REVIEW ARTICLE

Effectiveness of Lifestyle Interventions in Child Obesity: Systematic Review With Meta-analysis

AUTHORS: Mandy Ho, MSc, APD, RN,a,b Sarah P. Garnett, PhD, RNutr, APD,a,b,c Louise Baur, MBBS, PhD, FRACP,a,c abstract d Tracy Burrows, PhD, AdvAPD, Laura Stewart, PhD, RD, BACKGROUND AND OBJECTIVES: The effects of lifestyle interventions RNutr,e,f Melinda Neve, PhD, APD,d and Clare Collins, PhD, FDAAd on cardio-metabolic outcomes in overweight children have not been reviewed systematically. The objective of the study was to examine the aThe Children’s Hospital at Westmead Clinical School, University of Sydney, Sydney, Australia; bInstitute of Endocrinology and impact of lifestyle interventions incorporating a dietary component on Diabetes, and cKids Research Institute, The Children’s Hospital at both weight change and cardio-metabolic risks in overweight/obese Westmead, Westmead, Australia; dPriority Research Centre in children. Physical Activity and Nutrition, School of Health Sciences, Faculty of Health, University of Newcastle, Newcastle, Australia; METHODS: English-language articles from 1975 to 2010, available from ePaediatric Overweight Service Tayside, Perth Royal Infirmary, 7 databases, were used as data sources. Two independent reviewers Perth, United Kingdom; and fThe Children’s Weight Clinic, Edinburgh, United Kingdom assessed articles against the following eligibility criteria: randomized controlled trial, participants overweight/obese and #18 years, comparing KEY WORDS child, adolescent, lifestyle interventions, obesity, weight loss, lifestyle interventions to no treatment/wait-list control, usual care, or cardio-metabolic risks, systematic review written education materials. Study quality was critically appraised by 2 ABBREVIATIONS reviewers using established criteria; Review Manager 5.1 was used for CI—confidence interval meta-analyses. HDL—high-density lipoprotein HOMA-IR—homeostasis model assessment of insulin resistance RESULTS: Of 38 eligible studies, 33 had complete data for meta-analysis LDL—low-density lipoprotein on weight change; 15 reported serum lipids, fasting insulin, or blood — WMD weighted mean difference pressure. Lifestyle interventions produced significant weight loss All authors were involved in study conception and design, compared with no-treatment control conditions: BMI (21.25kg/m2, 95% interpretation of data, critical revision, and final approval of the submitted manuscript; Ms Ho was involved in quality confidence interval [CI] 22.18 to 20.32) and BMI z score (20.10, 95% CI assessment, data extraction, data analysis, and manuscript 20.18 to 20.02). Studies comparing lifestyle interventions to usual care preparation; Prof Collins, Dr Baur, and Dr Garnett were involved also resulted in significant immediate (21.30kg/m2, 95% CI 21.58 to in quality assessment, data extraction and study supervision; 2 2 2 2 2 and Drs Burrow and Steward were involved in quality 1.03) and posttreatment effects ( 0.92 kg/m , 95% CI 1.31 to 0.54) assessment and data extraction. on BMI up to 1 year from baseline. Lifestyle interventions led to www.pediatrics.org/cgi/doi/10.1542/peds.2012-1176 significant improvements in low-density lipoprotein cholesterol (20.30 doi:10.1542/peds.2012-1176 mmol/L, 95% CI 20.45 to 20.15), triglycerides (20.15 mmol/L, 95% CI 2 2 2 2 2 Accepted for publication Aug 7, 2012 0.24 to 0.07), fasting insulin ( 55.1 pmol/L, 95% CI 71.2 to 39.1) and blood pressure up to 1 year from baseline. No differences were Address correspondence to Ms Mandy Ho, Institute of Endocrinology and Diabetes, The Children’s Hospital at Westmead, found for high-density lipoprotein cholesterol. Locked Bag 4001, Westmead, NSW 2145, Australia. E-mail: mandy. CONCLUSIONS: Lifestyle interventions can lead to improvements in [email protected] weight and cardio-metabolic outcomes. Further research is needed PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). to determine the optimal length, intensity,andlong-termeffectiveness Copyright © 2012 by the American Academy of Pediatrics of lifestyle interventions. Pediatrics 2012;130:e1647–e1671 FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose. FUNDING: Ms Ho is supported by an Australian National Health and Medical Research Council Dora Lush Postgraduate Research Scholarship (APP 1017189). Dr Garnett is supported by a Cancer Institute NSW Early Career Development Fellowship Grant (10/ ECF/2-11). Dr Neve is supported by a Priority Research Centre in Physical Activity and Nutrition Postdoctoral Fellowship. Prof Collins is supported by a National Health and Medical Research Council Career Development Fellowship.

PEDIATRICS Volume 130, Number 6, December 2012 e1647 Downloaded from www.aappublications.org/news by guest on September 27, 2021 Obesity in children and adolescents is First, a detailed literature search was English language through CINAHL, a global public health concern and is conducted in September 2010 to iden- Cochrane Reviews, Current Concepts, associated with a range of short- and tify studies published between 2003 DARE, Embase, Premedline, and Medline. long-term health complications.1–4 Al- and 2010. Eligible studies from the The Medical Subject Headings of the though prevention of obesity is impor- previous review13 covering 1975 and National Library of Medicine keyword tant, so too are effective treatments for 2003 were then combined in the data search terms used were dietetic, those already affected. Lifestyle inter- synthesis with those from the current paediatric (pediatric), child, adoles- ventions, involving a combination of search. cent, family, parent, school, over- diet, exercise, and/or behavior modifi- weight, obesity, intervention, weight cation, are an essential element of Eligibility Criteria control, weight management, weight obesity management.5–7 Several sys- loss, and healthy weight (Supplemental Eligible studies were randomized con- tematic reviews of childhood obesity Appendix 1). In addition, the reference trolled trials of treatment of overweight have been published and lifestyle inter- lists of retrieved articles and key sys- and obesity in children and adolescents ventions targeting treatment of child tematic reviews of childhood obesity #18 years of age comparing the ef- and adolescent obesity are reported as treatments were scanned for relevant fectiveness of lifestyle intervention efficacious in weight loss in the short to references.8,9,11,12 programs incorporating a nutrition or medium term.8–12 The first specificre- dietary component with no treatment view of dietary interventions, published or wait-list control, usual care, or Study Selection in 2006,13,14 included studies published minimal advice or written diet and All studies identified in the database up to 2003, and found positive effects of physical activity education materials. search were assessed for relevance interventions that included a dietary Programs that involved the whole from the title and abstract by 2 in- component.14,15 family or were directed exclusively at dependent reviewers. Articles that met, The previously mentioned systematic parents of overweight or obese chil- or appeared to meet, the inclusion reviews and others have all presented dren and adolescents were also in- criteria were retrieved. All retrieved data on weight change outcomes; cluded. Additional inclusion criteria studies were assessed for relevance by however, obese children and adoles- were a follow-up period from baseline 2 independent reviewers. In case of cents also carry an increased risk for of at least 2 months, and inclusion of disagreement, a third independent re- cardio-metabolic complications, includ- the outcome measures of body weight viewer made the final decision. ing dyslipidemia, insulin resistance, and or body composition. Participants were – hypertension.15 20 To our knowledge, free living or attending obesity clinical Quality Assessment no systematic review has examined units, community programs, camps, Full copies of all included studies were the effects of lifestyle interventions schools, or one-off programs. Studies assessed for methodological quality by on cardio-metabolic outcomes in over- were excluded if they were targeted at 2 independent reviewers using the weight children and adolescents. obesity prevention or maintenance of Joanna Briggs Institute critical ap- Therefore, the aim of this review was to weight loss, were drug trials or inter- praisal of study quality tool (Supple- present the best available evidence ventions that dealt with eating dis- mental Appendix 2). Studies were rated from randomized controlled studies of orders, or if they focused on children as positive, negative, or neutral based lifestyle interventions incorporating a with obesity attributable to a second- on responses to 10 items. Discrep- dietary component to assess their im- ary or syndromal cause. Studies that ancies were resolved by discussion or pact on both weight loss and cardio- were not written in English, or included consultation with a third reviewer to metabolic risks. This review covers children who were within the healthy achieve consensus. literature published between 1975 and weight range at baseline, were ex- 2010. cluded. No restrictions were placed on Data Extraction intervention settings or who delivered Data in relation to methodology, in- the interventions. METHODS tervention effect, compliance, and in- The protocol and search strategy for tensity were extracted by the first this systematic review was based on Data Source and Search Strategy reviewer by using a standardized form the previous peer-reviewed protocol13 The search strategy involved a litera- developed specifically for this review. registered with the Joanna Briggs In- ture search conducted by a medical This was verified by a second reviewer stitute. It involved a 2-stage process. librarian of published literature in the for accuracy and a consensus reached

e1648 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 REVIEW ARTICLE where disagreement existed. Data de- systolic and diastolic blood pres- studies ranged from 16 to 258, with scribing interventions that were re- sures. a median of 72 participants per study. ported in more than 1 article were Where outcomes could not be quantita- Twenty-seven studies had 2 study arms, 9 extracted together. tively combined in a meta-analysis, they had 3 study arms, and 2 had 4 study are described in a narrative summary. arms (Table 1). Among the no treatment Data Synthesis For forest plots with sufficient studies or wait-list control comparisons (n = 22) . (Table 1), the intervention lengths Review Manager (RevMan5.1, The included ( 10), we generated funnel varied from 1 month (n =1)to2years. Cochrane Collaboration, Oxford, England) plots to examine for the publication bias. Twelve studies did not follow the par- was used for meta-analyses. All the RESULTS ticipants after the intervention pro- outcomes in this review were continu- gram ended.22–24,26,28,29,31,32,36,63,64,68 For ous outcomes and a weighted mean Search Result the comparison of lifestyle intervention difference (WMD) was calculated if the The literature search identified 4713 with usual care (n = 11), the inter- same measurement scale was used. references (Fig 1), and 434 full articles vention lengths varied from 3 months to When different outcome measurement were retrieved. Forty-one articles re- 1 year, and 6 studies conducted sub- scales were reported, we conducted lating to 30 different studies met all sequent follow-up,41–43,45,65,69 ranging the meta-analysis by using the stan- 22–62 inclusion criteria. Eight additional from 2 months to 4 years from the end dardized mean difference approach. studies (12 articles) from the previous of the active intervention component. Heterogeneity was assessed by I2 sta- review that met the comparison crite- Among the 5 written information studies, tistics. Heterogeneity is considered to 63–74 ria were also included. The total 1 had a varied intervention length, 2 had be low if I2 is #40%, and high if I2 is number of studies included in this re- an intervention length of 6 months,48,49 $75%.21 We used a random effects view is 38. and another 2 were 1-year50,51 interven- model for meta-analysis if there was tion programs with the outcome evalua- significant heterogeneity (I2 .40%), Description of Included Studies tion at the end of the intervention. and fixed effects for homogeneous Study characteristics and weight- (I2 # 40%). BMI and BMI z score were related outcomes are summarized in Methodological Quality used as the primary weight loss out- Table 1. Nearly half of the studies were No studies fulfilled all requirements comes. We also examined the effects of conducted in the United States (n = listed in the study quality critical ap- interventions on body composition by 18),22,24,29,33,41,46–49,50,56,63,64,66–69,71 5in praisal tool, although 8 studies met 8 using percentage body fat. By using Australia,25,30,35–37 and the others in Israel ofthe10requirements25,27,30,32,37,40,49,69 the last time point of weight loss (n =3),39,43,44 Germany (n =2),32,51 the (Table 2). Twenty-four studies did measurement for each study, we per- United Kingdom (n =2),34,40 Belgium,65 not specify the method of randomi- formed meta-analyses among sub- China,28 Finland,42 Iran,45 Korea,31 zation.22–24,26,28,29,31,33,36,39,42,45,46,48–51,63–68,71 groups by age (child defined as mean Mexico,38 Taiwan,26 and Tunisia.23 Eigh- Details of allocation conceal- age at baseline #12 years and ado- teen studies targeted obese children ex- ment23,24,26,28,29,31,33–36,41,43–46,50,51,63,66–68,71 and lescent as .12 years), and the length clusively,22,23,26,28,29,31,34,38,40–42,44,46,51,64–66,71 study blinding24,27–31,33,36,37,39,41–43,45,48,50,51,63–69,71 of the follow-up from baseline. Where whereas the others targeted both over- were not adequately reported for most key details or data were missing, au- weight and obese children. Most studies studies. Blinding of participants in di- thors were contacted, or data im- (n = 14) were conducted in a hospital etary and lifestyle interventions is puted based on methods described in environment,22,25,29,32,36,38,40–41,43–44,46,63,65,68 usually not possible. Only 5 studies the Cochrane Handbook.21 The follow- followed by the community (n=6),23,27,33–35,51 reported that outcome assessors were ing cardio-metabolic outcomes were school (n =6),26,42,48–50,64 and primary blinded to participants’ treatment al- examined: care setting (n =6).30,37,39,47,67,69 Thir- location.25,30,37,38,40 Overall, retention rates serum lipids, including total choles- teen studies were conducted in chil- for all included studies ranged from 38% terol, low-density lipoprotein (LDL) dren,25,30,33–35,37,40–42,44,47,50,66 7in to 100%. Most studies (29/38) had a re- cholesterol, high-density lipoprotein adolescents,22–24,28,29,31,36 and others tention rate of $70% at 6 months or (HDL) cholesterol, and triglycerides; enrolled both children and adoles- .60% at 1 year. Only 9 studies used fasting glucose, fasting insulin, and cents.26,27,32,38,39,43,45,46,48,49,51 Only 1 study intention-to-treat analysis.32,41,44,46,47,62,67–69 insulin resistance as determined by included children aged ,5years.35 Four Six studies did not report dropout rates the homeostasis model assessment adolescent studies specifically targeted and it was therefore not clear if they used of insulin resistance (HOMA-IR); and girls.24,29,31,45 Thesamplesizeofincluded an intention-to-treat analysis.23,26,35,45,48,63

PEDIATRICS Volume 130, Number 6, December 2012 e1649 Downloaded from www.aappublications.org/news by guest on September 27, 2021 FIGURE 1 Flowchart for identification of trials for inclusion in a systematic review of lifestyle interventions incorporating a dietary component in overweight and obese children and adolescents. aCollins CE, Warren JM, Neve M, McCoy P, Stokes B. Systematic review of interventions in the management of overweight and obese children which include a dietary component. International Journal of Evidence-Based Healthcare. 2007;5(1):2–53.

Dietary Interventions reported baseline energy intake,23 and ranged from 20 minutes per month25 23 Ten studies used the Traffic Light or another imposed caloric restrictions to 6 hours per week, most providing 22 fi ∼ modified Traffic Light diet as their di- on snacks and beverages. One- fth 1.5 to 2.0 hours of training each week etary intervention.27,28,32,38,40,41,48,49,66,69 of the included studies inadequately (Table 1). Three studies used pedome- 27,33,36 The Traffic Light diet is a calorie- described the details of dietary inter- ters to promote physical activity. 31,33,36,39,48,50,71 controlled approach in which foods in ventions. Other studies each category are color-coded ac- provided general healthy eating ad- Effects of Lifestyle Intervention cording to their calorie density per vice (Table 1). A dietitian was reported Compared With No Treatment or average serving: green for low-calorie to be involved in the delivery of the Wait-Listed Control foods that can be eaten freely; yellow dietary interventions in 13 stud- Eighteen of the 22 studies that com- 25,35,38,40,42–47,51,63,67 for moderate-calorie foods that can be ies. pared lifestyle intervention with no eaten occasionally; and red for high- treatment or a wait-listed control calorie foods that should be eaten Exercise Interventions group reported a positive effect on rarely. Four studies used a hypocaloric Nineteen studies conducted super- weight loss22,23,25–28,31,32,34–36,63,64,66–69,71 diet or a calorie restriction approach. vised physical activity sessions or (Table 1). In the meta-analysis, which Two aimed for a 30% of reported en- exercise training as part of the inter- included 19 studies (24 comparisons) ergy intake deficit or 15% less than vention,22–26,31,32,34,42–46,48,49,51,63,64,68 although and 1234 participants, there was estimated daily requirements.43,44 One the intensity and variety varied. The total a significantly larger effect on weight study targeted 500 kcal less than the duration of physical activity sessions and body composition (standardized

e1650 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 EITISVlm 3,Nme ,Dcme 2012 December 6, Number 130, Volume PEDIATRICS TABLE 1 Study Characteristics and Weight-Related Outcomes (Structured by Year of Publication) Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Lifestyle intervention compared with no-treatment or wait-list control Botvin 1979,64, e, 119 12 to 17 Obese (120% 10 wk, (1) Control (2) Weekly session (1 class period each) (1) 74% % of participants Yes: Significant School, USA F+M of their 10 wk (2) DA + PA + BM conducted by allied health professionals and (2) 100% $130% and decrease in ,

Downloaded from ideal BW) by school faculty members. 130% of ideal % overweight DA: Healthy food selection, prudent diet, pre- weight of the planning and changing eating patterns. intervention PA: 10-min supervised session each week (P , .05) (jogging, jumping rope, bicycling) No significant change in % overweight of www.aappublications.org/news the control group Epstein 53 families 8 to 12 Obese ($20% 6mo, (1) Wait-list (2) 15 treatment sessions, parent and child in Overall: Change in % % overweight: Yes 1984,66,70,73,e, F+M overweight and 1, 5, control separate groups (session 1–8: weekly; 94% at 1 y overweight (3 .1 at 6 mo) Setting: tricep skinfolds and 10 y (2) DA session 9–11: biweekly; session 12–15: unclear, USA .85th (3) DA + PA monthly). Parents encouraged to lose weight percentile) too. Traffic Light Diet (1200–1500 kcal/d, limit to 4 red foods/wk). (3) DA: as group 2 PA: lifestyle change exercise program - increase byguest on September27, 2021 energy expenditure 200–400 kcal/d Kirshenbaum 40 families 9 to 13 Obese ($20% 9wk, (1) Wait-list (2) 9 weekly 90-min treatment sessions. Focus (1) 89% % overweight, % overweight: Yes 1984,71, F+M overweight) 3 and control on cognitive-behavioral treatment. (2) 67% adjusted weight at 3 mo (2.1) Setting: 12 mo (2) DA + PA + CBT DA and PA: no details given. Only children (3) 87.5% unclear, USA (Child only) attended the group (6–9 children per group) (3) DA + PA + CBT (3) as group 2, both parents and children (Parent & child) attended all sessions together (4–5 dyads per group) Mellin 1987,67 66 12 to 18 Overweight 3mo, (1) Control (2) 14 weekly sessions (90 min each) for Overall: 84% Change in relative Relative weight: “Shapedown F+M and obese 6 and (2) DA + PA + CBT adolescentsplus2parentsessions.Usesself- weight Yes (2.1) Program,” (113% to 213% 15 mo directed change format and encourages Primary care, relative weight) participants to make successive, USA sustainable, small modifications in diet, exercise, lifestyles, and attitudes. Avoid very low calorie or restrictive diets. Becque 1988,63 36 12 to 13 Overweight 20 wk, (1) Control (2) DA: American Dietetic Association Exchange NR % body fat % body fat: Yes e Hospital F+M and obese 20 wk (2) DA + BM Program with kcal prescription to give (hydrostatic (3.1) ARTICLE REVIEW environment, (BW and triceps (3) DA + BM + PA weight loss of 1–2 lb/wk. Weekly nutrition weighing) USA skinfold .75th education materials and adjustment of percentile) eating pattern and kcal intake.

e1651 (3) DA: as group 2 PA: 50 min flexibility, muscle strengthening and aerobic exercise for (3 times/wk) e1652 TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference

Oe al et HO a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Rocchini 1988,68,74 72 10 to 17 Overweight and 20 wk, (1) Control (2) Weekly 1-h class (1) 82% % body fat % body fat: Yes Hospital F+M obese (Weight 20 wk (2) DA + BMe DA: modified kcal exchange program to (2) 85% (hydrostatic (3.1) environment, for height (3) DA + PA + BM produce a weight loss of approximately (3) 92% weighting) USA .75th 1 lb/wk percentile; (3) DA: as group 2

Downloaded from triceps and PA: supervised exercise 3 times/wk (1 h each, 10 subscapular min stretching and muscle strengthening skin folds exercise plus 40 min aerobic exercise) .80th percentile) Balagopal 16 14 to 18 Obese (BMI $30) 3 mo, (1) No-treatment (2) DA: caloric restriction (snacks and Overall: 95% BMI, % body fat BMI and % body www.aappublications.org/news 2003,22,52,53 F+M 3mo control beverage); met with a nutritionist once/wk (DXA) fat: Yes (2.1) “Shapedown (2) DA+PA+SB+BM PA: 45 min 3 times/wk; 1 monitored PA session/ Program,” wk in the presence of parent Hospital SB: limiting TV viewing environment, USA Jiang 2005,28 75 families 7th to 9th Obese (Chinese 2y, (1) No-treatment (2) Home visit by pediatrician once per month (1) 90% BMI BMI: Yes (2.1) Family, China F+M grade references 2y control DA: modified traffic light diet (2) 92% weight for (2) DA+PA+SB+BM PA: 20–30 min/d, 4 d/wk height $120%) SB: general advice

byguest on September27, 2021 Rooney 2005,33,e, 98 families, 5 to 12 Overweight and 12 wk, (1) No-treatment (2) family members received a pedometer and Overall: 89% BMI percentile No “Growing 353 people obese (CDC 1y control instructed to walk 10 000 steps daily and Healthy Family F+M Growth Chart (2) PA received a biweekly newsletter study”, BMI .84th (3) DA+PA (3) as group 2 plus attended 6, 1-h biweekly Community, USA percentile) sessions concerning nutrition, physical activity, and other parenting issues. Golley 2007,25,54 111 6 to 9 Overweight and 6mo, (1) Wait-list (2) Parent attended 4 weekly 2-h group sessions (1) 86% BMI z score, WC z BMI z score: No Hospital F+M obese (IOTF cut- 12 mo control on parenting skills training (family lifestyle (2) 78% score WC z score: Yes environment, off and BMI z (2) PS (parents change) followed by 4 weekly, then 3 monthly (3) 82% (2.1; 3.1) Australia score # 3.5) only)e 15- to 20-min individual telephone sessions. (3) PS + DA + PA (3)Parentscompleted thesameparentingskills training as group 2 plus an additional 7 intensive lifestyle support group sessions. Children attended 7 supervised activity sessions (focused on aerobic activity and motor skills development). DA: general eating guidelines and core food serve recommendations EITISVlm 3,Nme ,Dcme 2012 December 6, Number 130, Volume PEDIATRICS TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Huang 2007,26 120 10 to 13 Obese (Taiwanese 12 wk, (1) No-treatment nutrition NR BMI, % body fat BMI and % body School, Taiwan F+M reference BMI 12 wk control (2) DA: children received 30-min nutrition (BIA) fat: Yes (2.1) $95th (2) DA + PA instruction twice/wk at school

Downloaded from percentile) PA: 40-min classroom-based noncompetitive aerobic activities 3 times/wk McCallum 163 5.0 to 9.9 Overweight and 12 wk, (1) No-treatment (2) 4 standard consultations by general (1) 85% BMI, BMI z score No 2007,30,58 F+M obese (IOTF cut- 9mo, control practitioners assisted by a 20-page “family (2) 94% “LEAP trial,” off and BMI z 15 mo (2) DA+PA+SB folder” written at a 12-y-old reading level, Primary care, score ,3.0) targeting change in nutrition, physical

www.aappublications.org/news Australia activity, and sedentary behavior. Park 2007,31 44 13 to 15 Obese (Korean 12 wk, (1) No-treatment (2) Lifestyle education offered by a trained (1) 95% BMI, % fat (BIA), WC BMI,%bodyfatand Setting: unclear, F reference BMI 12 wk control counselor once/wk. (2) 91% and waist/hip waist/hip ratio: Korea $95th (2) DA+PA+BM DA: general advice ratio Yes (2.1) percentile) PA: walking 6 d/wk (10 min for 3 d and 30 to 40 min for 3 d) Shelton 2007,35 43 3 to 10 Overweight and 4wk, (1)Wait-list control (2) Parents attended 4, 2-h weekly group NR BMI BMI: Yes (2.1) Community, F+M obese (CDC 3mo (2) DA+PA+BM sessions and received parent treatment Australia growth chart (parent only) manual. BMI $85th percentile) byguest on September27, 2021 Janicke 2008,27,55,56 93 8 to 14 Overweight and 4mo, (1) Wait-list (2) Weekly group sessions (90 min each) for 8 (1) 81% BMI z score BMI z score: Yes “Project STORY”, F+M obese (CDC 10 mo control wk, followed by biweekly for 8 wk. Parent and (2) 73% (2.1; 3.1) Community (rural growth chart (2) DA+PA child participated in separate groups, then (3) 77% setting), USA BMI .85th (3)DA+PA(parents broughttogetherattheendofeachsessionto percentile) only) develop goals for the wk and specific plans. DA: modified version of the Stoplight Diet PA: pedometer-based step program (3) as group 2, but only parents attended group meetings. Tsiros 2008,36 47 12 to 18 Overweight and 20 wk, (1) No-treatment (2) Core program (first 10 wk): 8 clinic sessions Overall: 38% BMI, WC, body fat BMI, body fat and “Choose Health F+M obese (IOTF 20 wk control (1 h each) plus 1 phone-call session and (DXA) abdominal fat: program,” cut-off) (2) DA+PA+BM+CBT provided with a book on healthy eating Yes (2.1) Hospital and the Australian Guide to Healthy Eating. environment, Maintenance phase (10 wk): 4 fortnightly Australia phone calls. DA: Australian Guide to Healthy Eating

PA: provided with a pedometer ARTICLE REVIEW e1653 e1654 TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference

Oe al et HO a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Davis 2009,24 50 14 to18 Overweight and 16 wk, (1) No-treatment (2) Weekly 90-min dietary session plus 4 Overall: 82% BMI, BMI z scores, BMI: Yes (1.3; Setting: unclear, F obese (CDC 16 wk control motivational interview sessions. BMI percentile 1=4) USA growth chart (2) DAe DA: #10% of energy from added sugar plus at and fat mass BMI z score and fat BMI .84th (3) DA + PA least 14 g·1000 kcal21 of dietary fiber a day (DXA) mass: No percentile) (strength (3) dietary intervention as group 2 plus twice/

Downloaded from training) wk 60 min strength training. (4) DA + PA (4) as group 3 but 30 min cardio and 30 min (strength and strength training. cardio training) Kitzman-Ulrich 42 12 to 15 Obese (CDC 4 mo, (1) Wait-list (2) DA: Food Guide Pyramid Overall: 83% BMI z score No 2009,29 F growth chart 4mo control PA: general advice www.aappublications.org/news Hospital BMI .95th (2) DA+PA+BM (3) as group 2 plus attended a 45-min environment, percentile) (3) DA+PA+BM+ multifamily therapy group. USA Multifamily therapy Wake 2009,37 258 5 to 9.99 Overweight and 12 wk, (1) No-treatment (2) 4 standard consultations by general (1) 98% BMI No “LEAP 2 trial,”, F+M obese (IOTF 6mo, control practitioners assisted by a 16-page “family (2) 96% Primary care, cut-off and 12 mo (2) DA+PA+SB folder” written at a 12-y-old reading level, Australia BMI z score targeting change in nutrition, physical ,3.0) activity, and sedentary behavior. Ben Ounis 2010,23 28 Mean age Obese (BMI .97th 8wk, (1) No-treatment (2) Nutrition education program 4 h/wk NR BMI, % body fat BMI and % body

byguest on September27, 2021 Community, F+M 13.160.8 percentile, 8wk control DA: calorie restriction (500 kcal less than the (skin fold) fat: Yes (2.1) Tunisia reference not (2) PA + DA reported energy intake at baseline, 15% specified) energy from protein, 55% carbohydrate, 30% fat) PA: supervised exercise training 4 times/wk (90 min each) Reinehr 2010,32 71 8 to 16 Overweight and 6mo, (1) Wait-list (2) Intensive phase (3 mo): 6 nutrition groups (1) 84% BMI, BMI z score, BMI z score, WC “Obeldicks Light” F+M obese 6mo control (1.5 h each) education session for children (2) 97% WC, % body fat and % body fat: Program, Hospital (Germany (2) DA+PA+SB+BM (stratified by gender and age) plus 6 parent (BIA) Yes (2.1) environment, reference evening sessions (1.5 h) plus 30 min Germany BMI 90th to 97th individual dietary counseling; Establishing percentile) phase (3 mo): 1 individual nutrition counseling session plus 3 counseling sessions. DA: Optimized mixed diet (food-based dietary guideline) and Traffic Light system PA: weekly physical activity training (1.5 h each, mainly aerobic exercise) for 6 mo SB: general advice EITISVlm 3,Nme ,Dcme 2012 December 6, Number 130, Volume PEDIATRICS TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Sacher 2010,34 116 8 to 12 Obese (UK 1990 6mo, (1) Wait-list (2) 18, 2-h biweekly sessions for children and (1) 68% BMI and BMI z BMI and BMI z MEND program, F+M reference 12 mo control their parents and siblings. (2) 70% score score: Yes Community, data BMI .98th (2)DA+PA+BM+12- DA: 8 sessions, general nutrition education and (2.1)

Downloaded from United Kingdom percentile) wk free family nondieting approaching swim pass PA:16sessions(1heach),noncompetitivegroup play Lifestyle intervention program compared with usual care or minimal advice Braet 1997,65,72,e, 259 children 7 to 17 Obese ($20% Varied, (1) Minimal (1) Advice in 1 session (3 h) and given treatment Overall: 81% 1yfollow-uppaper, No: Significant Hospital (1 y follow- overweight) 1.0 and 4.6 y therapeutic manual for parents and workbook for at 1 y Mean % weight decrease in %

www.aappublications.org/news environment, up study) contact: DA, PA children 80% at 4.6 y loss overweight in Belgium 136 children (2) Individual (2) Received the same information packages as 4.6 y follow-up all groups at 3, 6 (4.6 y treatment: CBT, group 1 plus children received 7 sessions of paper, % mo, 1.0 and 4.6 y follow- DA, PA 90 min (twice/mo) and 7 monthly family overweight and up study) (3) Group follow-up sessions on an individual basis. mean % weight F+M treatment: CBT, DA: Healthy eating + 3 mid-meal snacks. loss DA, PA) Unhealthy meals limited to 1/mo or small (4) Summer camp amount weekly. No kcal counting. training: CBT, PA: moderate exercise 30 min/d DA, PA (3) same as group 2 but on a group approach (4) 10-d summer camp: Balanced healthy food byguest on September27, 2021 1500 kcal/d plus daily lifestyle exercises 5 h/ d and families encouraged to attend monthly follow-up sessions for 1 y Saelens 2002,69, 44 12 to 16 Overweight 4mo, (1) Typical care: (1) Single appointment with a pediatrician at (1) 90% BMI, % overweight BMI: Yes (2.1) Primary care, USA F+M and obese 7mo DA, PA baseline (nontailored counseling session) (2) 78% (20% to 100% (2) Intensive DA: Food Guide Pyramid above the follow-up: DA, PA: 60 min/d median for BMI) BM, PA, SB (2) Computer-based program: assessment of eating, PA and SB. At baseline, a pediatrician provided tailored counseling session based on the computer-generated individualized action plan. Weekly telephone counseling for 8 wk then biweekly until 14 to 16 wk. Also received behavior modification manual. DA: 1200–1500 kcal/d and weekly goal of 40 green (#1 g fat, ,150 kcal and nutrient

dense) and ,15 red foods (5 g or more fat ARTICLE REVIEW or diet versions of high-fat foods) PA: $60 min moderate intensity exercise 5 d/wk

e1655 SB: general advice e1656 TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference

Oe al et HO a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Nemet 2005,43,60 54 6 to 16 NR 3 mo, (1)Minimaladvice: (1) At least 1 ambulatory nutritional (1) 67% BMI, % body fat BMI and % body fat Hospital F+M 1y DA, PA consultation plus general exercise advice. (2) 83% (skin fold), BMI Yes (2.1) environment, (2) Lifestyle (2) DA: hypocaloric diet (30% deficient based on percentile Israel program: DA, PA reported intake or 15% less than estimated daily requirement). Met dietitian 6 times in 3

Downloaded from mo (2.5 h in total). PA: twice-weekly training (mostly endurance training, 1 h/session) Gillis 2007,39 27 7 to 16 Overweight and 6mo, (1)Minimaladvice: (1) A 0.5-h talk on exercise and diet at baseline (1) 54% BMI z score No Primary care, F+M obese (BMI 6mo DA, PA (2) Talks of 0.5 h at baseline and 3 mo, plus weekly (2) 79% Israel $90th (2) Lifestyle follow up phone calls. Also, instructed to www.aappublications.org/news percentile, program: DA, PA record diet and exercise on 1 day of each wk reference not specified) Kalavainen 70 7 to 9 Obese (Finnish 6mo, (1) Routine (1) 2 individual counseling sessions by school (1) 100% BMI and BMI z BMI and BMI z 2007,42,59 F+M reference 12 mo Program: DA nurse (30 min each) plus information (2) 97% score score: Yes School, Finland weight for (2) Lifestyle booklets for family. (2.1) height 120% to program: DA, (2) Family-based group treatment, 15 sessions 200%) PA, SB, BM, CBT (90 min each) with separate sessions for parentsandchildrenand1jointsession.First 10 sessions were held weekly and then

byguest on September27, 2021 fortnightly. DA: general healthy eating PA: noncompetitive activities Savoye 2007,46,61 209 8 to 16 Obese (CDC 1y, (1) Standard care: (1) Met dietitian every 6 mo, general healthy (1) 64% BMI and % body BMI and body fat: “Yale Bright F+M growth chart 1y DA eating and physical activity advice. (2) 71% fat (BIA) Yes (2.1) Bodies Weight BMI .95th (2) Lifestyle (2) Intensive family-based nutrition and Management percentile) program: DA, behavior modification (weekly 40 min for the Program”, PA, BM first 6 mo then every other wk). Hospital DA: a nondieting approach, focused on better environment, USA food choices PA: twice 50-min sessions of high-intensity aerobicexercisefor thefirst6mofollowedby 100 min twice/month. Plus encouraged to exercise 3 additional days at home per wk EITISVlm 3,Nme ,Dcme 2012 December 6, Number 130, Volume PEDIATRICS TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Hughes 2008,40,57 134 5 to 11 Obese (UK 12 mo, (1) Standard (1)3to4appointmentswithdietitianover6to10 (1) 63% BMI z score, No: both groups “SCOTT,” Hospital F+M reference BMI 12 mo care: DA mo (1.5 h in total). (2) 65% WC z score had a environment, $98th (2) Lifestyle DA: general healthy eating and mainly directed significant

Downloaded from United Kingdom percentile) program: DA, toward parents. decrease in BMI PA, BM (2) Family-centered lifestyle intervention z score. program, 8 appointments over 26 wk (5 h in total). DA: modified Traffic Light guide PA: not specified #

www.aappublications.org/news SB: 2 h per day Nemet 2008,44 22 8 to 11 Obese (CDC 3mo, (1)Minimaladvice: (1) At least 1 ambulatory nutritional Overall: BMI, % body fat BMI: No % body fat Hospital F+M growth chart 3mo DA, PA consultation plus general exercise advice. 100% (BIA), BMI and BMI environment, BMI . 95th (2) Lifestyle (2) DA: hypocaloric diet (30% deficient based on percentile percentile: Yes Israel percentile) program: DA, reported intake or 15% less than estimated (2.1) PA, BM daily requirement). The children met with the dietitian weekly and the parents met separately with the dietitian biweekly. PA: twice-weekly 1-h training (mostly endurance training) Kalarchian 2009,41 192 8 to 12 Obese (CDC 6mo, (1) Usual care: DA (1) 2 nutrition sessions based on Stop Light (1) 85.3% BMI, WC, and % At 6 mo, BMI and % byguest on September27, 2021 Hospital F+M growth chart 12 mo, (2) Lifestyle eating plan (2) 83.5% body fat (DXA) body fat: Yes environment, USA BMI .97th 18 mo program: DA, (2) First 3 mo: 20 group meetings (60 min each). (2.1) percentile) PA, SB, BM Parents and children in separate group, then At 12 and 18 mo, % joined to set weekly goals body fat: Yes 6 to 12 mo: 3 group sessions plus 3 telephone (2.1) follow-up 12–18 mo: no contacts DA: modified Stop Light eating plan SB: ,15 h/wk Sarvestani 60 11 to 15 NR 16 wk, (1) Standard care: (1) 3 behavioral modification intervention NR BMI BMI: Yes (2.1) 2009,45,e F 6mo DA, PA, BM sessions (same program as group 2). Setting: (2) Lifestyle (2) 16 behavioral modification intervention unclear, Iran program: DA, sessions (2 h behavior modification or PA, BM dietary instruction plus 2 h yoga therapy). Plus 4 individual diet sessions and required to keep 24-h food records. EIWARTICLE REVIEW e1657 e1658

Oe al et HO TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Diaz 2010,38 76 9 to17 Obese (CDC 12 mo, (1) Usual care: DA (1) Monthly 10- to 15-min consultations with (1) 58% BW, BMI, BMI z BMI, BMI z score, Shapedown F+M growth chart 12 mo (2) Lifestyle primary care physician (2) 55% score, WC, WC, and body

Downloaded from Program, Hospital BMI .95th program: DA, DA: Food Guide Pyramid body fat (DXA) fat: Yes (2.1) environment, percentile or PA, SB, BM PA: 30 min most days of the wk Mexico BMI .90th SB: #2 h/d percentile with (2) 12 weekly 2-h behavior group program and WC .90th weekly dietitian counseling sessions for the percentile) first 3 mo, then monthly until 12 mo. Plus 12 www.aappublications.org/news monthly physician consultations. Parents received 6 education sessions and were encouraged to lose weight if they were overweight. DA: Traffic Light diet approach pluseducation on glycemic index and provided with an individualized diet plan of 1200 to 1800 kcal/d PA: not specified SB: not specified Face-to-face education compared with written education materials 48 byguest on September27, 2021 Fullerton 2007, 80 6th and 7th Overweight and 6mo, (1) Written (1) To follow a 12-wk parent-guided manual of NR BMI z score BMI z score: Yes School, USA F+M graders obese (CDC 6mo materials Trim Kids. (2.1) growth chart (2) Intensive (2) Children received an instructor-led daily BMI $85th instructor-led intervention during school days (1 nutrition percentile) intervention class and 4 physical activity classes weekly) group: DA, PA, for 12 wk, then monthly booster sessions. BM Parents offered monthly evening training sessions. Johnston 71 10 to 14 Overweight and 6mo, (1) Written (1) To follow a 12-wk parent-guided manual of (1) 88% BMI and % body BMI: Yes (2.1) 2007,49,62 F+M obese (CDC 6mo materials Trim Kids. (2) 95% fat (BIA) % body fat: No School, USA growth chart (2) Intensive (2) Children received an instructor-led daily BMI $85th instructor-led intervention (1 nutrition class and 4 physical percentile) intervention activity classes weekly, 35 to 40 min each) group: DA, for 12 wk, then biweekly during the last PA, BM period of school. Plus received snack bars for a daily afternoon school snack and a cereal for breakfast. Parent attended monthly meetings. EITISVlm 3,Nme ,Dcme 2012 December 6, Number 130, Volume PEDIATRICS

TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Downloaded from DA: food groups were labeled “safety” (most fruits and nonstarchy vegetables), “caution” (low-fat meat, low-fat dairy and complex carbohydrate) and “danger” ($ 5 grams of fat or $ 15 grams of sugar per serving) zone www.aappublications.org/news food PA: 2-stage approach, wk 1 to 6 aimed to develop a basic level of physical fitness; wk 7 to 12 to focus on sport skill development Weigel 2008,51 73 7 to 15 Obese (WHO 1y, (1) Written dietary (1) Written advice (Food Guide Pyramid) from (1) 83% BMI, BMI z score BMI z score and Community, F+M growth 1y advice: DA a pediatrician during an outpatient visit at (2) 97% BMI: Yes (2.1) Germany standard) (2) Group baseline and 6 mo. sessions:DA,PA, (2) Children received weekly group nutrition BM education (adapted from the Food Guide Pyramid fruit and vegetable template, 45–60

byguest on September27, 2021 min/session) and coping strategy training. Parents received monthly 2 h support sessions. PA: weekly session, alternating swimming and indoor sports EIWARTICLE REVIEW e1659 e1660 Oe al et HO

TABLE 1 Continued Study, Program Participants Intervention Study Arms and Interventions Retention Weight-Related Significance Name (When Length, Study Components Ratec Outcome Difference a Applicable), N Age, y Selection Follow-up Reported Between b Setting, Country Gender Criteria From Groupsd Baseline Downloaded from Estabrooks 220 8 to 12 Overweight and Intervention (1) Written (1) Received a 61-page workbook designed to (1) 72% BMI z score BMI z score: Yes 2009,47 F+M obese (CDC length materials: DA, promote physical activity, increase (2) 66% (3.1; 1.2) Primary growth chart varied, PA, SB consumption of fruits and vegetables and (3) 74% care, USA BMI $85th follow up at 6 (2) Written decreased sugared drinks, TV viewing, and percentile) and 12 mo materials + sedentary behavior.

www.aappublications.org/news Group session (2) As group1 plus 2 weekly small group with dietitian: education sessions (2 h each). DA, PA, SB, BM (3) As group 2 plus 10 weekly follow-up phone (3) Written calls delivered by interactive voice response materials + system and based on goals set by the Group session participants. with dietitian + interactive voice response technology: DA, PA, SB, BM byguest on September27, 2021 Resnick 2009,50,e, 46 families Kindergarten Overweight 1y, (1) Written (1) 6mailingsofeducational materialsat∼5-wk (1) 100% BMI percentile No: both groups CHEER Program, F+M and 5th and obese 1y materials intervals. (2) 86% lost weight School, USA graders (CDC (2) Written (2) As group 1 plus at least 1 home visit or growth materials + phone call by the community health chart BMI home visit or workers. $85th phone call percentile) F, female; M, male; BW, body weight; DA, dietary advice; PA, physical activity; BM, behavioral modification; kcal, kilocalorie; CBT, cognitive behavior therapy; NR, not reported; SB, sedentary behavior; DXA, dual-energy x-ray absorptiometry; CDC, Centers for Disease Control and Prevention; IOTF, International Obesity Taskforce; WC, waist circumference; PS, parenting skill; BIA, bio-impedance analyzer; Project STORY, Project sensible treatment of obesity in rural youth; SCOTT, Scottish Childhood Overweight Treatment Trial; WHO, World Health Organization; CHEER Program, Communicating about Health, Exercising, and Eating Right Program. a Number of participants at randomization. b Classification of overweight and obesity standardized using the following definitions: Overweight = BMI 85th to 95th percentile for age and sex; obese = BMI $95th percentile for age and gender or weight $ 120% of average weight for height. c Retention rates reported post intervention if no follow-up, or at latest point of follow-up. d 2.1 means group 2 had a greater reduction than group 1 (P , .05). e Not included in the meta-analyses. REVIEW ARTICLE

TABLE 2 Selected Methodological Quality of Included Studies: Results Are Presented as Number of Studies (%) True Allocation Blinding of Outcome Baseline Retention Rate $70% at Intention-to- Randomization Concealment Measurement Comparability 6moor$60% at 1 y Treat Analysis Lifestyle interventions compared Yes 7 (32) 5 (23) 3 (14) 18 (82) 16 (73) 3 (14) with no-treatment/wait-list No 0 (0) 0 (0) 2 (9) 1 (4) 1 (4) 13 (59) control, n = 22 studies Unclear 15 (68) 17 (77) 17 (77) 3 (14) 5 (23) 6 (27) Lifestyle interventions compared Yes 5 (45) 3 (27) 2 (18) 10 (91) 8 (73) 4 (36) with standard care/minimal No 0 (0) 1 (0) 1 (9) 1 (9) 2 (18) 5 (46) advice, n = 11 studies Unclear 6 (55) 7 (73) 8 (73) 0 (0) 1 (9) 2 (18) Lifestyle interventions compared Yes 2 (40) 1 (20) 0 (0) 4 (80) 4 (80) 2 (40) with written education materials, No 0 (0) 0 (0) 0 (0) 1 (20) 0 (0) 2 (40) n = 5 studies Unclear 3 (60) 4 (80) 5 (100) 0 (0) 1 (20) 1 (20) mean difference 20.97, 95% confidence Effects of Lifestyle Intervention a 2.52 kg/m2 greater reduction in interval (CI) 21.39 to 20.55) for lifestyle Program Compared With Usual pooled BMI (95% CI: 0.91–5.95, I2 = interventions compared with control Care or Minimal Intervention 97%) and 0.06 greater reduction in over 2 years (Supplemental Fig 8). Eight of the 11 studies reported a posi- pooled BMI z score (95% CI: 0.02–0.10, 2 However, the studies were significantly tive effect of the lifestyle intervention as I = 99%) for the lifestyle intervention 2 heterogeneous (I = 90%) and the effect compared with usual care or minimal programs compared with written ed- size varied by age and length of study. interventions.38,41–46,69 The overall effect ucational materials only over 1 year There was no evidence of publication size in the meta-analysis, which included (Fig 4A and B). bias or small-study effects with visual 7 studies (586 participants),38,41–44,46,69 inspection of the funnel plot. 2 was a decrease in BMI of 1.30 kg/m at Effects of Lifestyle Interventions on We also conducted meta-analyses of the end of active intervention (95% CI: Cardio-metabolic Outcomes 12 studies (899 participants) that re- 1.03–1.58, I2 =0%to48%)(Fig3A).Stud- Table 3 summarizes the metabolic ported BMI (Fig 2A) and 7 studies (493 ies with longer intervention periods (.6 outcomes reported by each study. participants) that reported BMI z score months)41,42 showed greater weight loss Fifteen of the 38 studies reported (Fig 2B). There was a pooled BMI re- than shorter term interventions.41–44,69 at least 1 cardio-metabolic outcome. duction of 1.25 kg/m2 (95% CI: 0.32– Four studies followed up participants at 7 22,24,26,28,31,32,34,39,41,43,46,49,51,63,68 All except 2 2.18, I2 =98%)anda0.10BMIz score months to 1 year from baseline and the small studies (with 7 to 15 participants in reduction (95% CI: 0.02–0.18, I2 =0%– pooled results indicate that weight loss each study arm)24,39 reported a positive 50%) greater for the lifestyle in- was sustained after program completion weight loss effect of lifestyle inter- tervention compared with the control (Fig 3B). Similar observations were ob- ventions compared with control groups. condition. tained for percentage body fat change, Eight studies reported blood lipids with the lifestyle intervention group los- The short-term study of children with results26,28,31,39,43,46,49,63; 6 studies re- ing 3.2% more body fat (95% CI: 1.39– thegreatestpostinterventioneffectwas ported results of fasting glucose, fast- 5.01) than the usual care group at the end a 6-month community-based program ing insulin, or HOMA-IR,22,24,26,31,46,49; 2 of active intervention (Supplemental Fig (mean BMI difference = 2.10 kg/m )in and 12 studies reported blood pres- 9A).The fat loss effect was sustained at which parents and children attended sure findings.26,28,31,32,34,41,43,46,49,51,63,68 eighteen 2-hour group education and 1 year follow-up (Supplemental Fig 9B). – exercise sessions held twice weekly in Four studies38 40,42 reported BMI z score sports centers and schools, followed change after the active intervention, with Total Cholesterol and Triglycerides by a 12-week free family swimming the pooled weight loss being 0.09 BMI Meta-analysis of 5 studies including 440 2 pass (Table 1).34 However, the greatest z score greater (0.02 to 0.15, I , 40%) participants between 8 and 16 years BMI reduction in the studies of ado- in the lifestyle intervention compared old (Supplemental Fig 10) showed that lescents (mean BMI difference 4.30 with usual care (Supplemental Fig 9C). lifestyle intervention had a significantly kg/m2) was in a 2-month community- greater impact on total cholesterol im- based intensive exercise training pro- Effects of Lifestyle Intervention provement compared with no treatment/ gram (4 times per week, 90 minutes Program Compared With Written wait-list control, usual care, or written each) combined with dietary restric- Educational Materials educational materials, both in the short- tion (500 kcal/d less than the reported Two of the 5 studies reported BMI and term (WMD 20.40 mmol/L, 95% CI: 20.51 baseline energy intake).23 3 reported BMI z score. There was to 20.30; I2 = 0%; study length: 4 to 6

PEDIATRICS Volume 130, Number 6, December 2012 e1661 Downloaded from www.aappublications.org/news by guest on September 27, 2021 FIGURE 2 Meta-analysis of studies comparing lifestyle intervention with no-treatment or wait-list controls.

e1662 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 REVIEW ARTICLE

FIGURE 3 Meta-analysis studies comparing lifestyle program to usual care or minimal intervention conditions. months)26,31,49 and the longer-term stud- Low-Density Lipoprotein and High- were found for HDL cholesterol (P = ies (WMD 20.24 mmol/L, 95% CI: 20.30 Density Lipoprotein Cholesterol .22) (Fig 5C). 2 to 20.17; I = 0%; study period: 1 to 2 Meta-analysis of 4 studies including years).28,46 The pooled intervention effect Fasting Glucose, Fasting Insulin, 372 participants with study length on triglycerides for the same group of and HOMA-IR between4and12monthsshoweda studies (Fig 5A) was 20.20 mmol/L in the Meta-analyses of 4 studies including significant improvement in LDL cho- short-term studies (95% CI: 20.35 to 2 372 participants showed a significant 2 2 2 0.05, I = 59%) and 20.09 mmol/L in the lesterol ( 0.30 mmol/L, 95% CI: 0.45 improvement in fasting insulin (255.1 2 longer-term studies (95% CI: 20.11 to 2 to 20.15, I = 59%) favoring lifestyle pmol/L, 95% CI: 271.2 to 239.1, I2 = 0.07, I2 =0%). intervention(Fig5B).Nodifferences 0%) in favor of lifestyle interventions

PEDIATRICS Volume 130, Number 6, December 2012 e1663 Downloaded from www.aappublications.org/news by guest on September 27, 2021 FIGURE 4 Meta-analysis of studies comparing lifestyle intervention program with written education materials. over 1 year (Fig 6A) and no differences outcome measures, including blood lip- importance of lifestyle interventions was found for fasting glucose (P = .08) ids, fasting glucose and insulin, HOMA-IR, incorporating a dietary component as a (Supplemental Fig 11). The pooled dif- and blood pressure, was a 12-week critical part of treatment of childhood ference for HOMA-IR was 22.32 (95% intensive school-based lifestyle inter- obesity. CI: 23.25 to 21.39) in favor of lifestyle vention program targeting 10- to 13-year- The meta-analyses indicate that life- intervention over 1 year; however, the old obese students.26 The participants style interventions incorporating a di- 2 heterogeneity was high (I = 79%) (Fig 6C). received 30-minutes of nutrition in- etary component led to significant struction twice per week at school plus Blood Pressure weight loss when compared with no 40-minutes of classroom-based non- treatment. These results support pre- Meta-analyses of 7 studies (554 par- competitive aerobic activity 3 times vious reviews,8,9,11–14 and extend the ticipants) showed that lifestyle inter- per week. The mean BMI and body evidence base on the use of lifestyle ventions led to a significantly greater fat percentage difference between interventions in the treatment of child- improvement in diastolic blood pressure the intervention group and the no- hood obesity, as this review includes in the short-term studies (WMD 21.69 treatment control at the end of active more trials, uses clearly defined no- mm Hg, 95% CI: 23.15 to 20.24, I2 = 26%, intervention was 21.5 kg/m2 and treatment, or wait-list controls, and study length: 6 months or less) but there 21.2% respectively. was no difference in the longer-term extends ascertainment to September studies(Fig7A).Onthecontrary,asig- DISCUSSION 2010. Studies comparing lifestyle inter- ventions with usual care also resulted in nificantly greater improvement in sys- This systematic review reports on life- significant immediate and posttreat- tolic blood pressure was shown only style intervention trials incorporating ment effects on BMI up to 1 year from in the studies with a study length of a dietary component aimed at treating baseline. The meta-analysis shows that 1yearormore(WMD23.72 mm Hg, overweight and obesity in children and 2 weight loss was greater when the du- 95% CI: 24.74 to 22.69, I = 0%) (Fig 7B). adolescents published between 1975 Five studies were not included in the and September 2010 (n = 38). It is the ration of treatment was longer than meta-analyses, as they reported the first review to summarize the effects 6 months. Lifestyle interventions also fi absolute values only at baseline and of lifestyle interventions on cardio- produced signi cant treatment effects follow-up.34,43,51,63,68 These studies re- metabolic outcomes in this age group on BMI and BMI z score, compared with ported a similar trend as those included and provides an improved understand- written information only, over a 6- to in the meta-analyses. ing of the effects of lifestyle interventions 12-month intervention period. The study that achieved the greatest im- on weight loss and cardio-metabolic Meta-analyses showed that lifestyle provement across all cardio-metabolic outcomes. The results support the interventions resulted in significant

e1664 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 EITISVlm 3,Nme ,Dcme 2012 December 6, Number 130, Volume PEDIATRICS

TABLE 3 Effects of Lifestyle Interventions on Anthropometric and Metabolic Outcomes (Structured by Year of Publication) Study Time of Anthropometric/Body Composition Metabolic Measurementa

Downloaded from BMI BMI z Body Fat % Waist Total LDL HDL Triglyceride Fasting Fasting Insulin SBP DBP Score Circumference Cholesterol Glucose Insulin Resistancei Lifestyle intervention compared with no-treatment or wait-list control Becque 198863 20 wk ✓b,d ✓✓✓b ✓✓✓b Rocchini 198868 20 wk ✓b,d ✓b ✓b Balagopal 200322 3mo ✓b ✓b,e ✓b, c ✓b, c www.aappublications.org/news Jiang 200528 24 mo ✓b ✓b ✓b ✓b ✓b Huang 200726 12 wk ✓b ✓b,f ✓b ✓b ✓b ✓b ✓b ✓b ✓b ✓✓ Park 200731 12 wk ✓b ✓b,f ✓b ✓b ✓b ✓✓b ✓b ✓b ✓b ✓b ✓ Davis 200924 16 wk ✓b ✓ ✓✓ Reinehr 201032 6mo ✓b ✓b ✓b,f ✓b ✓b ✓b Sacher 201034 6mo ✓b ✓b ✓g ✓b ✓✓b Lifestyle intervention compared with usual care Nemet 200543 3mo ✓b ✓b,h ✓b ✓b ✓✓ ✓✓b Nemet 200543 12 mo ✓b ✓b,h Gillis 200739 6mo ✓✓✓✓ byguest on September27, 2021 Savoye 200746 6mo ✓b ✓b,f ✓b ✓✓ ✓ ✓ ✓b ✓b ✓✓ Savoye 200746 12 mo ✓b ✓b,f ✓b ✓✓ ✓ ✓ ✓b ✓b ✓✓ Kalarchian 200941 6mo ✓b ✓b,e ✓b ✓b ✓ Kalarchian 200941 12 mo ✓✓b,e ✓b ✓b ✓ Lifestyle intervention compared with written education materials only Johnston 200749 6mo ✓b ✓f ✓b ✓b ✓✓ ✓✓ ✓✓ Weigel 200851 12 mo ✓b ✓b ✓b ✓ DBP, diastolic blood pressure; SBP, systolic blood pressure; ✓, the outcomes were reported. a Counted from baseline. b Indicates a statistically significant improvement in that outcome when compared with no-treatment control/usual care/written education materials. c Findings reported in graph. d Body fat % measured by hydrostatic weighting. e Body fat % measured by dual-energy x-ray absorptiometry. f Body fat % measured by bio-impedance analyzer. g Body fat % determined by deuterium dilution. h Body fat % determined by skin folds. i Insulin resistance determined by the homeostasis model assessment of insulin resistance (HOMA-IR). EIWARTICLE REVIEW e1665 FIGURE 5 a,Forestplotofmeandifferencesintriglyceridesconcentrations(mmol/L).B,ForestplotofmeandifferencesinLDLcholesterolconcentrations(LDL,mmol/L).C, Forest plot of mean differences in HDL concentrations (HDL, mmol/L). improvements in total cholesterol and tervention resulted in improvement in demonstrated significant improvements triglycerides up to 2 years from base- plasma lipid concentrations, insulin in triglycerides26,28,31 or HDL choles- line, as well as improvements in fasting sensitivity, and blood pressure in obese terol.26,63 High triglycerides and low insulin and HOMA-IR up to 1 year from children, even in the absence of weight HDL cholesterol levels are the impor- baseline; however, the improvements loss or body composition change.75,76 tant risk factors of cardiovascular were not uniformly associated with the The absence of individual participants’ disease. Future studies should ex- extent of weight loss or body fat re- data on weight and cardio-metabolic plore effective strategies to improve duction. It is uncertain whether the outcome changes makes it impossible triglycerides and HDL cholesterol positive effects were attributable to to characterize the relationship be- concentrations. weight loss per se or attributable to tween the extent of weight loss and aspects of the lifestyle intervention that changes in various cardio-metabolic The impact of lifestyle interventions on were independent of weight loss, such outcomes. Although most studies blood pressure is less certain from the as reduction in saturated fat intake showed a significant improvement in included studies. Most overweight or or increased physical activity. Some total cholesterol (6/7)26,28,31,43,46,49 or obese children are likely to be normo- studies have reported that lifestyle in- LDL cholesterol,26,31,43,49 fewer than half tensive. In addition, blood pressure is

e1666 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 REVIEW ARTICLE

FIGURE 6 a, Forest plot of differences in mean fasting insulin (pmol/L). B, Forest plot of mean differences in homeostasis model assessment of insulin resistance (HOMA-IR).

strongly related to age and height77–79 in targeted parents as the sole agent of cise training component.22,31,43,46,48,49,64,68 children and adolescents, and therefore change,35 encouraged parents to lose Again, the varied strategies, intensity, direct comparison of blood pressure weight if they were overweight,38,66 or and duration of intervention make it reading possibly underestimates the provided a free family swim pass to difficult to conduct direct comparisons intervention effects and limits our abil- participants.34 and to identify the most effective exer- ity to draw definitive conclusions on the cise intervention for weight loss in this effects of lifestyle interventions on blood Dietary Intervention age group. pressure. We found that dietary interventions were rarely evaluated as a sole com- Strengths and Limitations Features of Effective Interventions ponentoftreatmentin comparisonwith This review comprehensively included The heterogeneity of the included a waited-list or no-treatment control lifestyle intervention trials published studies makes it difficult to give de- group. Dietary interventions were between 1975 and 2010 during which finitive recommendations for practice. usually part of a broader lifestyle in- time childhood obesity became preva- However, the studies provide evidence tervention program. Not all studies lent. Strengths of the study include the to support a variety of dietary and adequately described the dietary in- reporting of mean differences in BMI lifestyle components in treating child- tervention. The most commonly re- and percentage body fat, weight change hood obesity across a wide range of ported dietary interventions were the indicators commonly used by clini- treatment settings, age groups, and modified Stop/Traffic Light approach cians, as well as cardio-metabolic out- severity of obesity. and a hypocaloric diet/calorie re- comes, Also, separate meta-analyses striction approach. Both dietary ap- were conducted to compare lifestyle Family Involvement proacheswere demonstrated to achieve interventions with clearly defined no- Family involvement in treatment of effective relative weight loss across treatment or wait-list controls, usual childhood obesity is widely advocated different age groups, settings, and care, or minimal advice and written diet 22,23,27,32,38,41–44,66 fl and discussed.9,80,81 Our review demon- countries. The in uen- and physical activity education mate- strated that almost all effective inter- ces on weight were sustained up to 1 rials respectively. This provides clini- 27,41–43 ventions (particularly in studies that year from baseline. cally meaningful information for future enrolled children ,12 years of age) pediatric obesity treatment service reported including a family component, Exercise Intervention planning. including separate education sessions Another frequent feature of effective stud- A number of limitations of the present for parent and child,25,27,32,41,42,44,48,49,51,66 ies is involvement of a structured exer- analyses should be acknowledged.

PEDIATRICS Volume 130, Number 6, December 2012 e1667 Downloaded from www.aappublications.org/news by guest on September 27, 2021 FIGURE 7 a, Forest plot of differences in mean diastolic blood pressure (mmcHg). B, Forest plot of differences in mean systolic blood pressure (mmcHg).

First, this review was confined to pub- studies and the lack of isolation of the absence of longer-term cardiovas- lished literature written in English; this the effects of the dietary intervention cular morbidity data. may have introduced publication bias components. In addition, there were and an overrepresentation of effective inadequate data reported to allow in- interventions. Second, a high degree of clusion of some studies in meta- CONCLUSIONS clinical and statistical heterogeneity analyses, and almost 40% of included Thebodyofresearchreviewedsuggests among the included studies means the studies (n = 19) reporting only abso- that lifestyle interventions incorpo- results should be interpreted with lute values of weight outcome. For rating a dietary component along with caution. We addressed statistical het- these studies, we calculated weight an exercise and/or behavioral therapy erogeneity by using a random effects change from absolute values and used component are effective in treating meta-analysisandbysubgroupanalysis. imputation methods to estimate the SD childhood obesity and improving the The potential sources of heterogeneity of the change. To facilitate future sys- cardio-metabolic outcomes under a wide include variations in the participant tematic reviews and meta-analyses, range of conditions at least up to 1 year. populations, the intensity and duration authors should be encouraged to re- To draw firm clinical recommendations, of interventions, and the variety of diet port both weight change and SD data. future studies should provide details of andexercise regimensused. The review Finally, the review was also limited by all intervention components, participant wasalsolimitedbythelessthanoptimal the use of intermediate outcomes, such characteristics, and the study design, methodological quality of the included as lipoprotein and blood pressure, in including the method of randomization,

e1668 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 REVIEW ARTICLE blinding, allocation concealment, and determine what magnitude of weight ACKNOWLEDGMENT attrition rates. Further work is required reduction in the pediatric population is We thank Ms Debbie Booth, librarian, not only to determine the optimal length, compatible with clinically significant ben- Faculty of Health, The University of New- intensity, and long-term effectiveness efits. Further, cost-effectiveness analyses castle, for assistance with the search- of lifestyle interventions, but also to need to be conducted. and-retrieve strategies.

REFERENCES

1. Lobstein T, Baur L, Uauy R; IASO In- 9. McGovern L, Johnson JN, Paulo R, et al. and cardiovascular risk factors. N Engl J ternational Obesity TaskForce. Obesity in Clinical review: treatment of pediatric Med. 2011;365(20):1876–1885 children and young people: a crisis in obesity: a systematic review and meta- 20. Bindler RC, Daratha KB. Relationship of public health. Obes Rev. 2004;5(suppl 1):4– analysis of randomized trials. J Clin Endo- weight status and cardiometabolic out- 104 crinol Metab. 2008;93(12):4600–4605 comes for adolescents in the TEAMS study. 2. Daniels SR, Arnett DK, Eckel RH, et al. 10. Kelly KP, Kirschenbaum DS. Immersion Biol Res Nurs. 2012;14(1):65–70 Overweight in children and adolescents: treatment of childhood and adolescent 21. Higgins JPT, Green S, eds. Cochrane Hand- pathophysiology, consequences, prevention, obesity: the first review of a promising in- book for Systematic Reviews of Inter- and treatment. Circulation. 2005;111(15): tervention. Obes Rev. 2011;12(1):37–49 ventions. Version 5.1.0 (updated March 1999–2012 11. Oude Luttikhuis H, Baur L, Jansen H, et al. 2011). The Cochrane Collaboration, 2011. 3. Weiss R. Cardiovascular Risk Clustering in Interventions for treating obesity in chil- Available at: www.cochrane-handbook.org. Obese Children. In: Debasis B, ed. Global dren. Cochrane Database Syst Rev. 2009; Accessed September 2011 Perspectives on Childhood Obesity. San (1):CD001872 22. Balagopal P, Bayne E, Sager B, Russell L, Diego: Academic Press; 2011:139–146 12. Wilfley DE, Tibbs TL, Van Buren DJ, Reach KP, Patton N, George D. Effect of lifestyle 4. Steinberger J, Daniels SR, Eckel RH, et al; Walker MS, Epstein LH. Lifestyle inter- changes on whole-body protein turnover in Progress and challenges in metabolic ventions in the treatment of childhood obese adolescents. Int J Obes Relat Metab syndrome in children and adolescents: overweight: a meta-analytic review of ran- Disord. 2003;27(10):1250–1257 a scientific statement from the American domized controlled trials. Health Psychol. 23. Ben Ounis O, Elloumi M, Zouhal H, et al. Heart Association Atherosclerosis, Hyper- 2007;26(5):521–532 Effect of individualized exercise training tension, and Obesity in the Young Commit- 13. Collins CE, Warren J, Neve M, McCoy P, combined with diet restriction on in- tee of the Council on Cardiovascular Stokes BJ. Measuring effectiveness of di- flammatory markers and IGF-1/IGFBP-3 in Disease in the Young; Council on Cardio- etetic interventions in child obesity: a sys- obese children. Ann Nutr Metab. 2010;56(4): vascular Nursing; and Council on Nutrition, tematic review of randomized trials. Arch 260–266 Physical Activity, and Metabolism. Circula- Pediatr Adolesc Med. 2006;160(9):906–922 24. Davis JN, Tung A, Chak SS, et al. Aerobic and tion. 2009;119(4):628–647 14. Collins CE, Warren JM, Neve M, McCoy P, strength training reduces adiposity in 5. National Institute for Health and Clinical Stokes B. Systematic review of inter- overweight Latina adolescents. Med Sci Excellence. NICE clinical guideline 43 Obe- ventions in the management of overweight Sports Exerc. 2009;41(7):1494–1503 sity: guidance on the prevention, identifi- and obese children which include a dietary 25. Golley RK, Magarey AM, Baur LA, Steinbeck cation, assessment and management of component. Int J Evid-Based Healthc. 2007;5 KS, Daniels LA. Twelve-month effectiveness overweight and obesity in adults and (1):2–53 of a parent-led, family-focused weight- children. London, UK:NICE; December 2006 15. Weiss R, Dziura J, Burgert TS, et al. Obesity management program for prepubertal (last updated in 2010). Available at www. and the metabolic syndrome in children children: a randomized, controlled trial. nice.org.uk/CG043. Accessed February 6, and adolescents. N Engl J Med. 2004;350 Pediatrics. 2007;119(3):517–525 2012 (23):2362–2374 26. Huang SH, Weng KP, Hsieh KS, et al. Effects 6. National Health and Medical Research 16. Freedman DS, Dietz WH, Srinivasan SR, of a classroom-based weight-control in- Council. Clinical Practice Guidelines for the Berenson GS. The relation of overweight to tervention on cardiovascular disease in Management of Overweight and Obesity in cardiovascular risk factors among children elementary-school obese children. Acta Children and Adolescents. Canberra, Aus- and adolescents: the Bogalusa Heart Study. Paediatr Taiwan. 2007;48(4):201–206 tralia: NHMRC; 2003 Pediatrics. 1999;103(6 pt 1):1175–1182 27. Janicke DM, Sallinen BJ, Perri MG, et al. 7. Logue J, Thompson L, Romanes F, Wilson 17. Reilly JJ, Methven E, McDowell ZC, et al. Comparison of parent-only vs family-based DC, Thompson J, Sattar N; Guideline De- Health consequences of obesity. Arch Dis interventions for overweight children in velopment Group. Management of obesity: Child. 2003;88(9):748–752 underserved rural settings: outcomes from summary of SIGN guideline. BMJ. 2010;340: 18. Garnett SP, Baur LA, Srinivasan S, Lee JW, project STORY. Arch Pediatr Adolesc Med. c154 Cowell CT. Body mass index and waist cir- 2008;162(12):1119–1125 8. Whitlock EP, O’Connor EA, Williams SB, Beil cumference in midchildhood and adverse 28. Jiang JX, Xia XL, Greiner T, Lian GL, TL, Lutz KW. Effectiveness of weight man- cardiovascular disease risk clustering in RosenqvistU.Atwoyearfamilybased agement interventions in children: a tar- adolescence. Am J Clin Nutr. 2007;86(3): behaviour treatment for obese children. geted systematic review for the USPSTF. 549–555 Arch Dis Child. 2005;90(12):1235–1238 Pediatrics. 2010;125(2). Available at: www. 19. Juonala M, Magnussen CG, Berenson GS, 29. Kitzman-Ulrich H, Hampson R, Wilson DK, pediatrics.org/cgi/content/full/125/2/e396 et al. Childhood adiposity, adult adiposity, Presnell K, Brown A, O’Boyle M. An adolescent

PEDIATRICS Volume 130, Number 6, December 2012 e1669 Downloaded from www.aappublications.org/news by guest on September 27, 2021 weight-loss program integrating family vari- able at: www.pediatrics.org/cgi/content/full/ 52. Balagopal P, George D, Patton N, et al. ables reduces energy intake. JAmDiet 121/2/e539 Lifestyle-only intervention attenuates the Assoc. 2009;109(3):491–496 41. Kalarchian MA, Levine MD, Arslanian SA, inflammatory state associated with obesity: 30. McCallum Z, Wake M, Gerner B, et al. Out- et al. Family-based treatment of severe a randomized controlled study in adoles- come data from the LEAP (Live, Eat and pediatric obesity: randomized, controlled cents. J Pediatr. 2005;146(3):342–348 Play) trial: a randomized controlled trial of trial. Pediatrics. 2009;124(4):1060–1068 53. Balagopal P, George D, Yarandi H, Funanage a primary care intervention for childhood 42. Kalavainen MP, Korppi MO, Nuutinen OM. V, Bayne E. Reversal of obesity-related hypo- overweight/mild obesity. Int J Obes (Lond). Clinical efficacy of group-based treatment adiponectinemia by lifestyle intervention: 2007;31(4):630–636 for childhood obesity compared with rou- a controlled, randomized study in obese 31. Park TG, Hong HR, Lee J, Kang HS. Lifestyle tinely given individual counseling. [see adolescents. J Clin Endocrinol Metab. 2005; plus exercise intervention improves meta- comment] Int J Obes (Lond). 2007;31(10): 90(11):6192–6197 bolic syndrome markers without change in 1500–1508 54. Golley RK, Perry RA, Magarey A, Daniels L. adiponectin in obese girls. Ann Nutr Metab. 43. Nemet D, Barkan S, Epstein Y, Friedland O, Family-focused weight management program 2007;51(3):197–203 Kowen G, Eliakim A. Short- and long-term for five- to nine-year-olds incorporating par- 32. Reinehr T, Schaefer A, Winkel K, Finne E, beneficial effects of a combined dietary- enting skills training with healthy lifestyle Toschke AM, Kolip P. An effective lifestyle behavioral-physical activity intervention information to support behaviour modifica- intervention in overweight children: find- for the treatment of childhood obesity. Pe- tion. Nutr Diet 2007;64(3):144–150 ings from a randomized controlled trial on diatrics. 2005;115(4). Available at: www. 55. Janicke DM, Sallinen BJ, Perri MG, Lutes LD, “Obeldicks light.” Clin Nutr. 2010;29(3):331– pediatrics.org/cgi/content/full/115/4/e443 Silverstein JH, Brumback B. Comparison of 336 44. Nemet D, Barzilay-Teeni N, Eliakim A. Treat- program costs for parent-only and family- 33. Rooney BL, Gritt LR, Havens SJ, Mathiason ment of childhood obesity in obese fami- based interventions for pediatric obesity in MA, Clough EA. Growing healthy families: lies. J Pediatr Endocrinol Metab. 2008;21 medically underserved rural settings. J family use of pedometers to increase (5):461–467 Rural Health. 2009;25(3):326–330 physical activity and slow the rate of obe- 45. Sabet Sarvestani R, Jamalfard MH, Kargar 56. Janicke DM, Sallinen BJ, Perri MG, et al. sity. WMJ. 2005;104(5):54–60 M, Kaveh MH, Tabatabaee HR. Effect of di- Sensible treatment of obesity in rural youth 34. Sacher PM, Kolotourou M, Chadwick PM, etary behaviour modification on anthropo- (STORY): design and methods. Contemp Clin et al. Randomized controlled trial of the metric indices and eating behaviour in Trials. 2008;29(2):270–280 MEND program: a family-based community obese adolescent girls. J Adv Nurs. 2009;65 57. Stewart L, Houghton J, Hughes AR, Pearson intervention for childhood obesity. Obesity (8):1670–1675 D, Reilly JJ. Dietetic management of pedi- (Silver Spring). 2010;18(suppl 1):S62–S68 46. Savoye M, Shaw M, Dziura J, et al. Effects of atric overweight: development and de- 35. Shelton D, Le Gros K, Norton L, Stanton-Cook a weight management program on body scription of a practical and evidence-based S, Morgan J, Masterman P. Randomised composition and metabolic parameters in behavioral approach. J Am Diet Assoc. controlled trial: a parent-based group edu- overweight children: a randomized con- 2005;105(11):1810–1815 cation programme for overweight children. trolled trial. [see comment] JAMA. 2007;297 58. Wake M, Gold L, McCallum Z, Gerner B, J Paediatr Child Health. 2007;43(12):799–805 (24):2697–2704 Waters E. Economic evaluation of a primary 36. Tsiros MD, Sinn N, Brennan L, et al. Cogni- 47. Estabrooks PA, Shoup JA, Gattshall M, care trial to reduce weight gain in tive behavioral therapy improves diet and Dandamudi P, Shetterly S, Xu S. Automated overweight/obese children: the LEAP trial. body composition in overweight and obese telephone counseling for parents of over- Ambul Pediatr. 2008;8(5):336–341 adolescents. Am J Clin Nutr. 2008;87(5): weight children: a randomized controlled 59. Kalavainen M, Karjalainen S, Martikainen J, 1134–1140 trial. Am J Prev Med. 2009;36(1):35–42 Korppi M, Linnosmaa I, Nuutinen O. Cost- 37. Wake M, Baur LA, Gerner B, et al. Outcomes 48. Fullerton G, Tyler C, Johnston CA, Vincent JP, effectiveness of routine and group pro- and costs of primary care surveillance and Harris GE, Foreyt JP. Quality of life in grams for treatment of obese children. intervention for overweight or obese chil- Mexican-American children following Pediatr Int. 2009;51(5):606–611 dren: the LEAP 2 randomised controlled a weight management program. Obesity 60. Nemet D, Berger-Shemesh E, Wolach B, trial. BMJ. 2009;339(7730):b3308 (Silver Spring). 2007;15(11):2553–2556 Eliakim A. A combined dietary-physical ac- 38. Díaz RG, Esparza-Romero J, Moya-Camarena 49. Johnston CA, Tyler C, Fullerton G, et al. tivity intervention affects bone strength in SY, Robles-Sardín AE, Valencia ME. Lifestyle Results of an intensive school-based weight obese children and adolescents. Int J intervention in primary care settings loss program with overweight Mexican Sports Med. 2006;27(8):666–671 improves obesity parameters among Mex- American children. Int J Pediatr Obes. 2007; 61. Shaw M, Savoye M, Cali A, Dziura J, ican youth. J Am Diet Assoc. 2010;110(2): 2(3):144–152 Tamborlane WV, Caprio S. Effect of a suc- 285–290 50. Resnick EA, Bishop M, O’Connell A, et al. The cessful intensive lifestyle program on insulin 39. Gillis D, Brauner M, Granot E. A community- CHEER study to reduce BMI in elementary sensitivity and glucose tolerance in obese based behavior modification intervention school students: a school-based, parent- youth. Diabetes Care. 2009;32(1):45–47 for childhood obesity. J Pediatr Endocrinol directed study in Framingham, Massachu- 62. Johnston CA, Tyler C, McFarlin BK, et al. Metab. 2007;20(2):197–203 setts. J Sch Nurs. 2009;25(5):361–372 Weight loss in overweight Mexican Ameri- 40. Hughes AR, Stewart L, Chapple J, et al. 51. Weigel C, Kokocinski K, Lederer P, Dötsch J, can children: a randomized, controlled Randomized, controlled trial of a best- Rascher W, Knerr I. Childhood obesity: trial. Pediatrics. 2007;120(6). Available at: practice individualized behavioral program concept, feasibility, and interim results of www.pediatrics.org/cgi/content/full/120/6/ for treatment of childhood overweight: a local group-based, long-term treatment e1450 Scottish Childhood Overweight Treatment program. J Nutr Educ Behav. 2008;40(6): 63. Becque MD, Katch VL, Rocchini AP, Marks Trial (SCOTT). Pediatrics. 2008;121(3). Avail- 369–373 CR, Moorehead C. Coronary risk incidence

e1670 HO et al Downloaded from www.aappublications.org/news by guest on September 27, 2021 REVIEW ARTICLE

of obese adolescents: reduction by exercise based treatment for childhood obesity. 77. Falkner B, Gidding SS, Portman R, Rosner B. plus diet intervention. Pediatrics. 1988;81 Health Psychol. 1994;13(5):373–383 Blood pressure variability and classifica- (5):605–612 71. Kirschenbaum DS, Harris ES, Tomarken AJ. tion of prehypertension and hypertension 64. Botvin GJ, Cantlon A, Carter BJ, Williams CL. Effects of parental involvement in behavioral in adolescence. Pediatrics. 2008;122(2): Reducing adolescent obesity through weight loss therapy for preadolescents. 238–242 a school health program. J Pediatr. 1979;95 Behav Ther. 1984;15(5):485–500 78. National High Blood Pressure Education (6):1060–1063 72. Braet C, Van Winckel M, Van Leeuwen K. Program Working Group on High Blood 65. Braet C, Van Winckel M. Long-term follow-up of Follow-up results of different treatment Pressure in Children and Adolescents. The a cognitive behavioral treatment program for programs for obese children. Acta Pae- fourth report on the diagnosis, evaluation, obese children. Behav Ther. 2000;31(1):55–74 diatr. 1997;86(4):397–402 and treatment of high blood pressure in 66. Epstein LHL, Wing RRR, Koeske RR, Valoski 73. Epstein LH, McCurley J, Wing RR, Valoski A. children and adolescents. Pediatrics. 2004; AA. Effects of diet plus exercise on weight Five-year follow-up of family-based behav- 114(2 suppl 4th report):555–576 change in parents and children. J Consult ioral treatments for childhood obesity. J 79. Rosner B, Prineas R, Daniels SR, Loggie J. Clin Psychol. 1984;52(3):429–437 Consult Clin Psychol. 1990;58(5):661–664 Blood pressure differences between blacks 67. Mellin LM, Slinkard LA, Irwin CE Jr. Ado- 74. Rocchini A, Katch V, Schork A, Kelch R. In- and whites in relation to body size among lescent obesity intervention: validation of sulin and blood pressure during weight US children and adolescents. Am J Epi- the SHAPEDOWN program. J Am Diet Assoc. loss in obese adolescents. Hypertension. demiol. 2000;151(10):1007–1019 1987;87(3):333–338 1987;10(3):267–273 80. Golan M, Kaufman V, Shahar DR. Childhood 68. Rocchini AP, Katch V, Anderson J, et al. 75. Bell LM, Watts K, Siafarikas A, et al. Exercise obesity treatment: targeting parents ex- Blood pressure in obese adolescents: effect alone reduces insulin resistance in obese clusively v. parents and children. Br J Nutr. of weight loss. Pediatrics. 1988;82(1):16–23 children independently of changes in body 2006;95(5):1008–1015 69. Saelens BE, Sallis JF, Wilfley DE, Patrick K, composition. J Clin Endocrinol Metab. 2007; 81. Shrewsbury VA, Steinbeck KS, Torvaldsen S, Cella JA, Buchta R. Behavioral weight con- 92(11):4230–4235 Baur LA. The role of parents in pre- trol for overweight adolescents initiated in 76. Watts K, Jones TW, Davis EA, Green D. Ex- adolescent and adolescent overweight primary care. Obes Res. 2002;10(1):22–32 ercise training in obese children and ado- and obesity treatment: a systematic review 70. Epstein LH, Valoski A, Wing RR, McCurley J. lescents: current concepts. Sports Med. of clinical recommendations. Obes Rev. Ten-year outcomes of behavioral family- 2005;35(5):375–392 2011;12(10):759–769

PEDIATRICS Volume 130, Number 6, December 2012 e1671 Downloaded from www.aappublications.org/news by guest on September 27, 2021 Effectiveness of Lifestyle Interventions in Child Obesity: Systematic Review With Meta-analysis Mandy Ho, Sarah P. Garnett, Louise Baur, Tracy Burrows, Laura Stewart, Melinda Neve and Clare Collins Pediatrics 2012;130;e1647 DOI: 10.1542/peds.2012-1176 originally published online November 19, 2012;

Updated Information & including high resolution figures, can be found at: Services http://pediatrics.aappublications.org/content/130/6/e1647 References This article cites 73 articles, 18 of which you can access for free at: http://pediatrics.aappublications.org/content/130/6/e1647#BIBL Subspecialty Collections This article, along with others on similar topics, appears in the following collection(s): Endocrinology http://www.aappublications.org/cgi/collection/endocrinology_sub Nutrition http://www.aappublications.org/cgi/collection/nutrition_sub Obesity http://www.aappublications.org/cgi/collection/obesity_new_sub Permissions & Licensing Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.aappublications.org/site/misc/Permissions.xhtml Reprints Information about ordering reprints can be found online: http://www.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news by guest on September 27, 2021 Effectiveness of Lifestyle Interventions in Child Obesity: Systematic Review With Meta-analysis Mandy Ho, Sarah P. Garnett, Louise Baur, Tracy Burrows, Laura Stewart, Melinda Neve and Clare Collins Pediatrics 2012;130;e1647 DOI: 10.1542/peds.2012-1176 originally published online November 19, 2012;

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://pediatrics.aappublications.org/content/130/6/e1647

Data Supplement at: http://pediatrics.aappublications.org/content/suppl/2012/11/14/peds.2012-1176.DCSupplemental

Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. Pediatrics is owned, published, and trademarked by the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2012 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.

Downloaded from www.aappublications.org/news by guest on September 27, 2021