Effectiveness and process evaluation in obesity and type 2 diabetes prevention programs in children: a systematic review and meta-analysis

RESEARCH ARTICLE Open Access E ec i eness and p ocess e alua ion in obesi y and ype 2 diabe es p e en ion p og ams in child en: a sys ema ic e iew and me a-analysis M. Se al-Co es 1,2 , P. De Miguel-E ayo 1,3* , P. Zapa a 1,4 , M. L. Miguel-Be ges 1,2 and L. A. Mo eno 1,2,3 Abs ac Backg ound: Obesi y in child en is one o he mos se e e public heal h challenges o he cu en cen u y and Type 2 Diabe es Melli us (T2DM) equency is also escala ing. Mo e so, he impo ance o p ocess e alua ion (PE) in complex in e en ions is inc easingly ecognized. The p esen e iew, aims o iden i y he e ec i eness in e ms o body composi ion pa ame e s in a gene a ion o a icles o p e en obesi y and T2DM in child en. We hypo hesise ha hose s udies epo ing PE applying he la es implemen a ion guidelines sugges ed by he esea che s would po en ially show posi i e changes in body composi ion compa ed o hose no epo ing i . Addi ionally, we will e alua e he implemen a ion deg ee o PE in hose a icles conside ing i and desc ibe he PE subcomponen s. Las ly, we aim o assess he in e en ion a ge used and i s esul s. Me hods: A li e a u e e iew was pe o med in pa allel by 2 independen e iewe s. A inal numbe o 41 s udies we e selec ed o inclusion c i e ia. Resul s: Me a-analysis o BMI and zBMI ound non-signi ican e ec s o he p oposed in e en ions. Sub-g oup analysis e ealed only a signi ican e ec in s udies which pe o med PE. Mo eo e , PE was epo ed in 42% e ec i e s udies and 57% non-e ec i e s udies. Fideli y and sa is ac ion we e he mos implemen ed PE subcomponen s, al hough he e was a gene ally low g ade o PE use (7/41). The highes p opo ion o e ec i eness (83%) was shown in in e en ions o physical ac i i y alone while he in e en ion mos used was 3-a m a ge (die , PA and BS). (Con inued on nex page) © The Au ho (s). 2021 Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle's C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle's C ea i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed in a c edi line o he da a. * Co espondence: pila dm@uniza .es 1 G ow h, Exe cise, NU i ion and De elopmen (GENUD) Resea ch G oup, Ins i u o Ag oalimen a io de A agón (IA2), Ins i u o de In es igación Sani a ia, A agón (IIS A agón), Uni e sidad de Za agoza, 50009 Za agoza, Spain 3 CIBER Fisiopa ología de la Obesidad y Nu ición (CIBERobn), Ins i u o de Salud Ca los III, Mad id, Spain Full lis o au ho in o ma ion is a ailable a he end o he a icle Se al-Co es e al. BMC Public Heal h (2021) 21:348 h ps://doi.o g/10.1186/s12889-021-10297-8 (Con inued om p e ious page) Conclusions: O e all, obesi y and T2DM p e en ion s udies included in his e iew a e no e ec i e in e ms o BMI and zBMI. Those s udies pe o ming PE epo ed o be e ec i e in e ms o BMI, while s udies no epo ing PE did no ha e posi i e esul s in e ms o BMI and zBMI. In addi ion, none o he in e en ion s udies included all PE indica o s and mos s udies, which included PE in hei in e en ions, did no p o ide ull epo o he PE componen s, acco ding o he guidelines used o he p esen e iew. PROSPERO egis a ion numbe : CRD42018093667. Keywo ds: P ocess e alua ion, Obesi y, Type 2 diabe es melli us, Body composi ion, Heal h plan implemen a ion, P ima y p e en ion Backg ound Obesi y has impo an e ec s on heal h in he sho , medium and long e ms whe e me abolic complica ions a e common [1]. In child en, o e weigh and obesi y’s p e alence has inc eased in e e y con inen o he wo ld in ecen decades [2]. Cu en ly o e 1.5 billion adul s and 170 million child en a e o e weigh o obese [3]. Obesi y in child en is one o he mos se e e global pub- lic heal h challenges o he cu en cen u y [4]. Type 2 diabe es melli us (T2DM) equency is apidly escala ing, wi h i al implica ions o mo bidi y and mo ali y. Wo ldwide p e alence igu es o T2DM in 2019 es i- ma es 463 million (9.3%) people a ec ed, escala ing o 578 million (10.2%) by 2030 and 700 million (10.9%) by 2045 [5]. As a esul o he inc easing a es o obesi y and seden a y li es yle, younge popula ions a e p omp ly diagnosed wi h T2DM [6] co esponding o he ex ensi e sp ead o childhood obesi y [7]. Li es yle in e en ion p og ams o he ea men o child en and adolescen s wi h obesi y had a long- e m success a e lowe han 10% [8]. The ine iciency o p e- en ion p og ams o li es yle in e en ion comp ehends se e al ac o s including lack o mul idisciplina y ap- p oaches, inc easing men al heal h ela ed issues o in- su icien in ol emen o he pa en s, which migh in e ac wi h he p og am’s adhe ence [9]. As T2DM is one o he mos equen me abolic complica ions o obesi y wi h impo an long- e m e ec s, combined ac- ions o p e en bo h obesi y and T2DM could be de eloped. A numbe o esea ch ac i i ies aim o build and e alua e e idence-based p og ams o p e en childhood obesi y [10]. The school is usually ega ded as a sui able and e ec i e se ing o ca y ou obesi y p e en ion p o- g ams aiming o e alua e s uden s’ene gy balance- ela ed beha io s (EBRBs) [11]. Howe e , mos o sys- ema ic e iews conduc ed in Uni ed s a es and Eu ope o school-based in e en ions p e en ing obesi y, p o- mo e physical ac i i y (PA) and dec ease seden a y be- ha io s, show mode a e e idence o e ec i eness [12, 13]. Mos o he obesi y p e en ion p og ams a e di i- cul o implemen and e alua e due o he mul iple in e ac ing componen s ha exis . Randomized con- olled ials (RCT) o hese p og ams a e o en known o i s di icul ies o ind ou he eason why he p o- g am wo ked o did no wo k wi hou examining unde - lying p ocesses [11]. Wi h public heal h depending on he impac o hese p og ams and hei implemen a ion in p ac ice, i is essen ial o in e p e whe he a p og am was implemen ed as in ended, o wha ex end and how hese concep s could modula e he e ec i eness o he p og am [14]. Ve i ying ha he in e en ions a e deli - e ed as planned as well as ac o s a ec ing implemen a- ion, allows an accu a e in e p e a ion o in e en ion ou comes and con ex by he esea che s and policy make s in o de o op imize u he implemen a ion o in e en ions in he u u e [15]. The impo ance o p ocess e alua ion (PE) in public heal h in e en ion esea ch is g adually ecognized [16]. The use o PE is o obse e and eco d p og am imple- men a ion as well as helping o unde s and he connec- ion be ween speci ic p og am elemen s and p og am ou comes. Se e al p ac ical amewo ks and models a e a ailable o lead di e en p o essionals o he de elop- men o an e alua ion plan wi h wide scope, including PE. The e a e comp ehensi e and sys ema ic app oaches o de eloping a PE plan o assess he implemen a ion o a pa icula p e en ion p og am in e en ion. These ap- p oaches a e di ided in di e en indica o s including e- c ui men , each, ideli y, dose and sa is ac ion [16]. Howe e , he e is no ag eemen on wha is he ideal s anda d o classi y he s udy o implemen a ion in o key pa s, o ins ance dose and each, and i is no possible o p oduce a de ini e s anda d among he a ious ame- wo ks used a p esen . Wi hou he p esence o PE, i is challenging o di e en ia e be ween ou comes ha a e, in heo y, ela ed o a lack o ideli y and hose ones a - ibu ed o he incompe ence o he in e en ion o achie e he expec ed esul s. Unsuccess ul a emp s o pe o m an in e en ion as in ended p omp s o mis- leading esul s and conclusions abou he e ec i eness o he in e en ion and is conside ed no alid o epli- ca e o use in u u e in es iga ions in he scien i ic com- muni y. Mul icomponen p e en ion p og ams a e Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 2 o 17 complex in e en ions, designed o wo k syne gis ically. Fo his eason, PE con ibu es o in e p e complex ou - come e ec s and helps o unde s and he analysis o he in e en ion deli e ed [17,18]. RCTs a e conside ed he bes s udy design o es ablish he e ec i eness o in e en ions wi h ce ain deg ee o complexi y. Howe e , he e is no in o ma ion a p esen o how an in e en ion migh be ep oduced in a pa - icula con ex , o whe he ial ou comes will be epli- ca ed. To ou knowledge, he e is no sys ema ic e iew and me a-analysis ocusing on he e alua ion o e ec - i eness and he de elopmen and epo o PE o in e - en ions in heal h p og ams p e en ing obesi y and T2DM in child en. The p esen e iew, aims o iden i y he e ec i eness in e ms o body composi ion pa ame- e s in a gene a ion o a icles o p e en obesi y and T2DM in child en. We hypo hesise ha hose s udies epo ing PE applying he la es implemen a ion guide- lines sugges ed by he esea che s [19,20] would po en- ially be mo e e ec i e in e ms o changes in body composi ion compa ed o hose no epo ing i . Add- i ionally, we will e alua e he implemen a ion deg ee o PE in hose a icles conside ing i and desc ibe he PE subcomponen s. Las ly, we aim o assess he in e en ion a ge used and i s esul s. Me hods Li e a u e sea ch A li e a u e e iew was pe o med in pa allel by 2 inde- penden e iewe s and a hi d independen e iewe was in ol ed when inconsis ency o disag eemen wi h he selec ion o a icles was iden i ied. The p o ocol was de- eloped acco ding o he p e e ed epo ing i ems o sys ema ic e iews and me a-analysis (PRISMA) guide- lines adap ed o he design o he p esen s udy [21]. Mo eo e , he sys ema ic li e a u e sea ch was egis e ed on he In e na ional P ospec i e Regis e o Sys ema ic Re iews (PROSPERO, egis a ion numbe : CRD42018093667). An in-dep h sea ch o elec onic da- abases was conduc ed in he PubMed, Scopus and Embase. (Mesh®) e ms we e used du ing he sea ch s a egy in PUBMED, based on medical subjec headings and ex wo ds o pee pape s iden i ied. Sea ch e ms and ex wo ds a e desc ibed in ull epo as ollows: ((((((“Diabe es Melli us, Type 2”[Mesh]) OR “Risk”[- Mesh]) OR (“Obesi y”[Mesh] OR “Pedia ic Obesi y”[- Mesh] OR “Obesi y, Abdominal”[Mesh] OR “Obesi y Managemen ”[Mesh])) AND “Heal h Plan Implemen a- ion”[Mesh]) OR (“p e en ion and con ol”[Subheading] OR “P ima y P e en ion”[Mesh])) OR (“Ou come and P ocess Assessmen (Heal h Ca e)”[Mesh] OR “P ocess Assessmen (Heal h Ca e)”[Mesh])) AND “Body Compo- si ion”[Mesh])))))). The e e ence lis s o all included pa- pe s we e doublechecked o iden i y po en ial missing a icles ha migh ha e been missed du ing he ini ial sea ch. The ocus was on he s udies assessing he e ec - i eness on changes in body composi ion: zBMI (body mass index z sco e) and/o BMI (body mass index) and/ o wais ci cum e ence as well as he conside a ion o any o he PE subcomponen s in he heal h p og am in e en ion. O he isk ac o s such as high and inc eased blood p essu e, high and inc eased blood glucose le el, insulin le el, a - ee mass, pe cen age o and oid mass and pe - cen age obesi y a we e also conside ed in he selec ion o a icles as seconda y ou comes. A icles we e also conside ed i any o he p ima y ou comes we e e e ed as seconda y ou come and ice e sa. Da a ex ac ion and quali y assessmen The sys ema ic selec ion p ocess was pe o med in 3 phases (Fig. 1). Final esul s a e p esen ed in he desc ip- ion o pape s. The inclusion c i e ia we e 1) p esence o obesi y and/o T2DM pa ame e s as p ima y o second- a y ou comes, 2) die , physical ac i i y (PA) and beha - io al suppo (BS) alone o combined wi h o he kind o in e en ion, 3) popula ion child en age 6–12 yea s old, 4) w i en in English, 5) published om 2008 and 6) ex- clusi ely andomized con ol ials. Any disc epancies wi h he inclusion c i e ia be ween e iewe s was dis- cussed o each a common inal consensus. As a as exclusion c i e ia is conce ned i was ex- cluded om he e iew 1) clinical popula ions (diso de s, disabled, ins i u ionalized) o p esence o Type 1 dia- be es melli us and o he ypes o diabe es (E.g. insulin- dependen -diabe es, p egnancy diabe es, ges a ional dia- be es), 2) no in e en ion applied, 3) child en younge han 6 yea s and olde popula ions han 12 yea s old, 4) no w o e in English language, 5) published be o e 2008 and 6) non andomized con ol ials and unpublished s udies. S udies we e classi ied depending on an es ablished p ocess acco ding o he QUALYST (S anda d Quali y Assessmen C i e ia o e alua ing p ima y esea ch pa- pe s om a a ie y o ields) checklis o measu ing quali y by 2 independen e iewe s. The men ioned checklis has 14 ques ions which ha e o be answe ed wi h “yes”,“pa ial”,“no”o “no applicable”depending on he quali y o each a icle. The summa y sco e is he o al o he accumula ed answe s ansc ibed in o a numbe be ween 0 and 1 indica ing he quali y o each publica ion, being 1 he highes possible esul . When compa ing he me hodological sco e be ween he 2 e- iewe s, a hi d e iewe in e ened in he e en o a nu- me ical di e ence o mo e han 0.2 wi hin he same publica ion e alua ed. The Coch ane Collabo a ions Tool o assessing isk o bias in andomized ials [22] and he Coch ane Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 3 o 17 Fig. 1 PRISMA low cha o sea ch s a egy p ocess Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 4 o 17 Handbook o Sys ema ic Re iews o In e en ions [23] we e used o assess me hodological isk o bias o an- domized con ol ials, which ecommend he explici epo ing o he ollowing domains: andom sequence gene a ion, alloca ion sequence concealmen , blinding (pa icipan s and pe sonnel), blinding (ou come assess- men ), comple eness o ou come da a, selec i e epo ing and o he sou ces o bias. Each i em was gi en a sco e as being a high, low o unclea isk o bias as pe c i- e ia p o ided [23]. Me a-analysis Wi hin included e iews, a me a-analysis o 30 s udies epo ing BMI and 16 s udies epo ing zBMI in an in e en ion popula ion e sus a compa a o popula ion we e unde aken. Me a-analysis o subg oups acco ding o PE epo ing was pe o med in o de o iden i y dis- pa i ies in s udies be ween he 2 g oups including PE o no and he epo o i s e ec i eness. Wi hin included s udies o me a-analysis, G eening e al. [24] and Kala- ainen e al. [25] we e no included on BMI and Lison e al. [26] was no conside ed as no mee ing he inclu- sion c i e ia o BMI and zBMI. S anda dized mean di e ence was he app op ia e me ic o he da a ype. The in e en ions compa ed in me a-analysis we e changes in BMI and zBMI in he in e en ion g oup e sus changes in BMI and zBMI in he con ol g oup. Sub-g oup analysis we e ca ied ou in each ou come s udying PE as ac o co- a ia e o obse e po en ial di e ences be ween he g oups implemen ing PE e sus hose no implemen - ing i . Da a on mean di e ence in BMI and zBMI be ween in e en ion and compa a o g oups and s anda d de i- a ion o he di e ence om s udies ha epo ed da a in a compa able way we e analyzed in OpenMe aAnalys so wa e using in e se a iance andom-e ec s me a- analysis. Con inuous andom-e ec s De Simonian-Lai d analysis we e selec ed o e lec di e en s udy g oups, se ing, and age among he included s udies. The con idence le el used was 95.0. I2 s a is ic was used o assess he he e ogenei y o he s udies [23]. This s a is ic explains he a iance wi hin s udies as a p opo - ion o he o al a iance. < 25% alue showed low he - e ogenei y, 25 o 50% alue showed mode a e he e ogenei y, > 50 o 75% alue showed high he e ogen- ei y and > 75% alue indica ed e y high he e ogenei y. Associa ed p- alues we e also displayed, and signi icance le el was se a p< 0.05, showing he e ogenei y when p- alues we e below 0.05. P ocess e alua ion indica o s and c i e ia The p esen sys ema ic e iew has e alua ed he PE implemen a ion based on he guidelines p o ided by Saunde s e al. [19] and Moo e e al. [20]. In sho , bo h guidelines sha e he ele ance o assessing idel- i y, dose and each indica o s. Saunde s e al. p o ides a mo e de ailed lis o indica o s and i s use, consid- e ing ec ui men and con ex and 2 dose ca ego ies (deli e ed and ecei ed). Moo e e al. desc ibed a amewo k o PE buil on 3 hemes desc ibed in he 2008 MRC guidance (implemen a ion, mechanisms and con ex ) [27]. When applying he c i e ia o hese guidelines in ou a icles, we obse ed ha ideli y was conside ed mainly om he ca egi e s eedback o assess he ex en o he in e en ion implemen a- ion acco ding o he ini ial s udy p og am; mo eo e , dose was epo ed o see he mode o he p og am deli e y in e ms o aining, in e en ion compo- nen s, ma e ials and con en h ough con ol sessions om he s a (deli e ed), and also in e ms o use and eac ion o child en and pa en s o he ac i i ies deli e ed h ough ques ionnai es ( ecei ed). Finally, each e alua ed he pa icipan s a endance and o as- sess he p og am’s e ec on he a ge ed g oup, also h ough ques ionnai es. Howe e , he e migh be di - e en e ec s depending on which con ex he in e - en ion is pe o med. Resul s Desc ip ion o pape s The selec ion p ocess is displayed in Fig. 1. In summa y, he sc eening p ocess was di ided in 3 s ages: iden i ica- ion, eligibili y and inclusion. Fi s , a e dele ing 24 du- plica es, iden i ica ion s age le 273 a icles o inclusion. A e i le and abs ac sc eening, 46 a icles we e included. Finally, he inclusion s age showed, a e ull ex eading, a inal 41 (15%) a icles [24–26,28–65] which main cha ac e is ics a e summa ized in Table 1. F om he selec ed a icles, 39 o hem ocused on obesi y and 2 a icles on T2DM, al hough 7 a icles om obesi y included glucose and insulin le els in hei main e- sea ch pa ame e s. E ec i eness in body composi ion pa ame e s Table 2shows he e ec i e pa ame e s whe e some deg ee o imp o emen was epo ed. The analysis shows 26/41 (63%) s udies epo ed o be e ec i e; F om hose 26 e ec i e a icles, 3/26 (11%) included PE in hei in e en ions, meaning ha , a leas , 1 o he PE subcomponen s has been used and e- po ed as pa o he PE implemen a ion. The mos epea ed s udied ou comes when epo ing e ec i e- ness we e BMI and zBMI alone (12/26) o combined o one ano he o wi h o he body composi ion pa- ame e s (9/26). Tha is 81% o he o al e ec i e a icles. Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 5 o 17 Table 1 Main cha ac e is ics o he included s udies Au ho /Yea Coun y Sample Size Gende Mean Age Se ing In e en ion du a ion C oke e al. 2012 [28] UK 72 No epo ed 10.3 ± 1.6 Hospi al 12 mon hs Danielsen e al. 2011 [29] No way 49 24 male; 25 emale 10.6 ± 1.2 Family based/p ima y ca e 12 mon hs Da is e al. 2012 [30] USA 222 93 male; 129 emale 9.4 ± 1.1 School 10–15 weeks Ea he e al. 2013 [31] Aus alia 213 108 male; 105 emale 10.7 ± 0.6 School 8 weeks Elde e al. 2013 [32] UK 541 243 male; 298 emale 6.6 ± 0.7 Rec ea ion Cen e 2 yea s Foley e al. 2016 [33] New Zealand 251 142 male; 109 emale 11.25 Family based/Home 6 mon hs Ge a ds e al. 2015 [34] Ne he lands 86 38 male; 48 emale 7.2 ± 1.4 School 14 weeks Hollis e al. 2016 [35] Aus alia 1150 male 543; emale 583 12.0 School 24 mon hs Jones e al. 2015 [36] Aus alia 37 male 20; emale 17 9.7 ± 0.8 School 7 mon hs Kala chian e al. 2009 [37] USA 192 83 male; 109 emale 10.2 ± 1.2 Medical Cen e 18 mon hs Ko acs e al. 2009 [38] Hunga y 51 23 male; 28 emale 9.9 ± 1.3 School 15 weeks La sen e al. 2016 [39] Denma k 115 51 male; 64 emale 12.0 ± 0.4 Day camp 1 yea Li e al. 2010 [40] China 4700 2242 emale; 2458 male 9.3 ± 0.7 School 2 yea s Lison e al. 2012 [26] Spain 110 no epo ed 11.9 ± 2.2 Hospi al 6 mon hs Maddison e al. 2011 [41] New Zealand 322 male 235; emale 87 11.6 ± 1.1 School 6 mon hs Maddison e al. 2012 [42] New Zealand 322 male 235; emale 87 11.6 ± 1.1 School 24 weeks Maddison e al. 2014 [43] New Zealand 251 male 142; 109 emale 11.2 School/Communi y cen e 24 weeks Magnusson e al. 2012 [44] Iceland 321 173 male; emale 148 7.3 ± 0.3 School 2 yea Mon ei o e al. 2015 [45] B azil 32 16 emale; 18 male 11.0 ± 1.5 Family based/Communi y 20 weeks Neme e al. 2008 [46] Is ael 22 14 emales; 8 males 10.2 ± 0.5 Heal h cen e and nu i ional clinic 3 mon hs Neme e al. 2011 [47] Is ael 795 437 male; 358 emale 5.2 ± 0.3 Kinde ga ens 12 weeks Nowicka e al. 2009 [48] Sweden 76 40 male; 36 emale 10.5 Obesi y Clinic 6 mon hs Sa die e al. 2013 [49] Mexico 830 415 male; 415 emale 9.7 ± 0.7 School 18 mon hs Simon e al. 2008 [51] F ance 954 527 males; 527 emale 11.7 ± 0.7 School 4 yea s Sighn e al. 2009 [50] Ne he lands 1108 549 male, 559 emale 12.7 School 20 mon h Wa e s e al. 2018 [52] Aus alia 2965 No epo ed 8.5 School 3.5 yea s Williamson e al. 2012 [53] USA 2060 male 854; Female 1206 10.5 ± 1.2 School 28 mon hs Yackobo i ch Ga an e al. 2009 [54] Is ael 162 81 male; 81 emale 8.3 ± 1.6 Medical Cen e 12 weeks Yin e al. 2012 [55] USA 574 270 male; 304 emale 8.7 ± 0.5 School 3 yea s Casazza e al. 2012 [56] USA 26 26 emale 12.4 ± 0.3 Family based 16 weeks Copeland e al. 2013 [57] USA 699 No epo ed 13.5 No epo ed 24 mon hs G eening e al. 2011 [24] USA 450 234 male; 216 emale 8.34 ± 1.3 School 9 mon hs Gu in e al. 2008 [58] USA 206 97 male; 109 emale 8.5 ± −0.6 School 3 yea Johns on e al. 2009 [59] USA 60 33 male; 27 emale 12.3 ± 0.7 School 2 yea s Kala ainen e al. 2011 [25] Finland 70 28 male; 42 emale 8.1 ± 0.8 Family o ien ed 6 mon hs Khan e al. 2014 [60] USA 220 117 male; 103 emale 8.8 ± 0.5 School 9 mon h Lau e al. 2014 [61] Hong Kong 48 36 male; 12 emale 10.4 ± 0.9 School 6 weeks Meye e al. 2014 [62] Swi ze land 289 153 male; 136 emale 10.6 ± 0.3 School 3 yea Ma ild e al. 2012 [63] Sweden 66 28 male; 38 emale 10.8 ± 1.1 School 12 mon h Rush e al. 2011 [64] New Zealand 1348 686 male; 662 emale 7.5 School 2 yea s Velazquez Lopez e al. 2014 [65] Mexico 49 23 male; 26 emale 11.2 ± 2.7 Family medicine Uni 16 weeks Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 6 o 17 Table 2 Type o in e en ion applied, s udied ou come and e ec i eness In e en ion applied S udied Ou come E ec i e E ec i e Ou come PA,D and BS PA and BS PA and D D and BS D PA BS C oke e al. 2012 [28]√BMI, WC, %BF X Danielsen e al. 2011 [29]√BMI √BMI Da is e al. 2012 [30]√zBMI, %BF, BG, IL √zBMI, %BF, BG, IL Ea he e al. 2013 [31]√zBMI, BMI √zBMI, BMI Elde e al. 2013 [32]√zBMI,BMI,WC,%BF X Foley e al. 2016 [33]√zBMI,BMI X Ge a ds e al. 2015 [34]√zBMI,WC X Hollis e al. 2016 [35]√zBMI,BMI √zBMI,BMI Jones e al. 2015 [36]√zBMI,BMI,WC,%BF √zBMI,BMI,WC,%BF Kala chian e al. 2009 [37]√BMI,WC √BMI, WC Ko acs e al. 2009 [38]√BMI,WC,BP √BMI,WC,BP La sen e al. 2016 [39]√zBMI,BMI,WC,BG √zBMI,BMI,WC BG Li e al. 2010 [40]√zBMI,BMI,%BF √zBMI,BMI,%BF Lison e al. 2012 [26]√zBMI,BMI,WC,%BF √zBMI,BMI, WC, %BF Maddison e al. 2011 [41]√zBMI,BMI,WC,%BF √zBMI,BMI,WC,%BF Maddison e al. 2012 [42]√BMI,%BF √BMI,%BF Maddison e al. 2014 [43]√zBMI,BMI,WC,%BF X Magnusson e al. 2012 [44]√BMI,WC,%BF X Mon ei o e al. 2015 [45]√%BF,FFM,%AF √%BF,FFM;%AF Neme e al. 2008 [46]√BMI,%BF √BMI, %BF Neme e al. 2011 [47]√BMI X Nowicka e al. 2009 [48]√zBMI,BMI X Sa die e al. 2013 [49]√BMI √BMI Simon e al. 2008 [51]√BMI √BMI Sighn e al. 2009 [50]√BMI,WC X Wa e s e al. 2018 [52]√zBMI,BMI X Williamson e al. 2012 [53]√zBMI,%BF √%BF Yackobo i ch Ga an e al. 2009 [54]√BMI,%BF √BMI,%BF Yin e al. 2012 [55]√%BF √%BF Casazza e al. 2012 [56]√zBMI,BMI,%BF,BG X Copeland e al. 2013 [57]√BMI,WC,%BF X G eening e al. 2011 [24]√BMI,WC,%BF √BMI,%BF Gu in e al. 2008 [58]√%BF √%BF Johns on e al. 2009 [59]√zBMI,BMI √zBMI,BMI Kala ainen e al. 2011 [25]√BMI, WC √BMI,WC Khan e al. 2014 [60]√zBMI,BMI,%BF √zBMI,BMI,%BF Lau e al. 2014 [61]√BMI X Meye e al. 2014 [62]√%BF X Ma ild e al. 2012 [63]√BMI √BMI Rush e al. 2011 [64]√BMI,%BF,BP X Velazquez Lopez e al. 2014 [65]√BMI,%BF,BG √BMI,%BF,BG *BMI body mass index, *BP Blood p essu e, *BG Blood glucose, *FFM a - ee mass, *IL insulin le els, *WC wais ci cum e ence *%BF pe cen age o body a , *%AF pe cen age o and oid a Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 7 o 17 zBMI and BMI me a-analysis and sub-g oups analysis conside ing PE Figu e 2shows he o e all s udy esul s and plo he glo- bal e ec o changes in BMI. Figu e 3shows he sub- g oup s udies acco ding o he pe o mance o PE. Fig- u es 4and 5show he same o e all and sub-g oup ana- lysis, in his case, wi h zBMI. Me a-analysis o he 30 s udies which epo ed changes om baseline o ollow up in BMI ound non-signi ican e ec s be ween con ol and in e en ion g oups (O e all mean di e ence in BMI: −0.055; 95% CI, −0.116 o 0.006). The esul s main ained he e y high he e ogenei y in BMI s udies (I2 = 90.27%, p< 0.001). Sub-g oup analysis o zBMI e- sul s showed signi icance when compa ing s udies in- cluding PE (−0.301 (−0.531, −0.071)) e sus no PE (0.064 (−0.086, 0.214)). He e ogenei y wi hin s udies was e y high (I2=. 90.27%; p< 0.001). Me a-analysis o he 16 s udies which epo ed changes om baseline o ollow up in zBMI ound non-signi ican e ec be ween s udies (O e all mean di e ence in zBMI: −0.055; 95% CI, −0.116 o 0.006). He e ogenei y among s udies was high (I2 = 61.18%, p< 0.001). The sub- g oup analysis e- sul s e ealed non-signi ican di e ences in PE (No PE: −0.038 (−0.097, 0.021); PE -0.115 (−0.361, −0.132). In sub-g oup analysis, he esul s main ained he high he - e ogenei y in he PE s udies (sub-g oup No PE: I2 = 56.23%, p= 0.009; sub-g oup PE: I2 = 75.77%, p= 0.006). P ocess e alua ion inclusion and indica o s desc ip ion In he p esen e iew all indica o s ha e been examined and displayed in Table 3. PE was included in 17% o he s udies (7/41). Fideli y and sa is ac ion in 4/7 (57%) s ud- ies we e he indica o s conside ed he mos , ollowed by dose 3/7 (43%), each 2/7 (29%) and ec ui men 1/7 (13%) espec i ely. We also analysed how he PE indica o s we e p e- sen ed in each pape . Ea he e al. [31] epo edPE in a sepa a e pa ag aph wi hin he me hods sec ion Fig. 2 Fo es plo o he o e all s udies using BMI Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 8 o 17 and ocused on in e en ion wo ke s and pa en s e- c ui men , e en ion, adhe ence and sa is ac ion by comple ing e alua ion ques ionnai es by eache s and s uden s. Rec ui men and e en ion we e again e al- ua ed sepa a ely wi h no signi ican di e ences be- ween s udy g oups. Elde e al. [32] desc ibed PE in hei me hodology, examining p ocess da a associa ed o in e en ion ideli y wi h di e en ideli y mea- su es o each amily: “ abula ions o he numbe and ypes o con ac s comple ed”and desc ibes i in a able wi h no u he men ion. Foley e al. [33] conside ed PE h oughou he a icle and akes on a comp ehensi e analysis o a gene al implemen a ion o PE. They ocused on PE in e en ion a 3 le els (“ om in es iga i e eam o communi y wo ke , om communi y wo ke o p ima y ca egi e and om p ima y ca e gi e o child”). Dose and sa is ac- ion we e assessed by in e iew wi h he communi y wo ke s which conclusions h ow ha i was a “poo up ake o in e en ion componen s, and weak e i- cacy o he in e en ion i sel ”. Ge a ds e al. [34] amed PE in hei esul s. Fi s , hey measu ed pa - en al a endance in he g oup and elephone sessions and added PE ques ions in he 4-mon h ques ion- nai e. Then, he pa icipa ing pa en s comple ed a sa is ac ion ques ionnai e. The epo s showed a “high each as majo i y o lessons which we e planned ac ually ook place and he pa en s who did isi a leas one g oup session, 81% (pa en s o 25 child en) we e p esen a 5 o mo e sessions”and pa en s had a good imp ession o he p og am and a ed i on 7.7 on a 10-poin scale. Maddison e al. [43] epo ed ideli y sepa a ely o PE in he me hods sec ion. Fideli y was assessed by moni o ing he ses- sions pe o med by a esea che om he communi y wo ke s wi h eedback o ensu e all componen s we e deli e ed. P ima y ca egi e s also comple ed a su ey o de e mine hei pe cep ions o he Fig. 3 Fo es plo o p ocess e alua ion on s udies using BMI. Subg oup A No p ocess e alua ion; Subg oup B P ocess e alua ion Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 9 o 17 29. Danielsen YS, No dhus IH, Juliusson PB, Maehle M, Pallesen S. E ec o a amily-based cogni i e beha iou al in e en ion on body mass index, sel -es eem and symp oms o dep ession in child en wi h obesi y (aged 7-13): a andomised wai ing lis con olled ial. Obes Res Clin P ac . 2013;7(2):e116–28. 30. Da is CL, Pollock NK, Walle JL, Allison JD, Dennis BA, Bassali R, Melendez A, Boyle CA, Gowe BA. Exe cise dose and diabe es isk in o e weigh and obese child en: a andomized con olled ial. JAMA. 2012;308(11):1103–12. 31. Ea he N, Mo gan PJ, Lubans DR. Imp o ing he i ness and physical ac i i y le els o p ima y school child en: esul s o he Fi -4- un g oup andomized con olled ial. P e Med. 2013;56(1):12–9. 32. Elde JP, C espo NC, Co de K, Ayala GX, Slymen DJ, Lopez NV, Moody JS, McKenzie TL. Childhood obesi y p e en ion and con ol in ci y ec ea ion cen es and amily homes: he MOVE/me Mue o p ojec . Pedia Obes. 2014;9(3):218–31. 33. Foley L, Ni Mhu chu C, Ma sh S, Eps ein LH, Olds T, Dewes O, Heke I, Jiang Y, Maddison R. Sc een ime weigh -loss in e en ion a ge ing child en a home (SWITCH): p ocess e alua ion o a andomised con olled ial in e en ion. BMC Public Heal h. 2016;16:439. 34. Ge a ds SM, Dagnelie PC, Gubbels JS, an Buu en S, Hame s FJ, Jansen MW, an de Goo OH, de V ies NK, Sande s MR, K eme s SP. The e ec i eness o li es yle iple P in he Ne he lands: a andomized con olled ial. PLoS One. 2015;10(4):e0122240. 35. Hollis JL, Su he land R, Campbell L, Mo gan PJ, Lubans DR, Na han N, Wol enden L, Okely AD, Da ies L, Williams A, e al. E ec s o a 'school- based' physical ac i i y in e en ion on adiposi y in adolescen s om economically disad an aged communi ies: seconda y ou comes o he 'Physical ac i i y 4 E e yone' RCT. In J Obes. 2016;40(10):1486–93. 36. Jones RA, Kelly J, Cli DP, Ba e ham M, Okely AD. Accep abili y and po en ial e icacy o single-sex a e -school ac i i y p og ams o o e weigh and a - isk child en: he Wollongong SPORT RCT. Pedia Exe c Sci. 2015; 27(4):535–45. 37. Kala chian MA, Le ine MD, A slanian SA, Ewing LJ, Houck PR, Cheng Y, Ringham RM, Shee s CA, Ma cus MD. Family-based ea men o se e e pedia ic obesi y: andomized, con olled ial. Pedia ics. 2009;124(4):1060–8. 38. Ko acs VA, Fajcsak Z, Gabo A, Ma os E. School-based exe cise p og am imp o es i ness, body composi ion and ca dio ascula isk p o ile in o e weigh /obese child en. Ac a Physiol Hung. 2009;96(3):337–47. 39. La sen KT, Huang T, Ried-La sen M, Ande sen LB, Heidemann M, Molle NC. A mul i-componen day-camp weigh -loss p og am is e ec i e in educing BMI in child en a e one yea : a andomized con olled ial. PLoS One. 2016;11(6):e0157182. 40. Li YP, Hu XQ, Schou en EG, Liu AL, Du SM, Li LZ, Cui ZH, Wang D, Kok FJ, Hu FB, e al. Repo on childhood obesi y in China (8): e ec s and sus ainabili y o physical ac i i y in e en ion on body composi ion o Chinese you h. Biomed En i on Sci. 2010;23(3):180–7. 41. Maddison R, Foley L, Ni Mhu chu C, Jiang Y, Jull A, P apa essis H, Hohepa M, Rodge s A. E ec s o ac i e ideo games on body composi ion: a andomized con olled ial. Am J Clin Nu . 2011;94(1):156–63. 42. Maddison R, Mhu chu CN, Jull A, P apa essis H, Foley LS, Jiang Y. Ac i e ideo games: he media ing e ec o ae obic i ness on body composi ion. In J Beha Nu Phys Ac . 2012;9:54. 43. Maddison R, Ma sh S, Foley L, Eps ein LH, Olds T, Dewes O, Heke I, Ca e K, Jiang Y, Mhu chu CN. Sc een-Time Weigh -loss In e en ion Ta ge ing Child en a Home (SWITCH): a andomized con olled ial. In J Beha Nu Phys Ac . 2014;11:111. h ps://doi.o g/10.1186/s12966-12014-10111-12962. 44. Magnusson KT, H a nkelsson H, Sigu gei sson I, Johannsson E, S einsson T. Limi ed e ec s o a 2-yea school-based physical ac i i y in e en ion on body composi ion and ca dio espi a o y i ness in 7-yea -old child en. Heal h Educ Res. 2012;27(3):484–94. 45. Mon ei o PA, Chen KY, Li a FS, Sa ai a BT, An unes BM, Campos EZ, F ei as IF J . Concu en and ae obic exe cise aining p omo e simila bene i s in body composi ion and me abolic p o iles in obese adolescen s. Lipids Heal h Dis. 2015;14:153. 46. Neme D, Ba zilay-Teeni N, Eliakim A. T ea men o childhood obesi y in obese amilies. J Pedia Endoc inol Me ab. 2008;21(5):461–7. 47. Neme D, Ge a D, Eliakim A. Heal h p omo ion in e en ion in low socioeconomic kinde ga en child en. J Pedia . 2011;158(5):796–801.e791. 48. Nowicka P, Lanke J, Pie obelli A, Api zsch E, Flodma k CE. Spo s camp wi h six mon hs o suppo om a local spo s club as a ea men o childhood obesi y. Scand J Public Heal h. 2009;37(8):793–800. 49. Sa die M, Jennings-Abu o N, Le esque L, Janssen I, Campi ano-Nunez F, Lopez-Olmedo N, Abu o T, Ri e a JA. Impac o a school-based in e en ion p og am on obesi y isk ac o s in Mexican child en. Salud Publica Mex. 2013;55(Suppl 3):374–87. 50. Singh AS, Chin APMJ, B ug J, an Mechelen W. Du ch obesi y in e en ion in eenage s: e ec i eness o a school-based p og am on body composi ion and beha io . A ch Pedia Adolesc Med. 2009;163(4):309–17. 51. Simon C, Schwei ze B, Oujaa M, Wagne A, A eile D, T iby E, Copin N, Blanc S, Pla a C. Success ul o e weigh p e en ion in adolescen s by inc easing physical ac i i y: a 4-yea andomized con olled in e en ion. In J Obes. 2008;32(10):1489–98. 52. Wa e s E, Gibbs L, Tadic M, Ukoumunne OC, Maga ey A, Okely AD, de Sil a A, A mi C, G een J, O'Conno T, e al. Clus e andomised ial o a school- communi y child heal h p omo ion and obesi y p e en ion in e en ion: indings om he e alua ion o un 'n heal hy in Mo eland! BMC Public Heal h. 2017;18(1):92. 53. Williamson DA, Champagne CM, Ha sha DW, Han H, Ma in CK, New on RL J , So he n MS, S ewa TM, Webbe LS, Ryan DH. E ec o an en i onmen al school-based obesi y p e en ion p og am on changes in body a and body weigh : a andomized ial. Obesi y (Sil e Sp ing). 2012;20(8):1653–61. 54. Yackobo i ch-Ga an M, Nagelbe g N, Phillip M, Ashkenazi-Ho nung L, He shko i z E, Shali in S. The in luence o die and/o exe cise and pa en al compliance on heal h- ela ed quali y o li e in obese child en. Nu Res. 2009;29(6):397–404. 55. Yin Z, Moo e JB, Johnson MH, Ve non MM, Gu in B. The impac o a 3-yea a e -school obesi y p e en ion p og am in elemen a y school child en. Child Obes. 2012;8(1):60–70. 56. Casazza K, Ca del M, Dulin-Kei a A, Hanks LJ, Gowe BA, New on AL, Wallace S. Reduced ca bohyd a e die o imp o e me abolic ou comes and dec ease adiposi y in obese pe ipube al A ican Ame ican gi ls. J Pedia Gas oen e ol Nu . 2012;54(3):336–42. 57. Copeland KC, Higgins J, El Gho mli L, Delahan y L, G ey M, K iska AM, Lipman TH, Pyle L, Shephe d J, Hi s K. T ea men e ec s on measu es o body composi ion in he TODAY clinical ial. Diabe es Ca e. 2013;36(6): 1742–8. 58. Gu in B, Yin Z, Johnson M, Ba beau P. P elimina y indings o he e ec o a 3-yea a e -school physical ac i i y in e en ion on i ness and body a : he medical College o Geo gia Fi kid P ojec . In J Pedia Obes. 2008;3(SUPPL. 1):3–9. 59. Johns on CA, Tyle C, Fulle on G, McFa lin BK, Pos on WS, Haddock CK, Ree es RS, Fo ey JP. E ec s o a school-based weigh main enance p og am o Mexican-Ame ican child en: esul s a 2 yea s. Obesi y (Sil e Sp ing). 2010;18(3):542–7. 60. Khan NA, Raine LB, D olle e ES, Scudde MR, Pon i ex MB, Cas elli DM, Dono an SM, E ans EM, Hillman CH. Impac o he FITKids physical ac i i y in e en ion on adiposi y in p epube al child en. Pedia ics. 2014;133(4): e875–83. h ps://doi.o g/10.1542/peds.2013-2246 Epub 2014 Ma 1531. 61. Lau PW, Wong del P, Ngo JK, Liang Y, Kim CG, Kim HS. E ec s o high- in ensi y in e mi en unning exe cise in o e weigh child en. Eu J Spo Sci. 2015;15(2):182–90. 62. Meye U, Schindle C, Zahne L, E ns D, Hebes ei H, an Mechelen W, B unne -La Rocca HP, P obs -Hensch N, Pude JJ, K iemle S. Long- e m e ec o a school-based physical ac i i y p og am (KISS) on i ness and adiposi y in child en: a clus e - andomized con olled ial. PLoS One. 2014; 9(2):e87929. 63. Ma ild S, G onowi z E, Fo sell C, Dahlg en J, F ibe g P. A con olled s udy o li es yle ea men in p ima y ca e o child en wi h obesi y. Pedia Obes. 2013;8(3):207–17. 64. Rush E, Reed P, McLennan S, Coppinge T, Simmons D, G aham D. A school-based obesi y con ol p og amme: p ojec ene gize. Two-yea ou comes. B J Nu . 2012;107(4):581–7. 65. Velazquez-Lopez L, San iago-Diaz G, Na a-He nandez J, Munoz-To es AV, Medina-B a o P, To es-Tamayo M. Medi e anean-s yle die educes me abolic synd ome componen s in obese child en and adolescen s wi h obesi y. BMC Pedia . 2014;14:175. 66. Simon C, Wagne A, Pla a C, A eile D, Schwei ze B, Schlienge JL, T iby E. ICAPS: a mul ile el p og am o imp o e physical ac i i y in adolescen s. Diabe es Me ab. 2006;32(1):41–9. 67. Hollis J, Su he land R, Campbell L, Mo gan PJ, Lubans DR, Na han N, Wol enden L, Okely AD, Da ies L, Williams A, Cohen KE, Oldmeadow C, Gillham K, Wigge s J. E ec s o a 'school-based' physical ac i i y in e en ion Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 16 o 17 on adiposi y in adolescen s om economically disad an aged communi ies: seconda y ou comes o he 'Physical ac i i y 4 E e yone' RCT. In J Obes (2005). 2016;40(10):1486. 68. Duman F, Kokacya MH, Dog u E, Ka ayi ci N, Canbay O, Aman F. The Role o Ac i e Video-Accompanied Exe cises in Imp o emen o he Obese S a e in Child en: A P ospec i e S udy om Tu key. J Clin Res Pedia Endoc inol. 2016;8(3):334–40. h ps://doi.o g/10.4274/jc pe.2284 Epub 2016 Ap 4218. 69. Mei Z, G umme -S awn LM, Pie obelli A, Goulding A, Go an MI, Die z WH. Validi y o body mass index compa ed wi h o he body-composi ion sc eening indexes o he assessmen o body a ness in child en and adolescen s. Am J Clin Nu . 2002;75(6):978–85. 70. F anken ield DC, Rowe WA, Cooney RN, Smi h JS, Becke D. Limi s o body mass index o de ec obesi y and p edic body composi ion. Nu i ion. 2001;17(1):26–30. 71. Cole TJ, Fai h MS, Pie obelli A, Heo M. Wha is he bes measu e o adiposi y change in g owing child en: BMI, BMI %, BMI z-sco e o BMI cen ile? Eu J Clin Nu . 2005;59(3):419–25. 72. Moo e GF, Aud ey S, Ba ke M, Bond L, Bonell C, Ha deman W, Moo e L, O'Ca hain A, Tina i T, Wigh D, e al. P ocess e alua ion o complex in e en ions: Medical Resea ch Council guidance. Bmj. 2015;350:h1258. 73. B own T, Summe bell C. Sys ema ic e iew o school-based in e en ions ha ocus on changing die a y in ake and physical ac i i y le els o p e en childhood obesi y: an upda e o he obesi y guidance p oduced by he Na ional Ins i u e o heal h and clinical excellence. Obes Re . 2009;10(1): 110–41. 74. Mead E, B own T, Rees K, Aze edo LB, Whi ake V, Jones D, Olajide J, Maina di GM, Co peleijn E, O'Malley C, e al. Die , physical ac i i y and beha iou al in e en ions o he ea men o o e weigh o obese child en om he age o 6 o 11 yea s. Coch ane Da abase Sys Re . 2017;6: Cd012651. 75. F uhbeck G, Sb accia P, Nisoli E, Woodwa d E, Yumuk V, Fa pou -Lambe NJ, Hal o d JG, Toplak H, Ca uba MO. 2015 Milan decla a ion: a call o ac ion on obesi y - an EASO posi ion s a emen on he occasion o he 2015 EXPO. Obes Fac s. 2016;9(4):296–8. 76. Liese AD, D'Agos ino RB J , Hamman RF, Kilgo PD, Law ence JM, Liu LL, Loo s B, Linde B, Ma co ina S, Rod iguez B, e al. The bu den o diabe es melli us among US you h: p e alence es ima es om he SEARCH o diabe es in you h s udy. Pedia ics. 2006;118(4):1510–8. 77. Al-Saeed AH, Cons an ino MI, Molyneaux L, D'Souza M, Limache -Gisle F, Luo C, Wu T, Twigg SM, Yue DK, Wong J. An in e se ela ionship be ween age o ype 2 diabe es onse and complica ion isk and mo ali y: he impac o you h-onse ype 2 diabe es. Diabe es Ca e. 2016;39(5):823–9. 78. Manios Y, And ou sos O, Lamb inou CP, Ca don G, Linds om J, Annemans L, Ma eo-Gallego R, de Saba a MS, Io o a V, Ki ela J, e al. A school- and communi y-based in e en ion o p omo e heal hy li es yle and p e en ype 2 diabe es in ulne able amilies ac oss Eu ope: design and implemen a ion o he Feel4Diabe es-s udy. Public Heal h Nu . 2018;21(17): 3281–90. 79. Bha M, Naha i A, Wang PW, Kea sley E, Falzone N, Chen S, Fu E, Jeyakuma Y, Zukowski J, Ban ield L, e al. The quali y o clinical p ac ice guidelines o managemen o pedia ic ype 2 diabe es melli us: a sys ema ic e iew using he AGREE II ins umen . Sys Re . 2018;7(1):193. Publishe ’sNo e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Se al-Co es e al. BMC Public Heal h (2021) 21:348 Page 17 o 17