and activity track • Daily monitoring and recording of physical activity via pedometer or other activity ers as lifestyle monitors. Apps and activity trackers monitor can give instant feedback on levels of physical activity, • Daily recording of screen time sleep and active versus sedentary time, with little input • Daily recording of eating habits (for example, eating breakfast and eating lunch) or of required from the user. Popular apps are also used for specific food intake (for example, vegetable intake and soda drink intake) tracking diet or providing healthy food alternatives. • Weekly recording of weight Technologies including image or photo upload of foods Stimulus control and/or barcode scanning are designed to minimize the • Parent agrees to not offer to drive the young person to school time required for input from the user and enable instant • Television, laptops, cell phones and other mobile devices are out of the bedroom at feedback on dietary intake or advice on more appro night priate choices. However, the capacity of such devices to • Parent and young person agree to not store unhealthy foods (for example, processed assist weight loss is not yet clear, with a study published snack foods and soda drinks) in the house in 2016 demonstrating that adults randomly assigned • Parent agrees to serve evening meal using smaller-sized plates to the ‘wearable device’ (activity tracker) arm achieved • Parent agrees to have a family rule about eating the evening meal at the table without less weight loss at 24 months than those in the con 79 the television being on trol arm . Furthermore, these technologies will never replace a clinician, but they might be a useful tool to enhance motivation and assist in weight-loss mainte participation in a 12-week exercise intervention of twice nance, particularly in adolescents, who generally enjoy weekly 45–60 minute sessions, including resistance and digital devices. Discoveries from research in mental aerobic exercise, supervised by a personal trainer62. health provide support for the use of the Internet to By increasing adolescents’ participation in physical deliver interventions80. In order to be engaging and activity, it is anticipated that a flow-on effect will occur, to be considered for long-term use, such interventions resulting in reduced recreational screen time. However, (including apps) should be co-designed with adolescents both issues need to be addressed. A recreational screen and undergo regular updates. Their commercial viability time allowance of 2 h per day is increasingly difficult to is questionable. implement given the prominence and accessibility of digital technologies. Effective interventions for reducing Long-term weight outcomes screen time in adolescents with obesity are still needed. Few high-quality trials report long-term outcomes of Utilization of technologies, such as smartphone apps, is obesity treatment in adolescents, as opposed to chil a popular avenue of research in this area70. dren17,18,81. However, in a large central European health- care database from 157 paediatric specialist centres Sleep behaviours managing obesity (patients aged 5–25 years), 2-year Cross-sectional studies show an association between outcomes were reviewed for 3,135 patients82. Compared short sleep duration and an increased risk of over with children aged 5–11 years, adolescents aged weight or obesity in children and adolescents, with a 12–15 years had a 41% lower odds of successful weight pooled odds ratio of 1.89 (95% CI 1.43–1.68) for short reduction and maintenance, defined as a decrease in sleep duration (different definitions used in the vari BMI z-score of at least 0.2 from baseline. These findings ous studies, including <5 h, <6 h and <10 h) and obesity emphasize the importance of early treatment of estab being noted in a systematic review of 12 studies71. In a lished obesity. In a 12 month and 24 month follow-up of systematic review of 22 longitudinal studies, an inverse adolescents participating in a trial of two different types association was found between short sleep duration and of dietary intervention, weight loss in the first 3 months subsequent BMI in both children and adolescents72. On predicted long-term weight outcomes83, a finding con the basis of adult studies73,74, interventions targeting sistent with those from other studies. Such studies sug sleep in the management of adolescents with obesity gest that interventions with a focus on early weight loss might result in improved weight and body composi might have better outcomes than interventions started tion outcomes. Trial data in adolescents are limited, and after the patient has had obesity for some time.

Red foods Eaten rarely Examples Topiramate is currently marketed for epileptic seizures Greater than 20 kcal of the • Sweets and migraine and could be considered in adolescents average serving for foods • Desserts with obesity who fulfil prescribing criteria. Topiramate within that food group • Fast food • High-fat meat or has been trialled for weight loss in adults in combination meat substitutes with phentermine, with a significant dose-dependent percentage change in body weight of 7–9% compared Yellow foods Eaten judiciously Examples with placebo85. However, a pilot RCT in 30 adolescents Within 20 kcal of the average • Full-fat milk consisting of 24 weeks of topiramate following 4 weeks serving for foods within that • Whole-grain breads food group (i.e. fruit=40 kcal and cereals of meal replacement therapy did not result in significant and milk and dairy=80 kcal) BMI reduction87. Glucagon-like peptide 1 (GLP1) is a naturally occur Green foods Eaten freely Examples ring incretin produced by the gut88. GLP1R agonists Less than 20 kcal of the • Fruits average serving within that • Vegetables were originally marketed for the treatment of T2DM. food group • Legumes An additional effect is appetite suppression and weight • Low-fat milk loss89, and obesity management has become an addi • Lean meat or meat substitutes tional indication for the use of this drug class. In adults, the effects of liraglutide, which is approved for weight loss in many countries as a once-a-day injection, are Fig. 2 | Traffic light diet categories18,36,37,113. Foods are categorized according to high (top panel), moderate (middle panel) or low (bottom panel) caloric density. These categories impressive compared with the effects of standard oral are used to guide frequency of intake, with the aim to reduce overall calorie intake. anti-obesity therapy. Published in 2017, the results of an RCT showed that the oral GLP1R agonist semaglu tide had a statistically significant effect on weight loss In adults, a comprehensive weight maintenance pro of 2–7 kg as well as control of diabetes mellitus in par gramme is recommended for those who have had initial ticipants with T2DM and a mean age of 57 years. This weight loss32. This programme might entail continued finding suggests that oral therapy is a future, adolescent- support from a therapist, via phone or face to face, pos attractive weight-loss option90. In a pilot RCT, 3 months sibly on at least a monthly basis for ≥12 months. The of exenatide therapy in 26 adolescent patients with sustainability of such an approach in usual health-care severe obesity led to a greater percentage reduction in settings and the applicability to the adolescent popula BMI, as well as absolute weight and BMI, than placebo91. tion remain unclear. However, given the importance of digital technologies to young people, this might prove to Bariatric surgery be of relevance in weight maintenance strategies. In adults, bariatric surgery is a well-recognized form of therapy for severe obesity as an adjunct to behavioural Pharmacotherapy therapy and pharmacotherapy92. The role of bariatric In the long term, safe anti-obesity drugs must be a surgery in adolescents is being established. therapeutic goal given that the condition is chronic and lifestyle therapies have had limited results. Current Evidence for bariatric surgery anti-obesity agents are universally limited in both num To date, there has been only one published RCT of bar ber and availability, with few subsidized by govern iatric surgery in adolescents: an Australian study of ment agencies (Table 1). Trials are mostly performed in 50 patients aged 14–18 years (25 patients in both the adults, and use in adolescents is often off-label. A 2016 intervention and the comparator groups)93. The study Cochrane Review provided data for trials of current compared laparoscopic adjustable gastric banding (and some previously available) anti-obesity agents in (LAGB) with a multicomponent lifestyle programme. children and adolescents84. The authors also identified At 2 years, weight in the surgery group was reduced by ongoing trials for topiramate and glucagon-like peptide a mean of 34.6 kg (95% CI 30.2–39.0) compared with 1 receptor (GLP1R) agonists. Metformin and orlistat 3.0 kg (95% CI 2.1–8.1) in the lifestyle group. Although both result in a reduction in weight in favour of the both groups had improvements in the quality-of-life active intervention, albeit around 2 kg. The review also subscore of general health, the surgical group had bet describes a greater overall attrition rate for adolescents ter performance in four additional subscores (physical than reported in adult trials. Metformin and orlistat functioning score, self-esteem, family activities and are available for adolescent use in most jurisdictions, change in health score), with no difference between the although use of metformin generally remains off label groups for three other subscores. Revisional procedures in those without diabetes mellitus. were required in 28% of the surgical group. The authors Some antidepressants have weight loss as a bene commented that although this incidence is within the fit but are not first-line obesity pharmacotherapies. In range of other studies, there might be a need, in adoles some countries, bupropion is used as an antidepressant cents, for additional education and supervision of eat and in others it is licensed only for addiction therapies. ing to ensure they eat small meals slowly. This approach When bupropion is combined with naltrexone as a slow- might help reduce the need for revision. This study was release preparation, there is benefit compared with pla the only one included in the 2015 Cochrane Review of cebo for weight loss85. The data for topiramate's effect bariatric surgery in adolescents94. on body weight in adults are positive but compromised A systematic review with broader inclusion cri by adverse effects such as dizziness and drowsiness86. teria than the 2015 Cochrane Review investigated

Carbohydrate Protein Fat Energy intake required surgical and nutritional complications, to determine for weight loss whether weight and cardiometabolic risk improvements are maintained well into adulthood and to identify which patients benefit most from surgery97. Increased-protein diet Clinical guidelines Consensus guidelines on bariatric surgery in adolescents have been established by various medical professional Low-carbohydrate diet bodies in several countries11,98–101. In terms of patient characteristics, the recommendations generally cover: a minimum age, usually in mid-adolescence; the patient to have reached sexual maturity; the presence of severe obe Conventional diet sity, usually BMI >35 kg/m2 with an obesity-associated complication, or BMI >40 kg/m2; the persistence of obe 020406080 100 sity despite involvement in a formal multidisciplinary Energy contribution from macronutrient (%) programme of lifestyle modification and pharmaco therapy; the patient being able to give informed consent; 39,42,43 Fig. 3 | Macronutrient distribution of common diet interventions . and the adolescent and family being willing to partici The percentage of energy contributions from carbohydrate, protein and fat can be pate actively in the treatment programme and agreeing manipulated to achieve an energy intake level for weight loss. The purple and green areas to ongoing review following the surgery. indicate overlap between the percentage energy contributions. Recommendations generally specify that the sur gery should be performed by a skilled bariatric surgeon prospective clinical trials and observational studies of affiliated with a team experienced in the medical and bariatric surgery in patients ≤18 years old published up behavioural management of adolescents with obesity. to early 2014 (ref.95). The review included 37 studies, The specific recommendations for the type of sur with 15 being prospective and 1 being the RCT noted in gery vary depending upon regional expertise but have the previous paragraph. The mean BMI loss after LAGB included LAGB, RYGB and LSG. Finally, in terms of was 11.6 kg/m2 (95% CI 9.8–13.4), compared with postoperative management and follow-up, recom 16.6 kg/m2 (95% CI 13.4–19.8) after Roux-en-Y gas mendations tend to specify the following: that more tric bypass (RYGB) and 14.1 kg/m2 (95% CI 10.8–17.5) frequent postoperative follow-up is generally required after laparoscopic sleeve gastrectomy (LSG). The review with adolescents than with adult patients; that there is authors noted that adverse effects and long-term com a need for ongoing care by a multidisciplinary team; plications, including nutritional deficiencies, were more that contraception be considered owing to improved frequent and serious after RYGB than the other two fertility following weight loss; that clinicians be aware procedures, although there were fewer data for LSG, a that adolescent patients, especially, might not adhere procedure that has gained prominence more recently to nutritional supplementation; that very long-term than RYGB or LAGB. All forms of surgery resulted in follow-up is required; and that the patient should be improvements in a range of obesity-associated com enrolled in a bariatric surgery registry. plications, such as dyslipidaemia and T2DM, and in One of the major challenges in many countries is quality of life. the need to ensure equitable access to such services for Subsequent to this systematic review, the 3-year affected adolescents, because, in most health systems, outcomes of the Teen-Longitudinal Assessment of bariatric surgery remains largely available only in the Bariatric Surgery (Teen-LABS) study were published. private sector, and bariatric surgery services for adoles The Teen-LABS study is a prospectively enrolled study cents are sparse25. Thus, specific strategies for the pro of adolescents undergoing bariatric surgery in five cen vision of bariatric surgery for adolescents in the public tres in the United States96. A total of 242 patients (75% sector are likely to be needed in most jurisdictions. female patients and 72% white patients) were enrolled, with 67% undergoing RYGB, 28% LSG and 6% LAGB. Treatment of complications At 3 years after surgery, there was an overall mean weight Many obesity-associated complications, such as dys loss of 27% (95% CI 25–29) and marked improvement lipidaemia, fatty liver disease and elevated blood pres or remission in T2DM, prediabetes, kidney disease, sure, will improve following weight loss. Nevertheless, elevated blood pressure and dyslipidaemia as well as depending on the type and severity of such complica improvement in quality-of-life measures. However, tions, it might be appropriate to manage these when the majority of patients had low iron concentrations identified and not to wait until weight loss has occurred, and 13% had undergone one or more additional intra- as weight loss might be difficult to achieve12. abdominal procedures. Such findings indicate that During adolescence, the prevalence of individual there are risks associated with bariatric surgery in this obesity-associated complications increases, as does the age group and highlight the importance of ensuring clustering of such complications. In addition, adoles that appropriate nutrient supplementation is provided cence sees the emergence of components of the met to adolescents after bariatric surgery. abolic syndrome and the development of sex-specific Further high-quality long-term studies of bariatric differences in risk factors for cardiovascular disease102. surgery in adolescents are required to monitor potential No long-term data are available in contemporary cohorts

Musculoskeletal Psychosocial Functional Cardiometabolic • Improved body composition • Improved cognition • Improved physical fitness • Improved insulin sensitivity • Increased caloric expenditure • Reduced depressive symptoms • Enhanced exercise tolerance • Improved blood pressure and and metabolic rate • Improved academic performance • Increased strength and endothelial function • Improved bone health • Better confidence and self-esteem proprioception • Reduced inflammation • Increased skeletal muscle mass • Improved lipid profile

Fig. 4 | Benefits of exercise in adolescents with obesity55–62. A range of elements contributes to the musculoskeletal, cardiometabolic, functional and psychosocial health improvements resulting from increased exercise.

that enable quantification of future combined or specific The most common obesity-associated complications risk of obesity-associated complications or the cost of in adolescence are primarily related to metabolic abnor that risk. malities and mental health concerns (Fig. 5). Here, we Multiple definitions of the metabolic syndrome in ado provide a brief overview of the management of the met lescence exist, all of which have agreed core elements — abolic complications associated with obesity. Readers obesity, dyslipidaemia, abnormal glucose metabo are referred to recent reviews or clinical guidelines for lism and hypertension — but these vary in the spe details of the management in the context of adolescent cifics of cut points and reference standards103–106. The obesity, obstructive sleep apnoea111, nonalcoholic fatty International Diabetes Federation definition, which uses liver disease112, slipped capital femoral epiphysis113, age-specific cut points for the core metabolic elements depression114 and anxiety115. listed above for 10–16 years of age (adult cut points for The International Society for Paediatric and those >16 years) and which specifies abdominal obe Adolescent Diabetes (ISPAD) Clinical Guideline sity as defined by waist measure, is probably the most acknowledges the limited evidence for diagnosis, man common in use106,107. Concerns about which definition agement and monitoring of prediabetes and T2DM in to use, or whether the metabolic syndrome is present adolescents116,117. The prevalence of T2DM varies with or not103–106, should never impede management of the ethnicity, with T2DM occurring in 6% of adolescents individual components. The prevalence of the met with diabetes mellitus in the United States and 40% of abolic syndrome rises over the decade of adolescence adolescents with diabetes mellitus in Asia and the Pacific and might be present in up to 50% of adolescents with Islands117. The prevalence of insulin resistance is less well obesity, depending on both the definition used and the characterized, but increasing age and weight, abdom population being studied103–106. inal distribution of obesity, the presence of acanthosis Medication doses used in the treatment of obesity- nigricans and a family history of T2DM should raise the associated complications will be closer to adult rec level of clinical suspicion118. Prediabetes might be tran ommendations given their body weight. An important sient, coinciding with the insulin resistance often seen in adolescent proviso is that some of these medications are puberty19, with continuing weight gain predicting pro not considered safe in pregnancy, and countries vary in gression to T2DM. Other than metformin, drugs used their labelling of safety recommendations. Many drugs in T2DM usually require off-label use in those under recommended for the treatment of metabolic morbid 18 years of age. The rising community prevalence of ity in obesity have not been trialled in the reproductive obesity and the tighter target range for diabetes mellitus age group; studies of safety in pregnancy will never be control mean that more than one-third of adolescents undertaken and translating animal exposure results to with confirmed type 1 diabetes mellitus might also have humans is not always reliable. Given that these newer overweight or obesity119, and metformin could be added medications remain the optimal choice for treatment to their therapy to address insulin resistance. of metabolic complications of obesity, a more nuanced In a study published in 2015 that examined trends approach than previously used is being advocated in dyslipidaemia and hypertension in youth, 20.2% had by the FDA, which introduced a new labelling system, adverse levels of total and HDL cholesterol and 11.0% the PLLR (Pregnancy and Lactation Labelling)108, pub had elevated or borderline blood pressure recordings120. lished in 2016. The European Medicines Agency is due The US National Lipid Association121 provides an to release its updated pregnancy guidelines for consul annual update of lipid-lowering therapies, hyperlinks to tation in 2018 (ref.109). Table 1 shows data from the cate resources and separate recommendations for children, gorization system of the Australian Therapeutic Goods adolescents and adults <21 years of age. The common Administration (TGA), which provides currently acces dyslipidaemias of adolescent obesity are low levels of sible and regularly updated110 ratings for some common HDL cholesterol, in which the effective treatments are medications used in the management of obesity and weight loss and physical activity, and elevated levels of its complications. triglycerides. The risk of pancreatitis is increased with

Table 1 | Medications for the management of obesity and its complications name indication comments Safety in pregnancy1 Orlistat Obesity Inhibits gut lipase with resulting steatorrhoea with high-fat B1 meals — easy to manipulate Phentermine Obesity Sympathomimetic amine that has addictive potential. B3 Rejected by some agencies in its combination with topiramate owing to mental health and cardiovascular system concerns Topiramate Obesity If epilepsy or severe migraine are present, this fulfils D prescribing indications. If prescribed off-label, cost might be prohibitive GLP1R agonists Obesity Mostly by daily injection. Nausea and vomiting are B3 significant adverse effects Metformin Type 2 diabetes mellitus, prediabetes Gastrointestinal upset if not up-titrated slowly. Induces Approved for use (C) and insulin resistance, polycystic ovary ovulation; consider contraception needs. Monitor vitamin 128 syndrome, acanthosis nigricans and B12 levels adjunctive treatment for weight loss Statins Hypercholesterolaemia (LDL cholesterol) No data on long-term adverse effects, including myopathy Teratogenic in and increased risk of type 2 diabetes mellitus animals, but risk appears lower in humans (D) ACE inhibitors and Hypertension and diabetic nephropathy Single daily dose and low adverse effect profile improve Might increase ARBs compliance malformation risk and pregnancy loss in second and third trimesters (D) SSRIs Depression and anxiety in which talking Package warnings that these antidepressants may increase C therapy is not successful risk of suicidal ideation and behaviours COCP Polycystic ovary syndrome: lower COCP-induced weight gain is disputed129. Contraindicated B3 oestrogen doses and newer gestagens in paralytic migraine or thromboembolism have fewer adverse effects Australian categories for prescribing medicines in pregnancy: the Australian Therapeutic Goods Administration (TGA) system is currently accessible and regularly updated110, providing ratings from A (the safest drugs to take during pregnancy ; no known adverse reactions) to D (adverse reactions have been found in humans)110. The TGA system provides guidance that aims to assist both clinicians and the women who use the medications and is reliable evidence-based information for patients who use the Internet for health advice. Health professionals should refer to the TGA website for further information (Australian Government Department of Health). Practitioners should also be aware of local approvals and prescribing rules for medications in pregnancy. ACE, angiotensin-converting enzyme; ARBs, angiotensin receptor blockers; COCP, combined oral contraceptive pill; GLP1R , glucagon-like peptide 1 receptor ; SSRIs, selective serotonin reuptake inhibitors.

concentrations of triglycerides >5 mmol/l, and if life insulin resistance at the receptor level is not present in style change is ineffective and aggravating factors are the ovary, and high insulin concentrations drive ovar excluded, omega-3 fatty acids, niacin or fibrates should ian androgen production and subsequent hirsutism and be trialled122. menstrual cycle disturbances. Metformin prescribed Elevated blood pressure is defined as blood pres together with the combined oral contraceptive pill might sure (adjusted for age, sex and height) >90th percen ameliorate any weight gain associated with the use of tile to <95th percentile or 120/80 mmHg (whichever combined oral contraceptives125. is lower), stage 1 hypertension is blood pressure >95th percentile to <95th percentile + 12 mmHg, or 130/80 to Access to health care 139/89 mmHg (whichever is lower) and stage 2 hyper One of the six major recommendations of the 2016 tension is blood pressure >95th percentile + 12 mmHg, Report of the WHO Ending Childhood Obesity or >140/90 mmHg (whichever is lower)123. Pharmaco Commission was the provision of weight management therapy should not be withheld if weight loss does services for both children and young people as part of not occur. The updated 2017 American Academy of Universal Health Coverage126. Given that health sys Pediatrics guidelines123 provide a management guide tems vary considerably between countries and that and discussion around drug choices. there is little published on obesity models of care in The Endocrine Society's Clinical Guideline for pol young people, there will be a range of challenges ycystic ovary syndrome124 considers adolescents sepa in the delivery of treatment services for this group. rately. The diagnosis should be based on clinical and/or Importantly, the report highlighted the potential for biochemical hyperandrogenism in the presence of per weight bias by health professionals and hence the need sistent oligomenorrhoea and should not rely on multi for strategies to avoid stigmatization of young people cystic ovarian change, which might be developmentally with obesity. normal. Although obesity and insulin resistance are Adolescent obesity can be managed in paediatric or present in most patients with polycystic ovary syndrome, adult care teams. Neither of these approaches is likely to

Benign intracranial hypertension Affective disorders, depression and anxiety Patients without visual defects are Depression and anxiety increase in prevalence in treated with acetazolamide, a carbonic adolescence, with a female preponderance. anhydrase inhibitor. Treatment in adolescents with obesity is along standard lines. Fluoxetine is less likely to be Skin associated with weight gain than the newer Skin tags might require physical removal. For antipsychotics, which are associated with weight intertrigo, keep dry and use topical antifungals. gain and insulin resistance, and should be avoided. Metformin seems to be effective for acanthosis nigricans, but much of the evidence is case-based. Clinically, this effect might be positive for Obstructive sleep apnoea The American Academy of Pediatrics adolescents and metformin might assist in Clinical Guidelines give the evidence for reduction in BMI in combined obesity treatment weight management in obstructive sleep regimens. apnoea as grade C. CPAP therapy should be initiated if symptoms and/or signs of Lipids and other cardiovascular objective evidence of obstructive sleep risk factors Statins are suggested for nonfamilial if adenotonsillectomy is not performed hypercholesterolaemia if LDL (grade B). Intranasal glucocorticoids might cholesterol is >4.1 mmol/l and if also assist as adjuvant therapy. additional risk factors are present, such as obesity, hypertension and diabetes mellitus. Concerns regarding long-term Hypertension safety exist, and data in younger Preferred agents are thiazide diuretics patients with secondary lipid disorders and/or ACE inhibitors or angiotensin are lacking. receptor blockers. Once-a-day dosing and a low adverse effect profile make these drugs particularly useful in Fatty liver adolescents. Weight loss is the only proven therapy. Insulin sensitizers and vitamin E have been used. Insulin resistance, prediabetes and diabetes mellitus Lifestyle intervention is essential in all Polycystic ovary syndrome three conditions. Metformin also has a Metformin will produce a more regular role in all three conditions. Metformin is menstrual cycle and induce ovulation, a first-line therapy for type 2 diabetes of which the adolescent should be mellitus in adolescents, but medical warned. Weight loss, physical hair instability at presentation might require removal, and the combined oral insulin. Continuation of insulin use contraceptive pill to suppress ovarian should be reviewed at regular intervals androgen production are all standard as weight gain, accompanied by therapies. Weight gain induced by the worsening insulin resistance, is an combined oral contraceptive pill might undesirable effect of long-term therapy. occur in susceptible individuals. Slipped capital femoral epiphysis High index of suspicion if pain and/or limp are present. Orthopaedic intervention. Strength and balance training might be useful. Pes planus Supportive shoes and orthotics.

Fig. 5 | Management of comorbidities122,129–134. Obesity-associated complications are multisystemic, and the summarized treatment recommendations should be considered when the complication is identified in conjunction with weight-loss advice. ACE, angiotensin-converting enzyme; CPAP, continuous positive airway pressure.

be optimal for adolescents, in whom health care is best challenges in conducting traditional RCTs at this devel served by clinicians with expertise in treating adoles opmental stage and in this condition and because prag cents and young adults127. Supported transitional care matic trials offer information that informs the real-world from paediatric to adult care has not been reported in implementation of therapies. Additionally, improved the literature128, which is of concern given the preva data are needed on how best to manage young people lence of obesity and the need for ongoing management with obesity who are from disadvantaged environments to avoid further weight gain in young adulthood. and those from low-income and middle-income coun tries. Finally, it is vital that obesity in young people is Conclusions not dismissed as being too difficult to be addressed The importance of effective management of adolescent by clinicians. Instead, there should be a recognition of obesity is underscored by its high prevalence in many the WHO concept of adolescence as “a key decade in the countries and the short-term and long-term impacts life-course” where there is “a second chance in the sec of the condition. Although the evidence base for the ond decade” to improve health outcomes and provide management of adolescent obesity is growing, more high-quality clinical services1. pragmatic clinical trials in young people are needed. Published online xx xx xxxx These trials are needed because of the acknowledged

1. World Health Organization. Health for the world’s 26. Berger, S. E., Huggins, G. S., McCaffery, J. M. & considerations for dietetic practice. Eur. J. Clin. Nutr. adolescents: a second chance in the second decade. Lichtenstein, A. H. Comparison among criteria to 71, 646–651 (2017). WHO http://www.who.int/maternal_child_adolescent/ define successful weight-loss maintainers and 48. Mulholland, Y., Nicokavoura, E., Broom, J. & Rolland, documents/second-decade/en/ (2014). regainers in the action for health in diabetes (Look C. Very-low-energy diets and morbidity: a systematic 2. Ng, M. et al. Global, regional, and national prevalence AHEAD) and diabetes prevention program trials. review of longer-term evidence. Br. J. Nutr. 108, of overweight and obesity in children and adults Am. J. Clin. Nutr. 106, 1337–1346 (2017). 832–851 (2012). during 1980–2013: a systematic analysis for the 27. MacLean, P. S. et al. NIH working group report: 49. Berkowitz, R. I. et al. Meal replacements in the global burden of disease study 2013. Lancet 384, innovative research to improve maintenance of weight treatment of adolescent obesity: a randomized 766–781 (2014). loss. Obesity 23, 7–15 (2015). controlled trial. Obesity (Silver Spring) 19, 3. NCD Risk Factor Collaboration (NCD-RisC). Worldwide 28. Rajjo, T. et al. The association of weight loss and 1193–1199 (2011). trends in body-mass index, underweight, overweight, cardiometabolic outcomes in obese children: 50. Gow, M. L., Baur, L. A., Johnson, N. A., Cowell, C. T. & and obesity from 1975 to 2016: a pooled analysis of systematic review and meta-regression. J. Clin. Garnett, S. P. Reversal of type 2 diabetes in youth who 2416 population-based measurement studies in Endocrinol. Metab. 102, 758–762 (2016). adhere to a very-low-energy diet: a pilot study. 128.9 million children, adolescents, and adults. 29. Ho, M. et al. Impact of dietary and exercise Diabetologia 60, 406–415 (2016). Lancet 390, 2627–2642 (2017). interventions on weight change and metabolic 51. Willi, S. M., Martin, K., Datko, F. M. & Brant, B. P. 4. Chung, A. et al. Trends in child and adolescent obesity outcomes in obese children and adolescents: a Treatment of type 2 diabetes in childhood using a prevalence in economically advanced countries systematic review and meta-analysis of randomized very-low-calorie diet. Diabetes Care 27, 348–353 according to socioeconomic position: a systematic trials. JAMA Pediatr. 167, 759–768 (2013). (2004). review. Obes. Rev. 17, 276–295 (2016). 30. Ruotsalainen, H., Kyngäs, H., Tammelin, T. & 52. Moldovan, A. R. & David, D. Effect of obesity 5. Skinner, A. C. & Skelton, J. A. Prevalence and trends Kääriäinen, M. Systematic review of physical activity treatments on eating behavior: psychosocial in obesity and severe obesity among children in the and exercise interventions on body mass indices, interventions versus surgical interventions. a United States, 1999–2012. JAMA Pediatr. 168, subsequent physical activity and psychological systematic review. Eat Behav. 12, 161–167 (2011). 561–566 (2014). symptoms in overweight and obese adolescents. 53. Webber, L., Hill, C., Saxton, J., Van Jaarsveld, C. & 6. Garnett, S. P., Baur, L. A., Jones, A. M. & Hardy, L. L. J. Adv. Nurs. 71, 2461–2477 (2015). Wardle, J. Eating behaviour and weight in children. Trends in the prevalence of morbid and severe obesity 31. Altman, M. & Wilfley, D. E. Evidence update on the Int. J. Obes. 33, 21 (2009). in Australian children aged 7–15 years, 1985–2012. treatment of overweight and obesity in children and 54. Ho, M. et al. Effect of a prescriptive dietary PLoS ONE 11, e0154879 (2016). adolescents. J. Clin. Child Adolesc. Psychol. 44, intervention on psychological dimensions of eating 7. Skinner, A. C., Perrin, E. M. & Skelton, J. A. 521–537 (2015). behavior in obese adolescents. Int. J. Behav. Nutr. Prevalence of obesity and severe obesity in US 32. Webb, V. L. & Wadden, T. A. Intensive lifestyle Phys. Act 10, 119 (2013). children, 1999–2014. Obesity 24, 1116–1123 intervention for obesity: principles, practices, and 55. O´Malley, G. et al. Physical activity and physical fitness (2016). results. Gastroenterology 152, 1752–1764 (2017). in pediatric obesity: what are the first steps for 8. Gordon-Larsen, P., Adair, L. S., Nelson, M. C. & 33. Oude Luttikhuis, H. et al. Interventions for treating clinicians? Expert conclusion from the 2016 ECOG Popkin, B. M. Five-year obesity incidence in the obesity in children. Cochrane Database Syst. Rev. 21, workshop. Int. J. Exerc. Sci. 10, 487–496 (2017). transition period between adolescence and adulthood: CD001872 (2009). 56. Ekelund, U. et al. Independent associations of physical the national longitudinal study of adolescent health. 34. Samdal, G. B., Eide, G. E., Barth, T., Williams, G. & activity and cardiorespiratory fitness with metabolic Am. J. Clin. Nutr. 80, 569–575 (2004). Meland, E. Effective behaviour change techniques for risk factors in children: the European youth heart 9. Patton, G. C. et al. Overweight and obesity between physical activity and healthy eating in overweight and study. Diabetologia 50, 1832–1840 (2007). adolescence and young adulthood: a 10-year obese adults; systematic review and meta-regression 57. Janssen, I. & LeBlanc, A. G. Systematic review of the prospective cohort study. J. Adolesc. Health 48, analyses. Int. J. Behav. Nutr. Phys. Act. 14, 42 (2017). health benefits of physical activity and fitness in 275–280 (2011). 35. Gow, M. L., Garnett, S. P., Baur, L. A. & Lister, N. B. school-aged children and youth. Int. J. Behav. Nutr. 10. Baur, L. A., Hazelton, B. & Shrewsbury, V. A. The effectiveness of different diet strategies to reduce Phys. Act. 7, 40 (2010). Assessment and management of obesity in childhood type 2 diabetes risk in youth. Nutrients 8, 486 58. Lubans, D. et al. Physical activity for cognitive and and adolescence. Nat. Rev. Gastroenterol. Hepatol. 8, (2016). mental health in youth: a systematic review of 635–645 (2011). 36. Díaz, R. G., Esparza-Romero, J., Moya-Camarena, S. Y., mechanisms. Pediatrics 33, e20161642 (2016). 11. National Health and Medical Research Council. Robles-Sardín, A. E. & Valencia, M. E. Lifestyle 59. Raitakan, O. T. et al. Effects of persistent physical Clinical practice guidelines for the management of intervention in primary care settings improves obesity activity and inactivity on coronary risk factors in overweight and obesity in adults, adolescents and parameters among Mexican youth. J. Am. Diet Assoc. children and young adults the cardiovascular risk in children in Australia. NHMRC www.nhmrc.gov.au/ 110, 285–290 (2010). young Finns study. Am. J. Epidemiol. 140, 195–205 guidelines/publications/n57 (2013). 37. Epstein, L. H., Wing, R. R., Koeske, R. & Valoski, A. (1994). 12. National Clinical Guideline Centre (UK). Obesity: Effects of diet plus exercise on weight change in 60. Centers for Disease Control and Prevention. identification, assessment and management. NICE parents and children. J. Consult. Clin. Psychol. 52, The association between school-based physical www.nice.org.uk/guidance/cg189 (2014). 429 (1984). activity, including physical education, and academic 13. Parkin, P. et al. Recommendations for growth 38. Jiang, J. X., Xia, X., Greiner, T., Lian, G. & Rosenqvist, performance. CDC. https://www.cdc.gov/healthyyouth/ monitoring, and prevention and management of U. A two year family based behaviour treatment for health_and_academics/pdf/pa-pe_paper.pdf (2010). overweight and obesity in children and youth in obese children. Arch. Dis. Child. 90, 1235–1238 61. Paes, S. T., Marins, J. C. B. & Andreazzi, A. E. primary care. CMAJ 187, 411–421 (2015). (2005). Metabolic effects of exercise on childhood obesity: a 14. Styne, D. M. et al. Pediatric obesity — assessment, 39. Gow, M. L. et al. Impact of dietary macronutrient current view. Rev. Paul Pediatr. 33, 122–129 (2015). treatment, and prevention: an endocrine society distribution on BMI and cardiometabolic outcomes in 62. Gow, M. L. et al. Sustained improvements in fitness clinical practice guideline. J. Clin. Endocrinol. Metab. overweight and obese children and adolescents: a and exercise tolerance in obese adolescents after a 12 102, 709–757 (2017). systematic review. Nutr. Rev. 72, 453–470 (2014). week exercise intervention. Obes. Res. Clin. Pract. 10, 15. Blakemore, S.-J. Imaging brain development: the 40. Demol, S. et al. Low-carbohydrate (low and high-fat) 178–188 (2016). adolescent brain. Neuroimage 61, 397–406 (2012). versus high-carbohydrate low-fat diets in the 63. Australian Government Department of Health. 16. Shulman, E. P. et al. The dual systems model: review, treatment of obesity in adolescents. Acta Paediatr. 98, Australia’s physical activity and sedentary behaviour reappraisal, and reaffirmation. Dev. Cogn. Neurosci. 346–351 (2009). guidelines. Australian Government Department of 17, 103–117 (2016). 41. Kirk, S. et al. Role of carbohydrate modification in Health. http://www.health.gov.au/internet/main/ 17. Al-Khudairy, L. et al. Diet, physical activity and weight management among obese children: a publishing.nsf/content/health-pubhlth-strateg- behavioural interventions for the treatment of randomized clinical trial. J. Pediatr. 161, 320–327 phys-act-guidelines (2014). overweight or obese adolescents aged 12 to 17 years. (2012). 64. Department of Health (UK). UK physical activity Cochrane Database Syst. Rev. 6, CD012691 (2017). 42. Westerterp-Plantenga, M. S., Nieuwenhuizen, A., guidelines. GOV.UK https://www.gov.uk/government/ 18. Ho, M. et al. Effectiveness of lifestyle interventions in Tome, D., Soenen, S. & Westerterp, K. R. Dietary publications/uk-physical-activity-guidelines (2011). child obesity: systematic review with meta-analysis. protein, weight loss, and weight maintenance. Annu. 65. The Public Health Agency of Canada. 24 h movement Pediatrics 130, e1647–e1671 (2012). Rev. Nutr. 29, 21–41 (2009). guidelines for children and youth: an integration of 19. Patton, G. C. & Viner, R. Pubertal transitions in health. 43. Papadaki, A. et al. The effect of protein and glycemic physical activity, sedentary behaviour, and sleep. Lancet 369, 1130–1139 (2007). index on children’s body composition: the DiOGenes Government of Canada. https://www.canada.ca/en/ 20. Moran, A. et al. Insulin resistance during puberty: randomized study. Pediatrics 126, E1143–E1152 public-health/services/health-promotion/healthy-living/ results from clamp studies in 357 children. Diabetes (2010). physical-activity/24-hour-movement-guidelines- 48, 2039–2044 (1999). 44. Schwingshackl, L., Hobl, L. P. & Hoffmann, G. Effects children-youth.html (2016). 21. Kelsey, M. M. & Zeitler, P. S. Insulin resistance of of low glycaemic index/low glycaemic load versus high 66. Office of Disease Prevention and Health Promotion. puberty. Curr. Diab. Rep. 16, 64 (2016). glycaemic index/ high glycaemic load diets on Physical Activity Guidelines for Americans. health.gov 22. Pont, S. J., Puhl, R., Cook, S. R. & Slusser, W. Stigma overweight/obesity and associated risk factors in https://health.gov/paguidelines/guidelines/ (2008). experienced by children and adolescents with obesity. children and adolescents: a systematic review and 67. Stearns, J. A., Carson, V., Spence, J. C., Faulkner, G. & Pediatrics 140, e20173034 (2017). meta-analysis. Nutr. J. 14, 87 (2015). Leatherdale, S. T. The role of peer victimization in the 23. Puhl, R. M. & Latner, J. D. Stigma, obesity, and the 45. Seimon, R. V. et al. Do intermittent diets provide physical activity and screen time of adolescents: a health of the nation’s children. Psychol. Bull. 133, physiological benefits over continuous diets for weight cross-sectional study. BMC Pediatr. 17, 170 (2017). 557 (2007). loss? a systematic review of clinical trials. Mol. Cell. 68. Tsiros, M. et al. Day-to-day physical functioning and 24. Tang-Péronard, J. & Heitmann, B. Stigmatization of Endocrinol. 418, 153–172 (2015). disability in obese 10-to 13-year-olds. Pediatr. Obes. obese children and adolescents, the importance of 46. Headland, M., Clifton, P. M., Carter, S. & Keogh, J. B. 8, 31–41 (2013). gender. Obes. Rev. 9, 522–534 (2008). Weight-loss outcomes: a systematic review and meta- 69. Tsiros, M. D., Coates, A. M., Howe, P., Grimshaw, P. & 25. Dietz, W. H. et al. Management of obesity: analysis of intermittent energy restriction trials lasting Buckley, J. D. Obesity: the new childhood disability? improvement of health-care training and systems for a minimum of 6 months. Nutrients 8, 354 (2016). Obes. Rev. 12, 26–36 (2011). prevention and care. Lancet 385, 2521–2533 47. Lister, N. et al. Nutritional adequacy of diets for 70. Fedele, D. A., Cushing, C. C., Fritz, A., Amaro, C. M. & (2015). adolescents with overweight and obesity: Ortega, A. Mobile health interventions for improving

health outcomes in youth: a meta-analysis. JAMA 95. Paulus, G. F. et al. Bariatric surgery in morbidly obese 121. Bays, H. E., Jones, P. H., Orringer, C. E., Brown, W. V. Pediatr. 171, 461–469 (2017). adolescents: a systematic review and meta-analysis. & Jacobson, T. A. National lipid association annual 71. Cappuccio, F. P. et al. Meta-analysis of short sleep Obes. Surg. 25, 860–878 (2015). summary of clinical lipidology 2016. J. Clin. Lipidol. duration and obesity in children and adults. Sleep 31, 96. Inge, T. H. et al. Weight loss and health status 3 years 10, S1–S43 (2016). 619–626 (2008). after bariatric surgery in adolescents. N. Engl. J. Med. 122. Oh, R. C. & Lanier, J. B. Management of 72. Fatima, Y. & Mamun, A. Longitudinal impact of sleep 374, 113–123 (2016). hypertriglyceridemia. Am. Fam. Physician. 75, on overweight and obesity in children and adolescents: 97. Beamish, A. J. & Reinehr, T. Should bariatric surgery 1365–1371 (2007). a systematic review and bias-adjusted meta-analysis. be performed in adolescents? Eur. J. Endocrinol. 176, 123. Flynn, J. T. et al. Clinical practice guideline for screening Obes. Rev. 16, 137–149 (2015). D1–D15 (2017). and management of high blood pressure in children 73. Thomson, C. A. et al. Relationship between sleep 98. Inge, T. H. et al. Bariatric surgery for severely and adolescents. Pediatrics 140, e20171904 quality and quantity and weight loss in women overweight adolescents: concerns and (2017). participating in a weight-loss intervention trial. recommendations. Pediatrics 114, 217–223 (2004). 124. Legro, R. S. et al. Diagnosis and treatment of Obesity 20, 1419–1425 (2012). 99. International Pediatric Endosurgery Group. IPEG polycystic ovary syndrome: an endocrine society 74. Chaput, J.-P. & Tremblay, A. Sleeping habits predict the guidelines for surgical treatment of extremely obese clinical practice guideline. J. Clin. Endocrinol. Metab. magnitude of fat loss in adults exposed to moderate adolescents. J Laparoendosc. Adv. Surg. Tech. 19 98, 4565–4592 (2013). caloric restriction. Obes. Facts 5, 561–566 (2012). (Suppl. 1), xiv–xvi (2009). 125. Glintborg, D. et al. Body composition is improved 75. Capers, P. L., Fobian, A. D., Kaiser, K. A., Borah, R. & 100. Baur, L. A. & Fitzgerald, D. A. Recommendations for during 12 months’ treatment with metformin alone or Allison, D. B. A systematic review and meta-analysis of bariatric surgery in adolescents in Australia and New combined with oral contraceptives compared with randomized controlled trials of the impact of sleep Zealand. J. Paediatr. Child Health 46, 704–707 (2010). treatment with oral contraceptives in polycystic ovary duration on adiposity and components of energy 101. Michalsky, M., Reichard, K., Inge, T., Pratt, J. & syndrome. J. Clin. Endocrinol. Metab. 99, 2584–2591 balance. Obes. Rev. 16, 771–782 (2015). Lenders, C. ASMBS pediatric committee best practice (2014). 76. Yoong, S. L. et al. Systematic review and meta-analysis guidelines. Surg. Obes. Relat. Dis. 8, 1–7 (2012). 126. World Health Organization. Report of the commission on of interventions targeting sleep and their impact on 102. Pulgarón, E. R. Childhood obesity: a review of ending childhood obesity. WHO http://www.who.int/end- child body mass index, diet, and physical activity. increased risk for physical and psychological childhood-obesity/publications/echo-report/en/ (2016). Obesity 24, 1140–1147 (2016). comorbidities. Clin. Ther. 35, A18–A32 (2013). 127. Steinbeck, K., Towns, S. & Bennett, D. Adolescent and 77. Chaput, J.-P. & Dutil, C. Lack of sleep as a contributor 103. Agirbasli, M., Tanrikulu, A. M. & Berenson, G. S. young adult medicine is a special and specific area of to obesity in adolescents: impacts on eating and Metabolic syndrome: bridging the gap from childhood medical practice. J. Paediatr. Child Health 50, activity behaviors. Int. J. Behav. Nutr. Phys. Act 13, to adulthood. Cardiovasc. Ther. 34, 30–36 (2016). 427–431 (2014). 103 (2016). 104. Cook, S., Weitzman, M., Auinger, P., Nguyen, M. & 128. Shrewsbury, V., Baur, L., Nguyen, B. & Steinbeck, K. 78. Carroll, J. K. et al. Who uses mobile phone health Dietz, W. H. Prevalence of a metabolic syndrome Transition to adult care in adolescent obesity: a apps and does use matter? a secondary data analytics phenotype in adolescents: findings from the third systematic review and why it is a neglected topic. Int. approach. J. Med. Internet Res. 19, e125 (2017). national health and nutrition examination survey, J. Obes. 38, 475 (2014). 79. Jakicic, J. M. et al. Effect of wearable technology 1988–1994. Arch. Pediatr. Adolesc. Med. 157, 129. Chapman, L., Darling, A. & Brown, J. Association combined with a lifestyle intervention on long-term 821–827 (2003). between metformin and vitamin B 12 deficiency in weight loss: the IDEA randomized clinical trial. JAMA 105. Ford, E. S. & Li, C. Defining the metabolic syndrome in patients with type 2 diabetes: a systematic review and 316, 1161–1171 (2016). children and adolescents: will the real definition please meta-analysis. Diabetes Metab. 42, 316–327 80. van der Zanden, R., Kramer, J., Gerrits, R. & Cuijpers, P. stand up? J. Pediatr. 152, 160–164 (2008). (2016). Effectiveness of an online group course for depression 106. Reinehr, T. Metabolic syndrome in children and 130. Gallo, M. F. et al. Combination contraceptives: effects in adolescents and young adults: a randomized trial. adolescents: a critical approach considering the on weight. Cochrane Database Syst. Rev. 25, J. Med. Internet Res. 14, e86 (2012). interaction between pubertal stage and insulin CD003987 (2014). 81. Mead, E. et al. Diet, physical activity and behavioural resistance. Curr. Diab. Rep. 16, 8 (20 16 ). 131. Romo, A. & Benavides, S. Treatment options in insulin interventions for the treatment of overweight or obese 107. Zimmet, P. et al. The metabolic syndrome in children resistance obesity–related acanthosis nigricans. Ann. children from the age of 6 to 11 years. Cochrane and adolescents–an IDF consensus report. Pediatr. Pharmacother. 42, 1090–1094 (2008). Database Syst. Rev. 6, CD012651 (2017). Diabetes 8, 299–306 (2007). 132. McDonagh, M. S., Selph, S., Ozpinar, A. & Foley, C. 82. Wiegand, S. et al. Predicting weight loss and 108. Pernia, S. & DeMaagd, G. The new pregnancy and Systematic review of the benefits and risks of maintenance in overweight/obese pediatric patients. lactation labeling rule. P. T. 41, 713 (2016). metformin in treating obesity in children aged 18 Horm. Res. Paediatr. 82, 380–387 (2014). 109. European Medicines Agency. Good pharmacovigilance years and younger. JAMA Pediatr. 168, 178–184 83. Gow, M. L. et al. Can early weight loss, eating practices. EMA http://www.ema.europa.eu/ema/index. (2014). behaviors and socioeconomic factors predict jsp?curl=pages/regulation/document_listing/document_ 133. Marcus, C. L. et al. Diagnosis and management of successful weight loss at 12- and 24-months in listing_000345.jsp&mid=WC0b01ac058058f32c childhood obstructive sleep apnea syndrome. adolescents with obesity and insulin resistance (2017). Pediatrics 130, e714–e755 (2012). participating in a randomised controlled trial? 110. Therapeutic Goods Administration (Australia). 134. National Institute for Health Care and Excellence. Int. J. Behav. Nutr. Phys. Act. 13, 1–11 (2016). Prescribing medicines in pregnancy database. TGA Depression in children and young people: identification 84. Mead, E. et al. Drug interventions for the treatment of https://www.tga.gov.au/australian-c ategorisation-system- and management: clinical guideline [CG28]. NICE. obesity in children and adolescents. Cochrane prescribing-medicines-pregnancy (2017). https://www.nice.org.uk/guidance/cg28 (2017). Database Syst. Res. 11, CD012436 (2016). 111. Andersen, I. G., Holm, J. C. & Homøe, P. Obstructive 85. Van Gaal, L. & Scheen, A. Weight management in type sleep apnea in obese children and adolescents, Author contributions 2 diabetes: current and emerging approaches to treatment methods and outcome of treatment–a All authors contributed to researching data for the article, treatment. Diabetes Care 38, 1161–1172 (2015). systematic review. Int. J. Pediatr. Otorhinolaryngol. discussion of content, writing the article and reviewing and/ 86. Paravattil, B., Wilby, K. J. & Turgeon, R. Topiramate 87, 190–197 (2016). or editing the manuscript before submission. monotherapy for weight reduction in patients with type 112. Doycheva, I., Watt, K. D. & Alkhouri, N. Nonalcoholic 2 diabetes mellitus: a systematic review and meta- fatty liver disease in adolescents and young adults: the Competing interests analysis. Diabetes Res. Clin. Pract. 114, 9–14 (2016). next frontier in the epidemic. Hepatology 65, The authors declare no competing interests. 87. Fox, C. K. et al. Meal replacements followed by 2100–2109 (2017). topiramate for the treatment of adolescent severe 113. Bitersohl, B., Hosalkar, H. S., Zilkens, C. & Krauspe, R. Publisher’s note obesity: a pilot randomized controlled trial. Obesity Current concepts in management of slipped capital Springer Nature remains neutral with regard to jurisdictional 24, 2553–2561 (2016). femoral epiphysis. Hip. Int. 25, 104–114 (2015). claims in published maps and institutional affiliations. 88. Joao, A., Reis, F. & Fernandes, R. The incretin system 114. Lewandowski, R. E. et al. Evidence for the ABCs in obesity and diabetes–novel therapeutic management of adolescent depression. Pediatrics Review criteria strategies for weight loss and beyond. Obes. Rev. 17, 132, e996–e1009 (2013). This narrative review is based upon data from systematic 553–572 (2016). 115. Wehry, A. M., Beesdo-Baum, K., Hennelly, M. M., reviews, review papers and individual studies known to the 89. Tong, J. & Sandoval, D. A. Is the GLP-1 system a viable Connolly, S. D. & Strawn, J. R. Assessment and authors. Other studies were identified by a MEDLINE search therapeutic target for weight reduction? Rev. Endocr. treatment of anxiety disorders in children and of English language papers published up to 2017 using the Metab. Disord. 12, 187–195 (2011). adolescents. Curr. Psychiatry Rep. 17, 52 search terms “obesity” and “adolescent” and then further 90. Davies, M. et al. Effect of oral semaglutide compared (2015). refined using such terms as “bariatric surgery”, “behaviour with placebo and subcutaneous semaglutide on glycemic 116. Zeitler, P. et al. Type 2 diabetes in the child and therapy”, “drug therapy”, “diet”, “exercise”, “physical activity”, control in patients with type 2 diabetes: a randomized adolescent. Pediatr. Diabetes 15, 26–46 (2014). “weight maintenance” or “sleep”. Where appropriate, reviews clinical trial. JAMA 318, 1460–1470 (2017). 117. International Society for Pediatric and Adolescent or studies of obesity treatment involving children or adults 91. Kelly, A. S. et al. The effect of glucagon-like peptide-1 Diabetes. ISPAD clinical practice consensus were included in the Review. Reference lists of key papers receptor agonist therapy on body mass index in guidelines 2014. ISPAD http://www.ispad. were also checked to identify additional papers of interest. adolescents with severe obesity: a randomized, org/?page=ISPADClinicalPract (2014). We also searched for clinical guidelines on the management placebo-controlled, clinical trial. JAMA Pediatr. 167, 118. Caprio, S., Perry, R. & Kursawe, R. Adolescent obesity of adolescent obesity. For the latter to be considered, the 355–360 (2013). and insulin resistance: roles of ectopic fat guidelines had to be developed or endorsed by a national or 92. Vidal, J., Corcelles, R., Jiménez, A., Flores, L. & accumulation and adipose inflammation. international health professional body or government agency Lacy, A. M. Metabolic and bariatric surgery for Gastroenterology 152, 1638–1646 (2017). and available in English. obesity. Gastroenterology 152, 1780–1790 (2017). 119. van Vliet, M. et al. Overweight is highly prevalent in 93. O’Brien, P. E. et al. Laparoscopic adjustable gastric children with type 1 diabetes and associates with Related links banding in severely obese adolescents: a randomized cardiometabolic risk. J. Pediatr. 156, 923–929 (2010). Australian Government Department of Health: https://www. trial. JAMA 303, 519–526 (2010). 120. Kit, B. K. et al. Prevalence of and trends in tga.gov.au/australian-categorisation-system- 94. Ells, L. J. et al. Surgery for the treatment of obesity in dyslipidemia and blood pressure among US children prescribing-medicines-pregnancy children and adolescents. Cochrane Database Syst. and adolescents, 1999–2012. JAMA Pediatr. 169, CC-BY-4.0: https://creativecommons.org/licenses/by/4.0/ Res. 6, CD011740 (2015). 272–279 (2015).