Burden of childhood obesity in Pacific Islands Countries and Territories Authors: Anthony Okely, Sarah T. Ryan, Sumantla Varman, Rebecca Bogarobu Emori, Olivier Galy, Rachel Novotny Published: 2026 ABSTRACT SummaryChildhood obesity is a major and escalating health challenge in Pacific Island Countries and Territories (PICTs), where prevalence rates among children rank among the highest globally. This first paper in a three-part series synthesises evidence on the prevalence and sociodemographic distribution of overweight and obesity among children aged 2.0–11.9 years across 22 PICTs. Twenty-nine unique studies involving more than 33,000 children were included. Prevalence varied widely, from as low as 3.3 percent in the Marshall Islands to more than 50 percent in Wallis and Futuna and American Samoa, depending on population characteristics and the definitions used. Rates were highest in Polynesian and Micronesian countries, often severalfold above global averages. Across studies, prevalence generally increased with age and was frequently higher among boys than girls, though patterns varied. Children in urban settings or higher socioeconomic households were typically more likely to have overweight or obesity, reflecting rapid nutrition transitions marked by increasing consumption of imported, energy-dense foods and reduced physical activity. Ethnic disparities were pronounced, with Pacific Islander children at greatest risk. These findings indicate a region-wide epidemic driven by social, dietary, and environmental change, underscoring the need for strengthened surveillance, culturally grounded prevention, and coordinated regional action. Introduction to series Childhood obesity has been described as a global crisis, affecting countries across all income levels. In the past 40 years, almost every nation has experienced a significant rise in prevalence rates.1,2 Childhood obesity has been the subject of two previous series in The Lancet in 20103 and 2015,4 which primarily focused on school-aged children and adolescents, highlighting the global significance of the condition and the need for coordinated system-level actions. The most recent series reported on rapid increases in many low- and middle-income countries (LMICs) where undernutrition remains prevalent. It called for a healthy growth strategy to address both under- and overnutrition by ensuring nutrition security, protecting children from the marketing of energy-dense nutrient-poor foods and sedentary activities, and strengthening food system governance to promote children’s health. While it has been known that Pacific Islands Countries and Territories (PICTs) have some of the highest rates of childhood obesity in the world,5 it is timely to synthesise the current body of evidence on the prevalence, influencing factors, and efficacy of interventions in childhood obesity in PICTs. While efforts to track key indicators of health, including child obesity, have been initiated, such as the Pacific Monitoring Alliance for NCD Action (MANA) dashboard hosted by the Secretariat of the Pacific Community,6 efforts to address this public health concern have been constrained by a lack of comprehensive and consistent surveillance data.7 The lack of reliable epidemiological data on childhood obesity prevalence, its correlates and determinants, and on effective local strategies is a major barrier to developing policies that can effectively respond to the unique sociocultural, environmental, and economic conditions in this region, where rapid social and dietary transitions, reliance on imported foods, dispersed island geographies, and limited resources influence children’s health.6 Although the current evidence base is patchy and inconsistent, and constrained by limitations in study design, age coverage, sample size, and representativeness, this series aims to synthesise the best available evidence on childhood obesity in PICTs, identify critical knowledge gaps, and propose actionable pathways to improve surveillance, inform policy, and implement regionally appropriate interventions. The first article aims to describe the prevalence and sociodemographic distribution of overweight and obesity among children living in PICTs. The second paper examines the complex and interconnected sociocultural, environmental, and economic determinants of childhood obesity across PICTs, highlighting the associations between traditional practices, globalisation, food systems, physical activity environments, and broader political and economic drivers. The paper aims to provide a contextualised understanding of how these factors shape obesogenic environments and influence children’s overweight and obesity. The third paper evaluates the efficacy and effectiveness of community- and school-based lifestyle interventions for preventing or treating childhood and adolescent obesity in PICTs. Incorporating peer-reviewed and grey literature, it examines how cultural relevance, delivery settings, and implementation approaches influence the success of interventions. The paper identifies promising practices and provides direction for strengthening future intervention design and evaluation in PICTs. For this series, we defined childhood as 2.0–11.9 years of age. Children were further categorised into early childhood (two to less than five years of age) and childhood (ages 5.0–11.9 years). For Paper 3, the age range was extended to include adolescents (12.0–15.0 years) because preliminary scoping indicated that restricting eligibility to children aged 2.0–11.9 years would yield too few intervention studies to support a meaningful synthesis of intervention effectiveness in PICTs. Methods This systematic literature review, which is part one of a three-part series, followed the PRISMA Statement for Reporting Systematic Reviews8 to describe the prevalence and sociodemographic distribution of overweight and obesity among children aged 2.0–11.9 years living in PICTs. Protocol and registration This review was registered with PROSPERO (2025 CRD420251090452). A systematic review protocol was developed to guide this review and can be found online via PROSPERO. Eligibility criteria To address the research questions, studies were selected based on predefined Population, Exposure, and Outcomes, and Study Design (PEOS) criteria. Population Children were defined as individuals aged 2.0–11.9 years. We used this as the definitions for overweight and obesity are different for children aged under two years and because the review was designed to focus specifically on childhood obesity. This was further categorised into early childhood (2.0–4.9 years) and childhood (5.0–7.9 and 8.0–11.9 years). Participants must have lived in one of the following Pacific Island Countries and Territories (PICTs), based on Pacific Community (SPC) membership: American Samoa, Cook Islands, Commonwealth of the Northern Mariana Islands, Federated States of Micronesia, Fiji, French Polynesia, Guam, Kiribati, Nauru, New Caledonia, Niue, Papua New Guinea, Pitcairn Islands, Marshall Islands, Palau, Samoa, Solomon Islands, Tonga, Tuvalu, Tokelau, Vanuatu, Wallis and Futuna. Studies conducted in Australia, New Zealand, Hawaii, Alaska, or involving children from PICTs living in Australia, New Zealand, or the USA were excluded, as were studies from Easter Island (Rapa Nui). Exposure The exposures of interest were standard sociodemographic and contextual factors commonly reported in epidemiological studies examining childhood overweight and obesity in PICTs, including age, sex, socioeconomic status, ethnicity, rurality or urbanicity, and country or territory of residence, where these data were reported. Outcome/s -Prevalence rates of overweight and obesity among children aged 2.0–11.9 years.-Overweight or obesity as measured using height and weight (to calculate body mass index [BMI]) or weight-for-height as a standard deviation (SD), waist circumference, or skinfold thickness. We did not calculate overweight or obesity for any of the studies but relied on the definition used by the authors for overweight and obesity in their papers. -Prevalence rates of overweight and obesity among children aged 2.0–11.9 years. Prevalence rates of overweight and obesity among children aged 2.0–11.9 years. -Overweight or obesity as measured using height and weight (to calculate body mass index [BMI]) or weight-for-height as a standard deviation (SD), waist circumference, or skinfold thickness. We did not calculate overweight or obesity for any of the studies but relied on the definition used by the authors for overweight and obesity in their papers. Overweight or obesity as measured using height and weight (to calculate body mass index [BMI]) or weight-for-height as a standard deviation (SD), waist circumference, or skinfold thickness. We did not calculate overweight or obesity for any of the studies but relied on the definition used by the authors for overweight and obesity in their papers. Study design Cross-sectional or longitudinal studies were included. Exclusion criteria Studies were excluded if they were reviews, editorials, conference abstracts, commentaries, or reported secondary analysis of previously published data (e.g., the Global School-Based Student Health Survey (GSHS)). Studies not reporting prevalence data or determinants/risk factors, studies on adult populations, and studies conducted outside the defined Pacific region were also excluded. Studies were excluded if the mean age of participants exceeded 12 years, as the review focused on populations predominantly within the target age range of 2–12 years. No exclusions were made based on sample size. Information sources A research librarian with expertise in systematic reviews advised on the search strategy and database selection. We used the same search strategy in how we listed the PICTs as in a previously published systematic review.9 We systematically searched seven electronic databases in English: PsycINFO, CINAHL, PROQUEST (including ERIC), Web of Science, MEDLINE, Scopus, and SPORTDiscus. In French, Google Scholar was searched. The searches were conducted from inception to 31st January 2024, without restrictions on the language of publication. Non-English papers were translated using online tools (Google Translate). The references of included studies and relevant review articles were also searched to identify studies. Search strategy The search strategy was designed to identify literature in both English and French. For the English searches, terms included: (“child∗” OR “pediatric∗” OR “paediatric∗”) AND (“obes∗” OR “overweight∗” OR “adipos∗” OR “body mass index” OR “BMI” OR “weight∗”) AND (“prevalen∗” OR “epidemiol∗” OR “determinant∗” OR “risk factor∗” OR “caus∗” OR “correlat∗” OR demograph∗ OR economic OR social OR commercial) AND (population OR statistics OR incidence OR frequency OR occurrence) AND (“Pacific Island∗” OR “American Samo∗” OR “Cook Island∗” OR “Federated States of Micronesi∗” OR Fiji OR “French Polynesi∗” OR Guam OR Kiriba OR Nauru OR “New Caledoni∗” OR Niue OR “Northern Mariana Island∗” OR “Marshall Island∗” OR Palau OR “Papua New Guine∗” OR “Pitcairn Island∗” OR Samo∗ OR “Solomon Island∗” OR Tokelau OR Tong∗ OR Tuvalu OR Vanuatu OR “Wallis and Futun∗”). For the French searches, due to character limits in the Publish or Perish software (Version 8.18.5091)10 used for Google Scholar, the original search string was divided into nine individual searches. The full French search terms included: (“enfant∗” OR “pédiatri∗”) AND (“obés∗” OR “surpoids∗” OR “adipos∗” OR “indice de masse corporelle” OR “IMC” OR “poids∗”) AND (“prévalen∗” OR “épidémiol∗” OR “déterminant∗” OR “facteur de risque∗” OR “caus∗” OR “corrélat∗” OR démograph∗ OR économique OR social OR commercial) AND (population OR statistiques OR incidence OR fréquence OR occurrence) AND (“Îles du Pacifique∗” OR “Samoa américaines∗” OR “Îles Cook∗” OR “États fédérés de Micronésie∗” OR Fidji OR “Polynésie française∗” OR Guam OR Kiribati OR Nauru OR “Nouvelle-Calédonie∗” OR Niue OR “Îles Mariannes du Nord∗” OR “Îles Marshall∗” OR Palaos OR “Papouasie-Nouvelle-Guinée∗” OR “Îles Pitcairn∗” OR Samoa∗ OR “Îles Salomon∗” OR Tokelau OR Tonga∗ OR Tuvalu OR Vanuatu OR “Wallis et Futuna∗”). MeSH terms were applied for Medline, CINAHL, and SPORTSDiscus databases, specifically using “Pediatric Obesity” or “Obesity” for the concept of childhood obesity. All search results were imported into Mendeley (Version 2.135.0, 2024) to remove duplicates and then exported to a .ris file for upload into Covidence. Study selection Covidence systematic review software (2024)11 was used to conduct this review. Two authors (AO, SV, RE, or SR) independently screened each title and abstract. Full texts of all included studies, based on title and abstract screening, were reviewed by the same two authors. Any disagreements were resolved through consensus or by consulting a third author. The study selection process is further detailed in Fig. 1, PRISMA Flow Diagram. Inter-rater reliability was examined separately for the title and abstract screening and full-text review stages using the proportion of decisions in agreement and Cohen’s kappa coefficient (κ). Agreement statistics were calculated for each reviewer pair because the number of records assessed by different pairs varied. All disagreements were resolved through discussion and consensus or, where necessary, consultation with a third reviewer.Fig. 1Flow chart. Burden of childhood obesity in Pacific Islands. Flow chart. Burden of childhood obesity in Pacific Islands. Data collection process Data from each study were also extracted by two authors (AO, SV, RE, or SR) independently using forms within Covidence. Data extraction forms were piloted by two review authors (AO, SR) prior to their publication and use by the whole review team. This initial piloting ensured consistency in data collection and reduced the need for later consensus. Disagreements during data extraction were resolved through discussion and consensus or, if necessary, by a third reviewer. Data items Data extracted from each included study encompassed a wide range of variables relevant to the review’s objectives:-General study information: Year of publication, country/ies where the study was conducted, language of publication (English/French), aim/s of study, source of funding, and reported conflicts of interest.-Population and setting: Detailed description of the population and setting, including demographics, location, and month/year of data collection. For age, rurality and socioeconomic status, data were extracted and reported according to the definitions used in the original studies, recognising that these classifications varied across settings.-Participant characteristics: Mean age (years) and standard deviation, proportion of boys/girls, proportion urban/rural, socioeconomic status (SES), and ethnicity.-Recruitment and sampling: Eligibility criteria of participants, how participants were selected/identified and recruited, sample size, how sample size was determined, and whether the sample was designed to be representative of a particular region or country.-Measurement of variables: How weight status was assessed and defined (e.g., using IOTF,12 WHO,13 CDC14 cut-offs). Details on who conducted the assessments and what training they received were also extracted.-Confounding factors: Information on whether confounding factors were identified and what strategies were used to deal with them in the analyses (e.g., statistical adjustments).-Statistical analysis: Details on the statistical analyses used.-Study findings: Response rate, main results, and results from any sub-analyses, specifically prevalence rates of non-overweight, overweight, and obesity, geographical distribution, and breakdown of prevalence by age group, sex, rurality, socio-economic status, and ethnicity.-Where studies did not report this information, NR was listed in the extraction. -General study information: Year of publication, country/ies where the study was conducted, language of publication (English/French), aim/s of study, source of funding, and reported conflicts of interest. General study information: Year of publication, country/ies where the study was conducted, language of publication (English/French), aim/s of study, source of funding, and reported conflicts of interest. -Population and setting: Detailed description of the population and setting, including demographics, location, and month/year of data collection. For age, rurality and socioeconomic status, data were extracted and reported according to the definitions used in the original studies, recognising that these classifications varied across settings. Population and setting: Detailed description of the population and setting, including demographics, location, and month/year of data collection. For age, rurality and socioeconomic status, data were extracted and reported according to the definitions used in the original studies, recognising that these classifications varied across settings. -Participant characteristics: Mean age (years) and standard deviation, proportion of boys/girls, proportion urban/rural, socioeconomic status (SES), and ethnicity. Participant characteristics: Mean age (years) and standard deviation, proportion of boys/girls, proportion urban/rural, socioeconomic status (SES), and ethnicity. -Recruitment and sampling: Eligibility criteria of participants, how participants were selected/identified and recruited, sample size, how sample size was determined, and whether the sample was designed to be representative of a particular region or country. Recruitment and sampling: Eligibility criteria of participants, how participants were selected/identified and recruited, sample size, how sample size was determined, and whether the sample was designed to be representative of a particular region or country. -Measurement of variables: How weight status was assessed and defined (e.g., using IOTF,12 WHO,13 CDC14 cut-offs). Details on who conducted the assessments and what training they received were also extracted. Measurement of variables: How weight status was assessed and defined (e.g., using IOTF,12 WHO,13 CDC14 cut-offs). Details on who conducted the assessments and what training they received were also extracted. -Confounding factors: Information on whether confounding factors were identified and what strategies were used to deal with them in the analyses (e.g., statistical adjustments). Confounding factors: Information on whether confounding factors were identified and what strategies were used to deal with them in the analyses (e.g., statistical adjustments). -Statistical analysis: Details on the statistical analyses used. Statistical analysis: Details on the statistical analyses used. -Study findings: Response rate, main results, and results from any sub-analyses, specifically prevalence rates of non-overweight, overweight, and obesity, geographical distribution, and breakdown of prevalence by age group, sex, rurality, socio-economic status, and ethnicity. Study findings: Response rate, main results, and results from any sub-analyses, specifically prevalence rates of non-overweight, overweight, and obesity, geographical distribution, and breakdown of prevalence by age group, sex, rurality, socio-economic status, and ethnicity. -Where studies did not report this information, NR was listed in the extraction. Where studies did not report this information, NR was listed in the extraction. Data were not extracted for those below a healthy weight range (that is, underweight). Risk of bias in individual studies The JBI Critical Appraisal Checklist for Analytical Cross-Sectional Studies15 was used to assess the methodological quality and risk of bias (ROB) for each of the included studies. Each study was assessed independently by two authors (AO, SV, RE, or SR). Inconsistencies were discussed to reach a consensus or resolved by a third reviewer where needed. For each question on the checklist, a “Yes” response was assigned one point, while “No,” “Unclear,” or “Not Applicable” responses were assigned zero points. A percentage score was calculated by dividing the total points obtained by the maximum possible points, with a higher percentage indicating a lower risk of bias. No studies were excluded based on their ROB assessment scores. The assessment considered aspects such as clearly defined inclusion criteria, detailed description of study subjects and setting, valid and reliable measurement of exposure and outcomes, use of objective standard criteria for condition measurement, identification of confounding factors, and appropriate statistical analysis. Summary measures The primary summary measures included the proportion of children categorised as non-overweight (a healthy weight), overweight, or obese. Outcomes were reported based on sex, age (2.0–4.9 years, 5.0–7.9 years, or 8.0–11.9 years), rurality, SES, and race or ethnicity to address the secondary research questions on sociodemographic distribution. Data analysis and synthesis of results Due to the anticipated heterogeneity among studies, particularly regarding the varying tools used for assessing weight status (e.g., WHO, CDC, IOTF), a formal meta-analysis was not feasible. Instead, a narrative synthesis approach was used to present and discuss the findings from the included studies. The synthesis was structured around the primary and secondary research questions: Prevalence of overweight and obesity For the primary outcome, prevalence rates of overweight and obesity among children aged 2.0–11.9 years were summarised narratively, providing ranges of reported prevalence rates across different PICTs. Specific definitions and guidelines (e.g., WHO, CDC, IOTF) used by each study were noted and appear in the Supplemental file. Prevalence data were presented both for combined overweight and obesity, as well as separately, where reported. Sociodemographic distribution Sociodemographic distribution of childhood overweight and obesity was synthesised by key characteristics, including age (categorised as 2.0–4.9 years, 5.0–8.9 years, and 9.0–11.9 years), sex, and rurality. Where reported, SES, race, and ethnicity were also discussed. The narrative highlights patterns and variations in prevalence rates across these sociodemographic factors. Presentation of results Results were presented primarily through descriptive text and tables. Study characteristics and extracted data (including country, participant demographics, assessment methods, and reported prevalence rates) are presented as a summary table to allow for comparison and to highlight methodological differences between studies. The risk of bias assessment results for each included study is also presented in a table to inform the interpretation of the overall quality of evidence. The narrative synthesis identifies consistent findings, highlights discrepancies, and identifies gaps across Pacific Islands and specific child populations. Role of the funding source There was no funding source (funder) for this study. The content is solely the responsibility of the authors. Results We retrieved 5844 potential records from the databases/registers search and other sources (Fig. 1)16 and 1979 were excluded due to duplicates. The titles or abstracts of studies were screened; 3690 studies were further excluded. We assessed the full text of 187 studies to determine their eligibility for inclusion in the reviews. Twenty-nine articles met the eligibility criteria and were included in this review. These represented 23 unique studies, totalling 33,809 participants. Twenty-five articles reported on cross-sectional studies, and four articles on longitudinal designs. Of the 22 PICTs, 14 were included across the 29 articles. Five of the articles were from Samoa, three from New Caledonia, Guam, and Tonga, two from the Marshall Islands (RMI), Vanuatu, Federated States of Micronesia (FSM), Papua New Guinea (PNG), Commonwealth of Northern Mariana Islands (CNMI) and French Polynesia, and one each from American Samoa, Fiji, and Wallis and Futuna. There were three multi-country studies conducted in the US-affiliated PICTs (CMNI, Guam, American Samoa, Palau, RMI, and FSM), Papua New Guinea and Vanuatu, and FSM, Palau, and RMI. The age of the study participants ranged from 2 to 15 years of age. Sample sizes ranged from 36 to 3383 participants. Characteristics from all the included studies are shown in Table 1.Table 1Characteristics of included studies.CountryAuthorSurvey YearAge (years)Sample sizeMeasurement methods Obesity: SES: Ethnicity; RuralityDefinition used for weight statusProportion overweight, obese, and overweight + obeseAmerican SamoaGurr (2009)2006–20086–11 years2308 (2006–2007) 1805 (2008–2009)CDC2006–2007Grade KBoysOverweight = 19.9%Obese = 23.9%Males Overweight + Obese = 43.8%GirlsOverweight = 27.0%Obese = 16.1%Females Overweight + Obese = 43.1%Grade 3BoysOverweight = 17.1%Obese = 27.1%Males Overweight + Obese = 44.2%GirlsOverweight = 17.9%Obese = 25.6%Females Overweight + Obese = 43.5%Grade 6BoysOverweight = 17.4%Obese = 37.0%Males Overweight + Obese = 54.4%GirlsOverweight = 22.3%Obese = 34.5%Females Overweight + Obese = 56.8%2008–2009Grade 2BoysOverweight = 20.9%Obese = 22.6%Males Overweight + Obese = 43.5%GirlsOverweight = 22.3%Obese = 16.8%Females Overweight + Obese = 39.1%Grade 5BoysOverweight = 15.6%Obese = 34.1%Males Overweight + Obese = 49.7%GirlsOverweight = 18.8%Obese = 32.4%Females Overweight + Obese = 51.2%Commonwealth of the Northern Mariana IslandsNovotny (2007)20052–10 y346Obesity:CDCTotalOverweight = 15%Obese = 19%Total Overweight + Obese = 34%Commonwealth of the Northern Mariana IslandsPaulino (2008)20052.0–4.9 y; 5–12 years393Obesity: Height and weight were measured using similar methods and quality control procedures as Novotny and colleagues (4). Children were measured for weight using a digital scale (Seca, Hamburg, Germany). A Shorrboard (Shorr Productions, Olney, MD) was used to measure recumbent length for children younger than 24 months and height for children 24 months and older. Measurements were used to categorize children by body mass index (BMI; calculated as kg/m2)-for-age percentiles.SES: NREthnicity: NRRurality: NRCDCTotal Overweight + Obese = 33.5%2–3 y Overweight = 12%2–3 y Obese = 13%2–3 y Overweight + Obese = 25%4–6 y Overweight = 13%4–6 y Obese = 13%4–6 y Overweight + Obese = 26%7–10 y Overweight = 18%7–10 y Obese = 27%7–10 y Overweight + Obese = 45%Federated States of Micronesia (Yap)Durand (2007)2006>2–5 years; >5–10 years; >10–15 years448Obesity: BMI (height and weight)SES: NREthnicity: NRRurality: NRCDCTotal Overweight = 10.6%Total Obese = 17.6%Total Overweight + Obese = 28.2%>2–5 y Overweight = 6%>2–5 y Obese = 21%>2–5 y Overweight + Obese = 27%>5–10 y Overweight = 12%>5–10 y Obese = 16%>5–10 y Overweight + Obese = 28%>10–15 y Overweight = 12%>10–15 y Obese = 16%>10–15 y Overweight + Obese = 28%Federated States of Micronesia (Yap State)Wa’ab CHC (2006–2007)Outer Island Household Survey (2008–2009)Ichiho (2013)20112–14 y1736 (2006–07)1948 (2008–2009)Obesity: height and weight were assessed.SES: NREthnicity: NRRurality: NRCDCWa’ab CDCTotalOverweight = 12%Obese = 20%Total Overweight + Obese = 32%Outer Island Household SurveyTotalOverweight = 15.6%Obese = 18.2%Total Overweight + Obese = 33.8%MalesOverweight = 11.8%Obese = 17.7%Males Overweight + Obese = 29.5%FemalesOverweight = 18.1%Obese = 16.9%Females Overweight + Obese = 35.0%FijiNational Food & Nutrition Centre (2017)2014–20152–12 years629 (Birth–5.9 y)1288 (6–12 y)BMIEthnic group iTaukeiFID (Fijian of Indian Descent)FOD (Fijian of Other Descent)SexMaleFemaleArea typeUrbanRural>2SD ≤ 3SD (Overweight)>3SD (obese)2.0–4.9 y (59 mo)TotalOverweight = 3.0% (95% CI 1.5–4.6)Obese = 0.2% (95% CI 0.0–0.6)2.0–4.9 y Overweight + Obese = 3.2%Birth–4.9 y (59 mo)MalesOverweight = 6.3% (95% CI 3.9–8.6)Obese = 0.9% (95% CI 0.00–11.3)Males Overweight + Obese = 7.2%FemalesOverweight = 3.4% (95% CI 1.6–5.2)Obese = 1.0% (95% CI 0.0–11.3)Females Overweight + Obese = 4.4%Ethnicity iTaukeiOverweight = 5.5% (95% CI 3.5–7.4)Obese = 1.0% (95% CI 0.1–1.9)Ethnicity (iTaukei) Overweight + Obese = 6.5%FIDOverweight = 2.3% (95% CI 0.4–4.3)Obese = 0.3% (95% CI 0.0–1.2)Ethnicity (FOD) Overweight + Obese = 2.6%FODOverweight = 12.3% (95% CI 1.5–23.2)Obese = 4.2% (95% CI 0.0–11.3)Ethnicity (FOD) Overweight + Obese = 16.5%UrbanOverweight = 7.7% (95% CI 5.0–10.3)Obese = 1.4% (95% CI 0.2–2.6)RuralOverweight = 2.3% (95% CI 0.8–3.7)Obese = 0.5% (95% CI 0.0–1.3)6–12 yTotalOverweight = 7.2% (95% CI 6.7–7.6)Obese = 1.7% (95% CI 1.5–1.9)6–12 y Overweight + Obese = 8.9%MalesOverweight = 7.6% (95% CI 7.0–8.2)Obese = 2.5% (95% CI 2.1–2.8)Males Overweight + Obese = 10.1%FemalesOverweight = 6.6% (95% CI 6.0–7.2)Obese = 0.8% (95% CI 0.6–1.0)Females Overweight + Obese = 7.4%Ethnicity iTaukeiOverweight = 6.1% (95% CI 5.6–6.5)Obese = 1.4% (95% CI 1.2–1.7)Ethnicity (iTaukei) Overweight + Obese = 7.5%FIDOverweight = 9.1% (95% CI 8.3–10.0)Obese = 2.1% (95% CI 1.7–2.6)Ethnicity (FOD) Overweight + Obese = 11.2%FODOverweight = 9.9% (95% CI 7.6–12.3)Obese = 2.7% (95% CI 1.4–4.0)Ethnicity (FOD) Overweight + Obese = 12.6%UrbanOverweight = 9.6% (95% CI 8.9–10.3)Obese = 2.4% (95% CI 2.0–2.8)RuralOverweight = 5.0% (95% CI 4.6–5.5)Obese = 1.1% (95% CI 0.9–1.3)French PolynesiaDaigre (2012)2007–20085–12 years296Obesity: Body weight was measured with a Tanita BF 522 W balance (Tanita Corporation, Tokyo, Japan), height with an electronic device (Soehnle Professional GmbH, Backnang, Germany), and waist circumference with a constant-tension measuring tape placed, according to World Health Organization (WHO) recommendations, midway between iliac crests and lowermost ribs [9].SES: NREthnicity: NRRurality: NRWHO references and IOTFWHO DefinitionMales Overweight = 31.0% (SE 3.8)Males Obese = 21.4% (SE 3.4)Males Overweight + Obese = 52.4%Females Overweight = 14.6% (SE 2.9)Females Obese = 19.2% (SE 3.2)Females Overweight + Obese = 33.8%IOTF DefinitionMales Overweight = 23.4% (SE 3.5)Males Obese = 15.2% (SE 2.9)Females Overweight = 11.3% (SE 2.6)Females Obese = 16.6% (SE 3.0)French PolynesiaMinistry of Health (2015)20147–9 years1768IOTFMalesOverweight = 18.7% (95% CI 16.2–21.2)Obese = 17.1% (95% CI 14.7–19.5)Males Overweight + Obese = 35.8%FemalesOverweight = 20.5% (95% CI 17.8–23.2)Obese = 15.3% (95% CI 12.9–17.7)Females Overweight + Obese = 35.8%GuamPaulino (2015)2010–20144–5 y6–11 y2010CDC2010–2011 (sex not reported separately for age)4–5 yOverweight = 15.5% (95% CI 13.9–17.0)Obese = 14.5% (95% CI 12.9–16.0)4–5 y Overweight + Obese = 30%6–11 yOverweight = 15.3% (95% CI 14.7–16.0)Obese = 23.9% (95% CI 23.1–24.6)6–11 y Overweight + Obese = 39.2%2011–2012 (sex not reported separately for age)4–5 yOverweight = 12.5% (95% CI 10.6–14.3)Obese = 16.5% (95% CI 14.4–18.6)4–5 y Overweight + Obese = 29%6–11 yOverweight = 16.3% (95% CI 15.6–17.0)Obese = 23.6% (95% CI 22.8–24.4)6–11 y Overweight + Obese = 39.9%2012–2013 (sex not reported separately for age)4–5 yOverweight = 13.1% (95% CI 11.6–14.6)Obese = 14.0% (95% CI 12.4–15.5)4–5 y Overweight + Obese = 27.1%6–11 yOverweight = 15.8% (95% CI 15.1–16.5)Obese = 23.0% (95% CI 23.1–24.7)6–11 y Overweight + Obese = 38.8%2013–2014 (sex not reported separately for age)4–5 yOverweight = 13.5% (95% CI 12.1–15.0)Obese = 12.3% (95% CI 11.0–13.7)4–5 y Overweight + Obese = 25.8%6–11 yOverweight = 15.8% (95% CI 15.2–16.4)Obese = 23.0% (95% CI 22.3–23.8)6–11 y Overweight + Obese = 38.8%GuamMatanane (2017)2012–20132–8 years466CDCTotalOverweight = 16%Obese = 13%Total Overweight + Obese = 29%MalesOverweight = 13%Obese = 14%Males Overweight + Obese = 27%FemalesOverweight = 18%Obese = 12%Females Overweight + Obese = 30%GuamLeon Guerrero (2020)20132–8 years865Obesity: Each child was measured for weight, height, and waist circumference (WC). These measures were used to compute Body Mass Index (BMI) [weight (kg)/height(m)2], WC (cm) to height (cm) ratios, and subsequent BMI z-score, BMI-for-age-percentiles, and waist circumference-for-age percentiles [24,25]. BMI percentiles and z-scores were calculated according to CDC reference data [26], and BMI categories were assigned accordingly: underweight (<5th percentile), healthy weight (5th–84th percentile), overweight (85th–94th percentile), and obese (≥95th percentile)Abdominal obesity was defined as a waist circumference greater than the International Diabetes Federation (IDF) cut-point (>90.2 percentile for age and sex)SES: Income levelEthnicity: Ethnicity was assessed using a structured survey instrument developed for the Children’s Healthy Living (CHL) study. Participants were permitted to select multiple race/ethnic categories. Classification procedures prioritized ethnic groups indigenous to the specific jurisdiction.For Guam, the indigenous ethnic group is Chamoru. Children who selected CHamoru—regardless of additional ethnic identifications—were classified as Chamoru. Children who did not identify as CHamoru but selected one or more Micronesian Pacific Islander ethnicities (e.g., Chuukese, Palauan, Yapese, Carolinian, Marshallese, or Kosraean) were categorized as Other Micronesian. This grouping was used due to the small sample sizes of these ethnic groups and the absence of Polynesian identifiers (e.g., Native Hawaiian, Samoan) in the Guam dataset.Rurality: NRCDC (BMI)IDF (WC)CDC (BMI)Overweight + ObeseTotal Overweight + Obese = 27.4% (95% CI 24.3–30.4)Males Overweight + Obese = 27.5% (95% CI 23.3–31.8)Females Overweight + Obese = 27.2% (95% CI 22.8–31.6)High SES Overweight + Obese = 21.7% (95% CI 11.8–31.7)Medium SES Overweight + Obese = 32.6% (95% CI 22.7–42.5)Low SES Overweight + Obese = 26.7% (95% CI 17.2–36.3)Very low SES Overweight + Obese = 29.3% (95% CI 24.4–34.2)Ethnicity (Pacific Islander) Overweight + Obese = 26.0% (95% CI 22.3–29.7)Ethnicity (Other Micronesian) Overweight + Obese = 28.9% (95% CI 22.7–35.2)Ethnicity (Filipino) Overweight + Obese = 33.8% (95% CI 23.0–44.6)Ethnicity (Other) Overweight + Obese = 20.0% (95% CI −10.2 to 50.2)2–5 y Overweight + Obese = 24.7% (95% CI 20.7–28.8)6–8 y Overweight + Obese = 30.4% (95% CI 25.8–35.0)IDF (WC)ObeseTotal = 8.9%Ethnicity (Pac Island) = 8.7%Ethnicity (Oth Micro) = 8.0%Ethnicity (Filipino) = 12.5%Ethnicity (Other) = 10.0%New CaledoniaTubert-Jeannin (2018)2011–20125–12 years2734Obesity: Assistants used tape measures and SECA digital scales. Children wore light clothing but no shoes. Body mass index (BMI in kg/m2) and waist-to-height ratio (WtHR) were calculated.SES: NREthnicity: Ethnicity was assessed through a questionnaire completed by 12-year-old participants, which collected socio-demographic and behavioural variables, including ethnic background. The population in New Caledonia includes ethnic groups such as: Melanesian/Polynesian/New Caledonian/OthersRurality: NRIOTFTotal Obese = 19.7%Males Obese = 18.9%Females Obese = 20.4%Ethnicity (white) Obese = 16.0%Ethnicity (Asian) Obese = 22.1%Ethnicity (>1 race) Obese = 20.2%Ethnicity (Pacific Islander) Obese = 25.1%6 y Overweight = 10.8% (95% CI 8.8–13.3)6 y Obese = 7.8% (95% CI 6.0–9.9)6 y Overweight + Obese = 18.6%9 y Overweight = 18.1% (95% CI 15.6–21.0)9 y Obese = 11.4% (95% CI 9.3–13.8)9 y Overweight + Obese = 29.5%12 y Overweight = 22.2% (95% CI 19.9–24.7)12 y Obese = 20.5% (95% CI 18.2–22.9)12 y Overweight + Obese = 42.7%New CaledoniaFrayon (2020)2015–20165–12 years699Obesity: Height was measured to the nearest 0.1 cm using a portable stadiometer. Weight was determined using a scale to nearest 0.1 kg, with adolescents being weighed in light clothing. Body mass index BMI was calculated.SES: SES was indexed based on the occupation of the household reference person (defined as the householder with the highest income) using the National Statistics Socio-Economic Classification (Krieger, 2003). In this study, three categories were retained: managerial and professional occupations (higher SES), intermediate occupations (intermediate SES), and routine and manual occupations (lower SES).Ethnicity: Ethnicity was self-reported by the adolescents using an anonymous survey and categorized as recommended in the INSERM report on New Caledonia (INSERM 2008)Rurality: In line with the data from the last census in New Caledonia (ISEE 2014), the degree of urbanization was determined using a European standard (European Communities 2003). Densely populated areas comprising at least 50,000 inhabitants in a continuous zone with more than 500 inhabitants per km2 were classified as urban. Areas with more than 50,000 inhabitants in a continuous area of over 100 inhabitants per km2 or areas adjacent to an urban area were classified as semi-urban. Rural areas were defined as those areas that did not fulfill the conditions required to qualify as urban or semi-urbanIOTFTotal Overweight + Obese = 33.8%Males Overweight = 19.1%Males Obese = 12.4%Males Overweight + Obese = 31.5%Females Overweight = 23.6%Females Obese = 12.2%Females Overweight + Obese = 35.8%Urban Overweight = 14.2%Urban Obese = 6.6%Urban Overweight + Obese = 20.8%Rural Overweight = 24.0%Rural Obese = 14.3%Rural Overweight + Obese = 34.3%High SES Overweight = 16.0%High SES Obese = 7.5%High SES Overweight + Obese = 23.5%Medium SES Overweight = 21.9%Medium SES Obese = 13.5%Medium SES Overweight + Obese = 35.4%Low SES Overweight = 24.9%Low SES Obese = 14.9%Low SES Overweight + Obese = 39.8%Ethnicity (white) Overweight = 14.2%Ethnicity (white) Obese = 7.9%Ethnicity (white) Overweight + Obese = 22.1%Ethnicity (>1 race) Overweight = 24.8%Ethnicity (>1 race) Obese = 13.2%Ethnicity (>1 race) Overweight + Obese = 38%Ethnicity (Pacific Island) Overweight = 30.6%Ethnicity (Pacific Island) Obese = 30.6%Ethnicity (Pacific Islander) Overweight + Obese = 61.2%5–8 y Overweight = 23.8%5–8 y Obese = 16.4%5–8 y Overweight + Obese = 40.2%>8 y Overweight = 20.1%>8 y Obese = 9.9%>8 y Overweight + Obese = 30.0%New CaledoniaWattelez (2020)2018–20195–12 years1011Obesity: BMI. International Obesity Task Force tables were used to determine weight status categories (underweight, normal, overweight, obese) based on sex, age, and BMI.SES: NREthnicity: NRRurality: NRIOTFMales Overweight = 16%Males Obese = 12%Males Overweight + Obese = 28%Females Overweight = 25%Females Obese = 8%Females Overweight + Obese = 33%Papua New GuineaMoase (1987)Not reportedGrades 5 and 61028WHO standards15% of sample were above standard BMIPapua New Guinea and VanuatuHo (2021)2012–20143–5 yearsPNG = 20Van = 540Obesity: Children’s height, weight, mid-upper arm circumference (AC), and triceps skinfold (TS) thickness were measured with a protocol used in the WHO Multicenter Growth Reference Study. Height was measured to the nearest 1 mm, and weight was measured to the nearest 0.1 kg using a standard stadiometer and scale. Mid-upper AC and TS thickness were measured using standard tape measures and callipers to the nearest 1 mm.SES: Analysis didn’t use SES against country-specific data.Ethnicity: NRRurality: NRWHO standards (overweight = weight-for-height z score >2)Papua New GuineaOverweight = 29.8% (95% CI 27.7–31.9)VanuatuOverweight = 53.7% (95% CI 50.3–57.1)Republic of Marshall IslandsBlankenship (2020)20172.0–4.9 years729Obesity: Children were classified as stunted when their height-for-age Z-score < −2 SD, and mothers with a BMI ≥25 kg/m 2 or a BMI ≥30 kg/m 2 were classified as overweight and obese, respectively.SES: Wealth index quintiles were calculated using principal component analysis of household characteristics from the national sample.Ethnicity: NRRurality: Geographic area was divided into 2 categories: rural and urban, as identified by the RMI 2011 censusWHO standards2.0–2.9 y Overweight + Obese = 3.5%3.0–3.9 y Overweight + Obese = 0.0%4.0–4.9 y Overweight + Obese = 0.7%Total Overweight + Obese (Birth–4.9 y) = 3.3%Republic of Marshall IslandsPassmore (2019)2017–20185–12 years3383Obesity: Weight status was categorized using age-specific body mass index percentiles (BMI-for-age) based on US Centers for Disease Control and Prevention child growth charts.16 BMI-for-age was classified as follows: underweight (under 5th percentile), healthy weight (5th to 84th percentile), overweight (85th to 94th percentile), and obese (95th percentile and above)SES: NREthnicity: NRRurality: NRCDCTotalOverweight = 8.2% (95% CI 7.3–9.2)Obese = 5.1% (95% CI 4.4–5.9)Total Overweight + Obese = 13.3%MalesOverweight = 9.1% (95% CI 7.7–10.5)Obese = 6.3% (95% CI 5.2–7.6)Males Overweight + Obese = 15.4%FemalesOverweight = 7.4% (95% CI 6.2–8.7)Obese = 3.9% (95% CI 3.1–5.0)Females Overweight + Obese = 11.3%4–6 y Overweight = 7.0% (95% CI 5.4–8.8)4–6 y Obese = 3.3% (95% CI 2.3–4.7)4–6 y Overweight + Obese = 10.3%7–9 y Overweight = 8.1% (95% CI 6.7–9.7)7–9 y Obese = 4.4% (95% CI 3.4–5.7)7–9 y Overweight + Obese = 12.5%10–12 y Overweight = 9.5% (95% CI 7.9–11.4)10–12 y Obese = 7.1% (95% CI 5.9–8.7)10–12 y Overweight + Obese = 16.6%SamoaChoy (2017)20152.0–4.9 y305Obesity: height and weight were used to calculate BMI.SES: Material lifestyle score. Categorised into four quartiles.Ethnicity: NRRurality: The study included ten villages from Upolu, specifically four from the Apia Urban Area (AUA, an urban region), three from Northwest Upolu (NWU, a peri-urban region), and three from the Rest of Upolu (ROU, a rural region).WHO standardsTotal Overweight + Obese = 16.1%Males Overweight + Obese = 16.6%Females Overweight + Obese = 15.5%Urban Overweight + Obese = 22%Rural Overweight + Obese = 14.6%Peri-urban Overweight + Obese = 11.8%High SES Overweight + Obese = 26%Medium SES Overweight + Obese = 14.4%Low SES Overweight + Obese = 14.7%Very low SES Overweight + Obese = 7.9%SamoaChoy (2020)2015 (Wave 1) and 2017 (Wave 2)2–7 years214Obesity: Standardised procedures were used to collect duplicate measures of standing height (Stadiometer Pfister Imports, New York, NY) and weight (HD 351 weight scale; Tanita Corporation of America, Arlington Heights, IL). Z-scores were calculated using WHO child growth standards or references, depending on the child’s age and sex.SES: NREthnicity: NRRurality: NRWHO standards2–4 y Overweight + Obese = 16.8%4–7 y Overweight + Obese = 24.8%SamoaThompson (2020)20173–7 y83WHO standards; IOTFWHOMalesOverweight = 27.5%Obese = 17.5%Males Overweight + Obese = 45.0%FemalesOverweight = 16.3%Obese = 4.6%Females Overweight + Obese = 20.9%IOTFMalesOverweight = 22.5%Obese = 5.0%Males Overweight + Obese = 27.5%%FemalesOverweight = 11.6%Obese = 4.7%Females Overweight + Obese = 16.3%SamoaChoy (2022)2015 (Wave 1) and 2017 (Wave 2)2–4 years197Obesity: Standardised procedures were used to collect duplicate measures of standing height (Stadiometer Pfister Imports, NY, USA) and weight (HD 351 Weight scale, Tanita Corporation of America, IL, USA) that were averaged for use in analyses.SES: Mothers were asked to report their ownership of consumer durables and calculate an 18-item household asset score, an approach that is sensitive to estimating household socioeconomic resources in Samoa. The household asset score was categorised into quartiles based on score distribution. In line with census categories, mothers reported their total annual household income category, which was dichotomized as < 10,000 versus >10,000 tala based on response distribution.Ethnicity: NRRurality: Since an initial goal was to examine variation in health status by urbanisation and nutrition transitions exposure, we aimed to recruit an equal number of pairs from three census regions: Rest of ‘Upolu (rural; low exposure to nutrition transitions), Northwest ‘Upolu (peri-urban; moderate), and Apia urban area (urban; high).WHO standardsTotal Overweight + Obese = 16.2%Males Overweight + Obese = 17%Females Overweight + Obese = 15%Urban Overweight + Obese = 13%Rural Overweight + Obese = 14%Peri-urban Overweight + Obese = 5%High SES Overweight + Obese = 43.8%Medium SES Overweight + Obese = 21.9%Low SES Overweight + Obese = 25.0%Very low SES Overweight + Obese = 9.4%SamoaChoy (2023)2019–20205–12 years437Obesity: Measured to the nearest 0.1 kg using a Tanita HD 351 digital scale, and height to the nearest 0.1 cm using a portable stadiometer.SES: NREthnicity: Based on primary caregiver report.Rurality: Based on the village. 3 rural, 3 peri-urban, and 5 Apia UrbanWHO standardsTotal Overweight + Obese = 36.2%Males Overweight = 21.2%Males Obese = 17.1%Males Overweight + Obese = 38.3%Females Overweight = 22.3%Females Obese = 11.6%Females Overweight + Obese = 33.9%Urban Overweight = 25%Urban Obese = 14.9%Urban Overweight + Obese = 39.9%Rural Overweight = 15.9%Rural Obese = 10.9%Rural Overweight + Obese = 26.8%TongaFukuyama (2005)2002–20035–11 years895Obesity: Body weight and height were measured to the nearest 0.1 kg and 0.1 cm, respectively and BMI was calculated. Skinfold thicknesses at biceps, triceps, and subscapular were measured with GPM skinfold callipers (Siber Hegner & Co. Ltd, Zurich, Switzerland), which are measurable up to 40 mm with a precision of 0.2 mm.SES: NREthnicity: NRRurality: Based on the geographical location of Tongatapu and Niuatoputapu. Participants’ location was grouped either as urban or rural, and whether they lived in rural or urban areas. Niuatoputapu Island was the rural setting for this study.IOTF and CDCAll proportions are for obesityIOTFMales (urban) = 2.6%Males (rural) = 2.1%Females (urban) = 7.1%Females (rural) = 7.0%CDCMales (urban) = 5.1%Males (rural) = 6.3%Females (urban) = 10.1%Females (rural) = 10.5%TongaSmith (2007)200111–12 years121Obesity: Height was measured using portable height scales by the stretch stature method, with shoes and headgear removed. Measurement was recorded to the nearest 0.1 cm. Weight was measured with digital bathroom scales, with shoes, heavy clothing, and personal objects removed from pockets. Measurement was recorded to the nearest 0.1 kg.SES: NREthnicity: NRRurality: NRWHO referencesMales Overweight = 28.3% (95% CI 18–39)Males Obese = 9.4% (95% CI 2.7–16.0)Males Overweight + Obese = 37.7%Females Overweight = 35.3% (95% CI 18–52)Females Obese = 4.4% (95% CI 0.5–8.3)Females Overweight + Obese = 39.7%TongaVeatupu (2019)201710–12 years36IOTFTotalOverweight = 14.3%Obese = 2.9%Total Overweight + Obese = 17.2%VanuatuDancause (2012)20076–12 years375Obesity: BMI (height and weight). Skinfolds and WC were also assessed, but not used to report overweight/obesitySES: NREthnicity: NRRurality: Anthropometric survey of children was completed from three islands varying in degree of economic development: Ambae (rural), Aneityum (rural with tourism), and Efate, home of the urban capital. On Ambae, as in most rural areas, families subsist largely through horticulture. Families on Aneityum have similar livelihoods to those on Ambae, but the island has experienced a rapid boom in the tourism industry. Families on Efate have, for many decades, had access to the benefits of urban life.WHO referencesMalesUrban Overweight + Obese = 6.0% (95% CI 2.1–16.2)Peri-Urban Overweight + Obese = 4.5% (1.8–11.0)Rural Overweight + Obese = 3.0% (0.5–15.3)FemalesUrban Overweight + Obese = 11.4% (5.9–21.0)Peri-Urban Overweight + Obese = 7.6% (3.7–14.9)Rural Overweight + Obese = 0% (0–10.7)Wallis & FutunaADS Wallis and SPC (2020)Not reported7–10 years406IOTF; WHO references; CDCIOTFMales Overweight = 24.8%Males Obese = 27.2%Males Overweight + Obese = 52.0%%Females Overweight = 24.0%Females Obese = 25.5%Females Overweight + Obese = 49.5%WHOMales Overweight = 22.3%Males Obese = 38.6%Males Overweight + Obese = 60.9%%Females Overweight = 20.6%Females Obese = 32.3%Females Overweight + Obese = 52.9%CDCMales Overweight = 13.9%Males Obese = 47.0%Males Overweight + Obese = 60.9%Females Overweight = 13.2%Females Obese = 39.2%Females Overweight + Obese = 52.4%Multiple Countries FSM, Palau, RMIPaulinoa (2017)2013–20152–8 years1201CDCTotal Overweight + Obese = 15.0% ± 2.6 (SE)aData are not reported separately for each country. Table 1Characteristics of included studies.CountryAuthorSurvey YearAge (years)Sample sizeMeasurement methods Obesity: SES: Ethnicity; RuralityDefinition used for weight statusProportion overweight, obese, and overweight + obeseAmerican SamoaGurr (2009)2006–20086–11 years2308 (2006–2007) 1805 (2008–2009)CDC2006–2007Grade KBoysOverweight = 19.9%Obese = 23.9%Males Overweight + Obese = 43.8%GirlsOverweight = 27.0%Obese = 16.1%Females Overweight + Obese = 43.1%Grade 3BoysOverweight = 17.1%Obese = 27.1%Males Overweight + Obese = 44.2%GirlsOverweight = 17.9%Obese = 25.6%Females Overweight + Obese = 43.5%Grade 6BoysOverweight = 17.4%Obese = 37.0%Males Overweight + Obese = 54.4%GirlsOverweight = 22.3%Obese = 34.5%Females Overweight + Obese = 56.8%2008–2009Grade 2BoysOverweight = 20.9%Obese = 22.6%Males Overweight + Obese = 43.5%GirlsOverweight = 22.3%Obese = 16.8%Females Overweight + Obese = 39.1%Grade 5BoysOverweight = 15.6%Obese = 34.1%Males Overweight + Obese = 49.7%GirlsOverweight = 18.8%Obese = 32.4%Females Overweight + Obese = 51.2%Commonwealth of the Northern Mariana IslandsNovotny (2007)20052–10 y346Obesity:CDCTotalOverweight = 15%Obese = 19%Total Overweight + Obese = 34%Commonwealth of the Northern Mariana IslandsPaulino (2008)20052.0–4.9 y; 5–12 years393Obesity: Height and weight were measured using similar methods and quality control procedures as Novotny and colleagues (4). Children were measured for weight using a digital scale (Seca, Hamburg, Germany). A Shorrboard (Shorr Productions, Olney, MD) was used to measure recumbent length for children younger than 24 months and height for children 24 months and older. Measurements were used to categorize children by body mass index (BMI; calculated as kg/m2)-for-age percentiles.SES: NREthnicity: NRRurality: NRCDCTotal Overweight + Obese = 33.5%2–3 y Overweight = 12%2–3 y Obese = 13%2–3 y Overweight + Obese = 25%4–6 y Overweight = 13%4–6 y Obese = 13%4–6 y Overweight + Obese = 26%7–10 y Overweight = 18%7–10 y Obese = 27%7–10 y Overweight + Obese = 45%Federated States of Micronesia (Yap)Durand (2007)2006>2–5 years; >5–10 years; >10–15 years448Obesity: BMI (height and weight)SES: NREthnicity: NRRurality: NRCDCTotal Overweight = 10.6%Total Obese = 17.6%Total Overweight + Obese = 28.2%>2–5 y Overweight = 6%>2–5 y Obese = 21%>2–5 y Overweight + Obese = 27%>5–10 y Overweight = 12%>5–10 y Obese = 16%>5–10 y Overweight + Obese = 28%>10–15 y Overweight = 12%>10–15 y Obese = 16%>10–15 y Overweight + Obese = 28%Federated States of Micronesia (Yap State)Wa’ab CHC (2006–2007)Outer Island Household Survey (2008–2009)Ichiho (2013)20112–14 y1736 (2006–07)1948 (2008–2009)Obesity: height and weight were assessed.SES: NREthnicity: NRRurality: NRCDCWa’ab CDCTotalOverweight = 12%Obese = 20%Total Overweight + Obese = 32%Outer Island Household SurveyTotalOverweight = 15.6%Obese = 18.2%Total Overweight + Obese = 33.8%MalesOverweight = 11.8%Obese = 17.7%Males Overweight + Obese = 29.5%FemalesOverweight = 18.1%Obese = 16.9%Females Overweight + Obese = 35.0%FijiNational Food & Nutrition Centre (2017)2014–20152–12 years629 (Birth–5.9 y)1288 (6–12 y)BMIEthnic group iTaukeiFID (Fijian of Indian Descent)FOD (Fijian of Other Descent)SexMaleFemaleArea typeUrbanRural>2SD ≤ 3SD (Overweight)>3SD (obese)2.0–4.9 y (59 mo)TotalOverweight = 3.0% (95% CI 1.5–4.6)Obese = 0.2% (95% CI 0.0–0.6)2.0–4.9 y Overweight + Obese = 3.2%Birth–4.9 y (59 mo)MalesOverweight = 6.3% (95% CI 3.9–8.6)Obese = 0.9% (95% CI 0.00–11.3)Males Overweight + Obese = 7.2%FemalesOverweight = 3.4% (95% CI 1.6–5.2)Obese = 1.0% (95% CI 0.0–11.3)Females Overweight + Obese = 4.4%Ethnicity iTaukeiOverweight = 5.5% (95% CI 3.5–7.4)Obese = 1.0% (95% CI 0.1–1.9)Ethnicity (iTaukei) Overweight + Obese = 6.5%FIDOverweight = 2.3% (95% CI 0.4–4.3)Obese = 0.3% (95% CI 0.0–1.2)Ethnicity (FOD) Overweight + Obese = 2.6%FODOverweight = 12.3% (95% CI 1.5–23.2)Obese = 4.2% (95% CI 0.0–11.3)Ethnicity (FOD) Overweight + Obese = 16.5%UrbanOverweight = 7.7% (95% CI 5.0–10.3)Obese = 1.4% (95% CI 0.2–2.6)RuralOverweight = 2.3% (95% CI 0.8–3.7)Obese = 0.5% (95% CI 0.0–1.3)6–12 yTotalOverweight = 7.2% (95% CI 6.7–7.6)Obese = 1.7% (95% CI 1.5–1.9)6–12 y Overweight + Obese = 8.9%MalesOverweight = 7.6% (95% CI 7.0–8.2)Obese = 2.5% (95% CI 2.1–2.8)Males Overweight + Obese = 10.1%FemalesOverweight = 6.6% (95% CI 6.0–7.2)Obese = 0.8% (95% CI 0.6–1.0)Females Overweight + Obese = 7.4%Ethnicity iTaukeiOverweight = 6.1% (95% CI 5.6–6.5)Obese = 1.4% (95% CI 1.2–1.7)Ethnicity (iTaukei) Overweight + Obese = 7.5%FIDOverweight = 9.1% (95% CI 8.3–10.0)Obese = 2.1% (95% CI 1.7–2.6)Ethnicity (FOD) Overweight + Obese = 11.2%FODOverweight = 9.9% (95% CI 7.6–12.3)Obese = 2.7% (95% CI 1.4–4.0)Ethnicity (FOD) Overweight + Obese = 12.6%UrbanOverweight = 9.6% (95% CI 8.9–10.3)Obese = 2.4% (95% CI 2.0–2.8)RuralOverweight = 5.0% (95% CI 4.6–5.5)Obese = 1.1% (95% CI 0.9–1.3)French PolynesiaDaigre (2012)2007–20085–12 years296Obesity: Body weight was measured with a Tanita BF 522 W balance (Tanita Corporation, Tokyo, Japan), height with an electronic device (Soehnle Professional GmbH, Backnang, Germany), and waist circumference with a constant-tension measuring tape placed, according to World Health Organization (WHO) recommendations, midway between iliac crests and lowermost ribs [9].SES: NREthnicity: NRRurality: NRWHO references and IOTFWHO DefinitionMales Overweight = 31.0% (SE 3.8)Males Obese = 21.4% (SE 3.4)Males Overweight + Obese = 52.4%Females Overweight = 14.6% (SE 2.9)Females Obese = 19.2% (SE 3.2)Females Overweight + Obese = 33.8%IOTF DefinitionMales Overweight = 23.4% (SE 3.5)Males Obese = 15.2% (SE 2.9)Females Overweight = 11.3% (SE 2.6)Females Obese = 16.6% (SE 3.0)French PolynesiaMinistry of Health (2015)20147–9 years1768IOTFMalesOverweight = 18.7% (95% CI 16.2–21.2)Obese = 17.1% (95% CI 14.7–19.5)Males Overweight + Obese = 35.8%FemalesOverweight = 20.5% (95% CI 17.8–23.2)Obese = 15.3% (95% CI 12.9–17.7)Females Overweight + Obese = 35.8%GuamPaulino (2015)2010–20144–5 y6–11 y2010CDC2010–2011 (sex not reported separately for age)4–5 yOverweight = 15.5% (95% CI 13.9–17.0)Obese = 14.5% (95% CI 12.9–16.0)4–5 y Overweight + Obese = 30%6–11 yOverweight = 15.3% (95% CI 14.7–16.0)Obese = 23.9% (95% CI 23.1–24.6)6–11 y Overweight + Obese = 39.2%2011–2012 (sex not reported separately for age)4–5 yOverweight = 12.5% (95% CI 10.6–14.3)Obese = 16.5% (95% CI 14.4–18.6)4–5 y Overweight + Obese = 29%6–11 yOverweight = 16.3% (95% CI 15.6–17.0)Obese = 23.6% (95% CI 22.8–24.4)6–11 y Overweight + Obese = 39.9%2012–2013 (sex not reported separately for age)4–5 yOverweight = 13.1% (95% CI 11.6–14.6)Obese = 14.0% (95% CI 12.4–15.5)4–5 y Overweight + Obese = 27.1%6–11 yOverweight = 15.8% (95% CI 15.1–16.5)Obese = 23.0% (95% CI 23.1–24.7)6–11 y Overweight + Obese = 38.8%2013–2014 (sex not reported separately for age)4–5 yOverweight = 13.5% (95% CI 12.1–15.0)Obese = 12.3% (95% CI 11.0–13.7)4–5 y Overweight + Obese = 25.8%6–11 yOverweight = 15.8% (95% CI 15.2–16.4)Obese = 23.0% (95% CI 22.3–23.8)6–11 y Overweight + Obese = 38.8%GuamMatanane (2017)2012–20132–8 years466CDCTotalOverweight = 16%Obese = 13%Total Overweight + Obese = 29%MalesOverweight = 13%Obese = 14%Males Overweight + Obese = 27%FemalesOverweight = 18%Obese = 12%Females Overweight + Obese = 30%GuamLeon Guerrero (2020)20132–8 years865Obesity: Each child was measured for weight, height, and waist circumference (WC). These measures were used to compute Body Mass Index (BMI) [weight (kg)/height(m)2], WC (cm) to height (cm) ratios, and subsequent BMI z-score, BMI-for-age-percentiles, and waist circumference-for-age percentiles [24,25]. BMI percentiles and z-scores were calculated according to CDC reference data [26], and BMI categories were assigned accordingly: underweight (<5th percentile), healthy weight (5th–84th percentile), overweight (85th–94th percentile), and obese (≥95th percentile)Abdominal obesity was defined as a waist circumference greater than the International Diabetes Federation (IDF) cut-point (>90.2 percentile for age and sex)SES: Income levelEthnicity: Ethnicity was assessed using a structured survey instrument developed for the Children’s Healthy Living (CHL) study. Participants were permitted to select multiple race/ethnic categories. Classification procedures prioritized ethnic groups indigenous to the specific jurisdiction.For Guam, the indigenous ethnic group is Chamoru. Children who selected CHamoru—regardless of additional ethnic identifications—were classified as Chamoru. Children who did not identify as CHamoru but selected one or more Micronesian Pacific Islander ethnicities (e.g., Chuukese, Palauan, Yapese, Carolinian, Marshallese, or Kosraean) were categorized as Other Micronesian. This grouping was used due to the small sample sizes of these ethnic groups and the absence of Polynesian identifiers (e.g., Native Hawaiian, Samoan) in the Guam dataset.Rurality: NRCDC (BMI)IDF (WC)CDC (BMI)Overweight + ObeseTotal Overweight + Obese = 27.4% (95% CI 24.3–30.4)Males Overweight + Obese = 27.5% (95% CI 23.3–31.8)Females Overweight + Obese = 27.2% (95% CI 22.8–31.6)High SES Overweight + Obese = 21.7% (95% CI 11.8–31.7)Medium SES Overweight + Obese = 32.6% (95% CI 22.7–42.5)Low SES Overweight + Obese = 26.7% (95% CI 17.2–36.3)Very low SES Overweight + Obese = 29.3% (95% CI 24.4–34.2)Ethnicity (Pacific Islander) Overweight + Obese = 26.0% (95% CI 22.3–29.7)Ethnicity (Other Micronesian) Overweight + Obese = 28.9% (95% CI 22.7–35.2)Ethnicity (Filipino) Overweight + Obese = 33.8% (95% CI 23.0–44.6)Ethnicity (Other) Overweight + Obese = 20.0% (95% CI −10.2 to 50.2)2–5 y Overweight + Obese = 24.7% (95% CI 20.7–28.8)6–8 y Overweight + Obese = 30.4% (95% CI 25.8–35.0)IDF (WC)ObeseTotal = 8.9%Ethnicity (Pac Island) = 8.7%Ethnicity (Oth Micro) = 8.0%Ethnicity (Filipino) = 12.5%Ethnicity (Other) = 10.0%New CaledoniaTubert-Jeannin (2018)2011–20125–12 years2734Obesity: Assistants used tape measures and SECA digital scales. Children wore light clothing but no shoes. Body mass index (BMI in kg/m2) and waist-to-height ratio (WtHR) were calculated.SES: NREthnicity: Ethnicity was assessed through a questionnaire completed by 12-year-old participants, which collected socio-demographic and behavioural variables, including ethnic background. The population in New Caledonia includes ethnic groups such as: Melanesian/Polynesian/New Caledonian/OthersRurality: NRIOTFTotal Obese = 19.7%Males Obese = 18.9%Females Obese = 20.4%Ethnicity (white) Obese = 16.0%Ethnicity (Asian) Obese = 22.1%Ethnicity (>1 race) Obese = 20.2%Ethnicity (Pacific Islander) Obese = 25.1%6 y Overweight = 10.8% (95% CI 8.8–13.3)6 y Obese = 7.8% (95% CI 6.0–9.9)6 y Overweight + Obese = 18.6%9 y Overweight = 18.1% (95% CI 15.6–21.0)9 y Obese = 11.4% (95% CI 9.3–13.8)9 y Overweight + Obese = 29.5%12 y Overweight = 22.2% (95% CI 19.9–24.7)12 y Obese = 20.5% (95% CI 18.2–22.9)12 y Overweight + Obese = 42.7%New CaledoniaFrayon (2020)2015–20165–12 years699Obesity: Height was measured to the nearest 0.1 cm using a portable stadiometer. Weight was determined using a scale to nearest 0.1 kg, with adolescents being weighed in light clothing. Body mass index BMI was calculated.SES: SES was indexed based on the occupation of the household reference person (defined as the householder with the highest income) using the National Statistics Socio-Economic Classification (Krieger, 2003). In this study, three categories were retained: managerial and professional occupations (higher SES), intermediate occupations (intermediate SES), and routine and manual occupations (lower SES).Ethnicity: Ethnicity was self-reported by the adolescents using an anonymous survey and categorized as recommended in the INSERM report on New Caledonia (INSERM 2008)Rurality: In line with the data from the last census in New Caledonia (ISEE 2014), the degree of urbanization was determined using a European standard (European Communities 2003). Densely populated areas comprising at least 50,000 inhabitants in a continuous zone with more than 500 inhabitants per km2 were classified as urban. Areas with more than 50,000 inhabitants in a continuous area of over 100 inhabitants per km2 or areas adjacent to an urban area were classified as semi-urban. Rural areas were defined as those areas that did not fulfill the conditions required to qualify as urban or semi-urbanIOTFTotal Overweight + Obese = 33.8%Males Overweight = 19.1%Males Obese = 12.4%Males Overweight + Obese = 31.5%Females Overweight = 23.6%Females Obese = 12.2%Females Overweight + Obese = 35.8%Urban Overweight = 14.2%Urban Obese = 6.6%Urban Overweight + Obese = 20.8%Rural Overweight = 24.0%Rural Obese = 14.3%Rural Overweight + Obese = 34.3%High SES Overweight = 16.0%High SES Obese = 7.5%High SES Overweight + Obese = 23.5%Medium SES Overweight = 21.9%Medium SES Obese = 13.5%Medium SES Overweight + Obese = 35.4%Low SES Overweight = 24.9%Low SES Obese = 14.9%Low SES Overweight + Obese = 39.8%Ethnicity (white) Overweight = 14.2%Ethnicity (white) Obese = 7.9%Ethnicity (white) Overweight + Obese = 22.1%Ethnicity (>1 race) Overweight = 24.8%Ethnicity (>1 race) Obese = 13.2%Ethnicity (>1 race) Overweight + Obese = 38%Ethnicity (Pacific Island) Overweight = 30.6%Ethnicity (Pacific Island) Obese = 30.6%Ethnicity (Pacific Islander) Overweight + Obese = 61.2%5–8 y Overweight = 23.8%5–8 y Obese = 16.4%5–8 y Overweight + Obese = 40.2%>8 y Overweight = 20.1%>8 y Obese = 9.9%>8 y Overweight + Obese = 30.0%New CaledoniaWattelez (2020)2018–20195–12 years1011Obesity: BMI. International Obesity Task Force tables were used to determine weight status categories (underweight, normal, overweight, obese) based on sex, age, and BMI.SES: NREthnicity: NRRurality: NRIOTFMales Overweight = 16%Males Obese = 12%Males Overweight + Obese = 28%Females Overweight = 25%Females Obese = 8%Females Overweight + Obese = 33%Papua New GuineaMoase (1987)Not reportedGrades 5 and 61028WHO standards15% of sample were above standard BMIPapua New Guinea and VanuatuHo (2021)2012–20143–5 yearsPNG = 20Van = 540Obesity: Children’s height, weight, mid-upper arm circumference (AC), and triceps skinfold (TS) thickness were measured with a protocol used in the WHO Multicenter Growth Reference Study. Height was measured to the nearest 1 mm, and weight was measured to the nearest 0.1 kg using a standard stadiometer and scale. Mid-upper AC and TS thickness were measured using standard tape measures and callipers to the nearest 1 mm.SES: Analysis didn’t use SES against country-specific data.Ethnicity: NRRurality: NRWHO standards (overweight = weight-for-height z score >2)Papua New GuineaOverweight = 29.8% (95% CI 27.7–31.9)VanuatuOverweight = 53.7% (95% CI 50.3–57.1)Republic of Marshall IslandsBlankenship (2020)20172.0–4.9 years729Obesity: Children were classified as stunted when their height-for-age Z-score < −2 SD, and mothers with a BMI ≥25 kg/m 2 or a BMI ≥30 kg/m 2 were classified as overweight and obese, respectively.SES: Wealth index quintiles were calculated using principal component analysis of household characteristics from the national sample.Ethnicity: NRRurality: Geographic area was divided into 2 categories: rural and urban, as identified by the RMI 2011 censusWHO standards2.0–2.9 y Overweight + Obese = 3.5%3.0–3.9 y Overweight + Obese = 0.0%4.0–4.9 y Overweight + Obese = 0.7%Total Overweight + Obese (Birth–4.9 y) = 3.3%Republic of Marshall IslandsPassmore (2019)2017–20185–12 years3383Obesity: Weight status was categorized using age-specific body mass index percentiles (BMI-for-age) based on US Centers for Disease Control and Prevention child growth charts.16 BMI-for-age was classified as follows: underweight (under 5th percentile), healthy weight (5th to 84th percentile), overweight (85th to 94th percentile), and obese (95th percentile and above)SES: NREthnicity: NRRurality: NRCDCTotalOverweight = 8.2% (95% CI 7.3–9.2)Obese = 5.1% (95% CI 4.4–5.9)Total Overweight + Obese = 13.3%MalesOverweight = 9.1% (95% CI 7.7–10.5)Obese = 6.3% (95% CI 5.2–7.6)Males Overweight + Obese = 15.4%FemalesOverweight = 7.4% (95% CI 6.2–8.7)Obese = 3.9% (95% CI 3.1–5.0)Females Overweight + Obese = 11.3%4–6 y Overweight = 7.0% (95% CI 5.4–8.8)4–6 y Obese = 3.3% (95% CI 2.3–4.7)4–6 y Overweight + Obese = 10.3%7–9 y Overweight = 8.1% (95% CI 6.7–9.7)7–9 y Obese = 4.4% (95% CI 3.4–5.7)7–9 y Overweight + Obese = 12.5%10–12 y Overweight = 9.5% (95% CI 7.9–11.4)10–12 y Obese = 7.1% (95% CI 5.9–8.7)10–12 y Overweight + Obese = 16.6%SamoaChoy (2017)20152.0–4.9 y305Obesity: height and weight were used to calculate BMI.SES: Material lifestyle score. Categorised into four quartiles.Ethnicity: NRRurality: The study included ten villages from Upolu, specifically four from the Apia Urban Area (AUA, an urban region), three from Northwest Upolu (NWU, a peri-urban region), and three from the Rest of Upolu (ROU, a rural region).WHO standardsTotal Overweight + Obese = 16.1%Males Overweight + Obese = 16.6%Females Overweight + Obese = 15.5%Urban Overweight + Obese = 22%Rural Overweight + Obese = 14.6%Peri-urban Overweight + Obese = 11.8%High SES Overweight + Obese = 26%Medium SES Overweight + Obese = 14.4%Low SES Overweight + Obese = 14.7%Very low SES Overweight + Obese = 7.9%SamoaChoy (2020)2015 (Wave 1) and 2017 (Wave 2)2–7 years214Obesity: Standardised procedures were used to collect duplicate measures of standing height (Stadiometer Pfister Imports, New York, NY) and weight (HD 351 weight scale; Tanita Corporation of America, Arlington Heights, IL). Z-scores were calculated using WHO child growth standards or references, depending on the child’s age and sex.SES: NREthnicity: NRRurality: NRWHO standards2–4 y Overweight + Obese = 16.8%4–7 y Overweight + Obese = 24.8%SamoaThompson (2020)20173–7 y83WHO standards; IOTFWHOMalesOverweight = 27.5%Obese = 17.5%Males Overweight + Obese = 45.0%FemalesOverweight = 16.3%Obese = 4.6%Females Overweight + Obese = 20.9%IOTFMalesOverweight = 22.5%Obese = 5.0%Males Overweight + Obese = 27.5%%FemalesOverweight = 11.6%Obese = 4.7%Females Overweight + Obese = 16.3%SamoaChoy (2022)2015 (Wave 1) and 2017 (Wave 2)2–4 years197Obesity: Standardised procedures were used to collect duplicate measures of standing height (Stadiometer Pfister Imports, NY, USA) and weight (HD 351 Weight scale, Tanita Corporation of America, IL, USA) that were averaged for use in analyses.SES: Mothers were asked to report their ownership of consumer durables and calculate an 18-item household asset score, an approach that is sensitive to estimating household socioeconomic resources in Samoa. The household asset score was categorised into quartiles based on score distribution. In line with census categories, mothers reported their total annual household income category, which was dichotomized as < 10,000 versus >10,000 tala based on response distribution.Ethnicity: NRRurality: Since an initial goal was to examine variation in health status by urbanisation and nutrition transitions exposure, we aimed to recruit an equal number of pairs from three census regions: Rest of ‘Upolu (rural; low exposure to nutrition transitions), Northwest ‘Upolu (peri-urban; moderate), and Apia urban area (urban; high).WHO standardsTotal Overweight + Obese = 16.2%Males Overweight + Obese = 17%Females Overweight + Obese = 15%Urban Overweight + Obese = 13%Rural Overweight + Obese = 14%Peri-urban Overweight + Obese = 5%High SES Overweight + Obese = 43.8%Medium SES Overweight + Obese = 21.9%Low SES Overweight + Obese = 25.0%Very low SES Overweight + Obese = 9.4%SamoaChoy (2023)2019–20205–12 years437Obesity: Measured to the nearest 0.1 kg using a Tanita HD 351 digital scale, and height to the nearest 0.1 cm using a portable stadiometer.SES: NREthnicity: Based on primary caregiver report.Rurality: Based on the village. 3 rural, 3 peri-urban, and 5 Apia UrbanWHO standardsTotal Overweight + Obese = 36.2%Males Overweight = 21.2%Males Obese = 17.1%Males Overweight + Obese = 38.3%Females Overweight = 22.3%Females Obese = 11.6%Females Overweight + Obese = 33.9%Urban Overweight = 25%Urban Obese = 14.9%Urban Overweight + Obese = 39.9%Rural Overweight = 15.9%Rural Obese = 10.9%Rural Overweight + Obese = 26.8%TongaFukuyama (2005)2002–20035–11 years895Obesity: Body weight and height were measured to the nearest 0.1 kg and 0.1 cm, respectively and BMI was calculated. Skinfold thicknesses at biceps, triceps, and subscapular were measured with GPM skinfold callipers (Siber Hegner & Co. Ltd, Zurich, Switzerland), which are measurable up to 40 mm with a precision of 0.2 mm.SES: NREthnicity: NRRurality: Based on the geographical location of Tongatapu and Niuatoputapu. Participants’ location was grouped either as urban or rural, and whether they lived in rural or urban areas. Niuatoputapu Island was the rural setting for this study.IOTF and CDCAll proportions are for obesityIOTFMales (urban) = 2.6%Males (rural) = 2.1%Females (urban) = 7.1%Females (rural) = 7.0%CDCMales (urban) = 5.1%Males (rural) = 6.3%Females (urban) = 10.1%Females (rural) = 10.5%TongaSmith (2007)200111–12 years121Obesity: Height was measured using portable height scales by the stretch stature method, with shoes and headgear removed. Measurement was recorded to the nearest 0.1 cm. Weight was measured with digital bathroom scales, with shoes, heavy clothing, and personal objects removed from pockets. Measurement was recorded to the nearest 0.1 kg.SES: NREthnicity: NRRurality: NRWHO referencesMales Overweight = 28.3% (95% CI 18–39)Males Obese = 9.4% (95% CI 2.7–16.0)Males Overweight + Obese = 37.7%Females Overweight = 35.3% (95% CI 18–52)Females Obese = 4.4% (95% CI 0.5–8.3)Females Overweight + Obese = 39.7%TongaVeatupu (2019)201710–12 years36IOTFTotalOverweight = 14.3%Obese = 2.9%Total Overweight + Obese = 17.2%VanuatuDancause (2012)20076–12 years375Obesity: BMI (height and weight). Skinfolds and WC were also assessed, but not used to report overweight/obesitySES: NREthnicity: NRRurality: Anthropometric survey of children was completed from three islands varying in degree of economic development: Ambae (rural), Aneityum (rural with tourism), and Efate, home of the urban capital. On Ambae, as in most rural areas, families subsist largely through horticulture. Families on Aneityum have similar livelihoods to those on Ambae, but the island has experienced a rapid boom in the tourism industry. Families on Efate have, for many decades, had access to the benefits of urban life.WHO referencesMalesUrban Overweight + Obese = 6.0% (95% CI 2.1–16.2)Peri-Urban Overweight + Obese = 4.5% (1.8–11.0)Rural Overweight + Obese = 3.0% (0.5–15.3)FemalesUrban Overweight + Obese = 11.4% (5.9–21.0)Peri-Urban Overweight + Obese = 7.6% (3.7–14.9)Rural Overweight + Obese = 0% (0–10.7)Wallis & FutunaADS Wallis and SPC (2020)Not reported7–10 years406IOTF; WHO references; CDCIOTFMales Overweight = 24.8%Males Obese = 27.2%Males Overweight + Obese = 52.0%%Females Overweight = 24.0%Females Obese = 25.5%Females Overweight + Obese = 49.5%WHOMales Overweight = 22.3%Males Obese = 38.6%Males Overweight + Obese = 60.9%%Females Overweight = 20.6%Females Obese = 32.3%Females Overweight + Obese = 52.9%CDCMales Overweight = 13.9%Males Obese = 47.0%Males Overweight + Obese = 60.9%Females Overweight = 13.2%Females Obese = 39.2%Females Overweight + Obese = 52.4%Multiple Countries FSM, Palau, RMIPaulinoa (2017)2013–20152–8 years1201CDCTotal Overweight + Obese = 15.0% ± 2.6 (SE)aData are not reported separately for each country. Characteristics of included studies. Data are not reported separately for each country. Agreement between authors was high during title and abstract screening. Across the three principal reviewer pairings, which each involved between 819 and 1662 records, the proportionate agreement ranged from 92.2% to 95.1%, and Cohen’s κ ranged from 0.08 to 0.37. When decisions from these three pairings were combined, reviewers agreed on 93.5% of 4151 paired screening decisions. The comparatively modest κ values despite high absolute agreement reflected the low proportion of records judged potentially eligible at this stage. Agreement was more variable during full-text review. Among the three principal reviewer pairings, involving between 38 and 80 articles, proportionate agreement ranged from 39.5% to 87.5%, and Cohen’s κ ranged from 0.05 to 0.43. Across these pairings, reviewers agreed on 67.4% of 178 paired full-text decisions. Two additional reviewer pairings involved only four and five articles, respectively, and therefore produced unstable agreement estimates. All discrepancies at both screening stages were resolved through consensus. Of the 29 included articles, 20 (71%) were designed to be representative of either a national population, an entire territory, or a defined geographical region. Seventeen studies aimed to obtain samples representative of a country, territory, or population of interest, while a further three studies were designed to represent specific sub-national regions or census areas. Three studies were not intended to be representative of the broader population, one study used a convenience sample, and four studies did not report whether the sampling strategy was designed to achieve representativeness. Overall, most studies employed population-based sampling approaches, although the representativeness of the evidence varied across PICTs. Twenty studies identified the personnel who conducted the anthropometric assessments, although only 11 provided information about the training or standardisation received by data collectors. Response rates were reported for 15 studies and, where a numerical percentage was available, generally ranged from approximately 62%–97%. Total overweight and obesity Across PICTs, the prevalence of total overweight and obesity among children varied substantially according to the definition used and the age group. Using WHO definitions, the prevalence among children aged 2.0–4.9 years ranged from 3.3% in RMI17 to 16.1–16.2% in Samoa.18,19 In primary school-aged children (5–12 years), WHO-defined prevalence was higher at 36.2% in Samoa.20 Among young children (birth to <5 years), the prevalence was 33.5% in CNMI21 and 15.0% (±2.6 SE) across the FSM, Palau, and RMI.22 In older children (5–12 years), combined totals were 13.3% in RMI,23 27.4% in Guam,24 and 29.0% in another Guam sample.25 In CNMI, the prevalence was 34.0%,26 and in Yap (FSM), it was 28.2%.27 The IOTF-defined estimates were generally comparable or slightly lower than CDC estimates but higher than WHO estimates for similar age groups. In New Caledonia, the prevalence among children aged 5–12 years was 33.8%,28 while in Tonga, among those aged 10–12 years, the prevalence was 17.2%.29 Overweight and obesity by sex Sex-specific differences in overweight and obesity prevalence were reported in 17 studies, with notable variation by age group and country. Among children aged 2.0–4.9 years, the prevalence was generally higher among boys. In Samoa, combined overweight and obesity were 16.6% in boys and 15.5% in girls18 and 17% and 15%, respectively, in a later sample.19 In Fiji, among children under five years, the prevalence was 7.2% in boys and 4.4% in girls.30 Among school-aged children, patterns were mixed but tended to show slightly higher prevalences in boys. In Samoa, 38.3% of boys and 33.9% of girls had overweight or obesity,20 and in Tonga, the rates were 37.7% among boys and 39.7% among girls aged 11–12 years.31 In French Polynesia, sex differences were small, with 35.8% of both boys and girls classified as having overweight or obesity,32 while in Wallis and Futuna, prevalence was high for both sexes—49.5%–60.9% depending on the definition used.33 In New Caledonia, rates were 31.5% for boys and 35.8% for girls using IOTF criteria,28 and in a separate study of similar-aged children, 28% of boys and 33% of girls.34 In Guam, the prevalence was nearly identical between boys (27.5%) and girls (27.2%).24,25 The Department of Health and Education surveillance in American Samoa found very high levels across primary school ages, ranging from 43.5% to 54.4% in boys and 39.1%–56.8% in girls.35 In the Federated States of Micronesia, rates were 29.5% in boys and 35.0% in girls,36 and in the Marshall Islands, 15.4% in boys and 11.3% in girls.23 In Fiji, among children aged 6.0–12.0 years, the prevalence was 10.1% in boys and 7.4% in girls.30 Overweight and obesity by age Across the 10 studies that reported results by age group, the prevalence of overweight and obesity generally increased with age, although this pattern was not uniform across all PICTs. In CNMI, 25% of 2–3-year-olds and 26% of 4–6-year-olds had overweight or obesity, increasing to 45% among 7–10-year-olds.21 In Guam, the prevalence was 24.7% among 2–5-year-olds and 30.4% among 6–8-year-olds.24 Similarly, in Samoa, longitudinal data showed that overweight and obesity affected 16.8% of 2–4-year-olds and 24.8% of 4–7-year-olds.37 In studies with broader childhood age ranges, similar age-related gradients were evident. In the Marshall Islands, the combined prevalence increased from 10.3% among 4–6-year-olds to 12.5% among 7–9-year-olds and 16.6% among 10–12-year-olds.23 In New Caledonia, the prevalence declined slightly with age, from 40.2% among 5–8-year-olds to 30.0% among those older than eight years,28 whereas in the same setting, another study found rates rising steadily from 18.6% at age six to 29.5% at age nine and 42.7% at age 12.38 In Guam, the prevalence among 4–5-year-olds ranged between 25.8% and 30.0% across survey years, compared with 38.8–39.9% among 6–11-year-olds.39 In FSM, combined overweight and obesity remained relatively stable across successive age bands, with 27% among children aged >2.0–5.0 years, 28% in >5.0–10.0 years, and 28% in >10.0–15.0 years.27 In Fiji, overweight and obesity were 3.2% among children aged 2.0–5.9 years and 8.9% among those aged 6.0–12.0 years.30 Overweight and obesity by SES Socioeconomic disparities in the prevalence of overweight and obesity were evident in the four studies that reported results by household socioeconomic status. In Samoa, among children aged 2.0–4.9 years, combined overweight and obesity was markedly higher in those from high socioeconomic households (26%) compared with those from medium- (14.4%), low- (14.7%), and very low-socioeconomic groups (7.9%).18 A later Samoan study using an expanded asset-based index similarly found a strong gradient, with prevalence highest in the highest socioeconomic quartile (43.8%) and lowest in the very low-socioeconomic group (9.4%).19 In New Caledonia, among children aged 5–12 years, total overweight and obesity increased from 23.5% in the high-socioeconomic group to 35.4% in the medium-socioeconomic group and 39.8% in the low-socioeconomic group.28 By contrast, in Guam, where children were aged 2–8 years, the pattern was less consistent: prevalence was lowest in the high-income group (21.7%), highest in the medium-income group (32.6%), and intermediate in the low- and very low-income groups (26.7% and 29.3%, respectively).24 Overweight and obesity by rurality Geographical differences in the prevalence of childhood overweight and obesity were observed across the six studies that reported data by urban and rural location. In Samoa, the prevalence was consistently higher in urban areas compared with rural and peri-urban regions. Among children aged birth to <5 years, the prevalence was 22% in urban areas, 11.8% in peri-urban areas, and 14.6% in rural areas.18 Similar patterns were seen in a later cohort of the same age group, where urban children had a prevalence of 13% compared with 5% in peri-urban areas and 14% in rural areas.19 In school-aged Samoan children (5–12 years), the difference was even more pronounced, with 39.9% of urban children classified as having overweight or obesity compared with 26.8% of rural children.20 In Vanuatu, where three settings differing in economic development were examined, prevalence among boys increased with urbanisation from 3.0% in rural areas to 4.5% in peri-urban areas and 6.0% in urban areas, while among girls the corresponding rates were 0%, 7.6%, and 11.4%.40 A similar trend was observed in New Caledonia, where the prevalence among 5–12-year-olds was 20.8% in urban areas and 34.3% in rural areas,28 indicating a possible reversal of the typical urban-rural gradient seen elsewhere. In Fiji, urban children had higher rates of overweight and obesity than rural children across both age groups: among those aged birth to 5.9 years, 9.1% were overweight or obese compared with 2.8% of rural children, and among those aged 6.0–12.0 years, 12.0% of urban children compared with 6.1% of rural children.30 Overweight and obesity by ethnicity Ethnic differences in childhood overweight and obesity were reported in three studies. In New Caledonia, among children aged 5–12 years, prevalence varied considerably across ethnic groups, from 22.1% among children of European (White) descent to 38.0% among those identifying with more than one race, and 61.2% among Pacific Islander children.28 In Guam, among children aged 2–8 years, combined overweight and obesity prevalence was 26.0% among CHamoru (Pacific Islander) children, 28.9% among Other Micronesian children, 33.8% among Filipino children, and 20.0% among those from other ethnic backgrounds.24 In Fiji, ethnic differences were evident in both early and middle childhood. Among children aged birth to 5.9 years, combined overweight and obesity prevalence was 6.5% among iTaukei, 2.6% among Fijians of Indian descent (FID), and 16.5% among Fijians of other descent (FOD). Among those aged 6.0–12.0 years, prevalence was 7.5% for iTaukei, 11.2% for FID, and 12.6% for FOD.30 Discussion Our systematic review found a high burden of childhood overweight and obesity in PICTs. Prevalence estimates varied widely between settings, ranging from relatively low levels in some Melanesian countries to extremely high levels in certain Polynesian and Micronesian populations.1,41 In several PICTs, more than one-third of children had overweight or obesity, with the highest reported prevalence over 50%. This indicates that in some Pacific communities, overweight and obesity have become highly prevalent among children. By comparison, recent global estimates suggest that approximately 8–10% of children are living with obesity globally.1 Notably, our review found that older children had a higher prevalence than younger ones, suggesting that weight problems compound with age during late childhood. Differences by sex were observed but were not consistent across studies. In some PICTs, boys showed slightly higher obesity rates, whereas in others, girls were equally or more affected. These findings are consistent with the NCD-RisC analysis of 200 countries and territories, which identified Pacific Island populations as having some of the highest BMI levels among children and adolescents worldwide while also highlighting substantial heterogeneity between countries.1 Our review extends these global observations by demonstrating considerable variation within PICTs according to age, urbanicity, socioeconomic position, and ethnicity, underscoring the need for context-specific surveillance and interventions. Socio-demographic patterning of child obesity was also evident in relation to SES, ethnicity, and rurality, although the relevance and variability of these factors likely differ across PICTs depending on local demographic, geographic, and socioeconomic contexts. Children from urbanised or higher-income environments were at greater risk, likely due to increased access to energy-dense foods and more sedentary lifestyles.25,42 Where socioeconomic gradients were observed, higher obesity prevalence among children from more advantaged households highlights the complex socioeconomic dynamics operating in some PICTs, where education and income may facilitate access to imported processed foods and sedentary entertainment, thereby increasing obesity risk despite greater parental knowledge.41 This pattern is also consistent with the Obesity Transition framework, which proposes that in the earlier stages of the obesity epidemic, obesity is often concentrated among higher socioeconomic groups before progressively shifting towards more disadvantaged populations as countries develop.43 We found that indigenous Polynesian or Micronesian children had higher obesity rates than those of Asian or European descent living in the same country. These patterns align with a broader trend in PICTs whereby Polynesian and Micronesian populations experience greater obesity burdens compared to Melanesian populations,44 although rising rates were seen across all groups. We observed multiple PICTs with over half of their older children having overweight or obesity, a proportion unheard of in most other regions.45 Our review also corroborates findings from multi-country surveillance initiatives. The WHO Western Pacific Region has warned that childhood obesity is increasing at an “alarming” rate in the Pacific.1 Data from the Global School-Based Student Health Surveys and other youth surveys in PICTs consistently show a high prevalence of obesity among school-age children and adolescents. For example, the Youth Risk Behaviour Survey in American Samoa found that about 40% of adolescent students were obese, and surveys in Guam and the Marshall Islands have reported roughly 25% obesity in adolescents.45 These figures are comparable with, or higher than, those reported in the worst-affected populations globally. They highlight that Pacific children and youth are bearing a disproportionate burden of obesity-related health risks. Our findings and the global literature both note a rapid pace of change. While global child obesity rates have increased fourfold in the past 40 years, PICTs appear to be undergoing an even more accelerated increase. Longitudinal data in our review from Samoa found that the prevalence of overweight and obesity was 16.8% among children aged 2–4 years and 24.8% when the same cohort was reassessed only two years later, indicating a higher prevalence at older ages within this cohort.37 Such steep rises within a short time frame are rarely seen elsewhere and highlight the intensity of the nutrition transition occurring in the Pacific. Our findings can be interpreted in light of the nutrition transition and broader social determinants of health. Over recent decades, PICTs have undergone profound changes in lifestyle and environment. Traditional diets based on local staples and high physical activity levels from active transportation and household and community activities have progressively given way to diets high in imported, ultra-processed foods and more sedentary living.46 These environmental and cultural shifts help explain the rising obesity among even very young children. Easy availability of energy-dense snacks, infant formula, and sugary drinks, combined with declines in traditional food consumption, has created an obesogenic environment in many PICTs.47 Notably, Pacific Island children are now exposed to intensive marketing of highly processed food and beverages, as global food companies extend their reach. UNICEF reports that across East Asia and the Pacific, children are targeted with advertisements for ultra-processed or ultra-formulated, high-sugar products via television, social media, and even in schools.48 This commercial pressure is contributing to excessive sugar, fat, and sodium intake (hyperpalatable foods).49 In Pacific islands, where regulatory controls on food marketing are often weak, the impact on children’s diets is likely substantial. The high prevalence of soft drink and instant noodle consumption noted in some Pacific surveys supports this hypothesis.41 Our findings of higher child overweight in urban and higher-SES groups likely reflect greater access to shelf-stable, lower-cost, and convenient processed foods and more sedentary entertainment (television, computers) in those populations. However, as a study in our review from Guam suggests, even education and awareness do not automatically protect against obesity when unhealthy food environments prevail.39 In some contexts, higher parental education was associated with higher child BMI, possibly because more affluent or educated families can afford more store-bought treats, fast foods, and screen-based devices, highlighting how obesogenic environments can override individual knowledge or intentions. Additionally, Given the high frequency of natural disasters and their substantial impacts on food systems and infrastructure in PICTs, access to shelf-stable foods is critical for disaster preparedness and response.50 An additional contributor to the high prevalence of childhood overweight and obesity in PICT settings is the intergenerational transmission of risk through maternal factors. Maternal obesity, gestational diabetes, excessive gestational weight gain, and related metabolic influences may increase a child’s susceptibility to obesity through both biological and shared environmental pathways.51 Although genetic factors contribute to individual susceptibility to obesity, the rapid rise in prevalence across recent decades and the variability by urbanity, socioeconomic status, and ethnicity suggest that environmental, social, cultural, and commercial determinants contribute more in shaping population patterns of obesity.52 Efforts to explain and address childhood obesity in PICTs should focus primarily on these modifiable determinants rather than on presumed genetic differences between populations. The high prevalence of childhood obesity in PICTs and its apparent sociocultural determinants have implications for policy and public health action. Foremost, our findings reinforce the need for comprehensive, multi-sectoral strategies to improve food environments and support healthy behaviours. Many Pacific governments have recognised this. All 22 PICTs endorsed the establishment of a regional network (Pacific ECHO—Ending Childhood Obesity network) in 2017 to coordinate action on this issue.53 However, progress has been patchy and inconsistent. For example, very few Pacific jurisdictions currently have robust regulations on the marketing of unhealthy foods and beverages to children. As of 2019, only two PICTs (French Polynesia and Kiribati) had implemented policies to restrict junk food advertising aimed at children.53 This is a clear area for improvement, as protecting children from the aggressive marketing of ultra-processed foods could help reduce demand for those products. Another policy lever is fiscal measures to influence food choices. Several PICTs have introduced taxes on sugar-sweetened beverages and unhealthy foods or provided subsidies to improve access to fruits and vegetables. Between 2017 and 2018, 13 PICT governments had some form of food taxation policy in place, but only four were rated as having strong, adequately scaled measures.53 Strengthening and expanding these fiscal policies would be beneficial. Evidence from other regions shows that raising the price of sugary drinks and junk food can curb consumption, especially among youth.54 Complementary measures, such as front-of-pack nutrition labelling and standards for school canteens, should also be prioritised to create healthier food environments for children. Schools are a crucial setting for intervention. Policies ensuring that school meals and snacks are nutritious, limiting the sale of sugary drinks on campus, and incorporating nutrition and physical education into curricula can directly shape healthier habits in children. An important implication of this review is the need to strengthen surveillance systems for childhood obesity. We found that epidemiological data on child obesity in PICTs are limited and often not directly comparable due to non-standardised methods.41 Published evidence of regular, standardised monitoring of child weight status beyond infancy was limited, making it difficult to track trends or evaluate interventions consistently across PICTs. While some surveillance data may exist in government reports or other grey literature, comparable and routinely reported data remain scarce. The Pacific ECHO network has highlighted the establishment of consistent primary-school BMI surveillance as a priority. Based on the available evidence identified in this review, surveillance systems that generate standardised and routinely reported data appear to remain patchy across many PICTs. Improving surveillance will require investing in routine data collection. For example, incorporating annual or biannual height and weight measurements in schools or health clinics. Standardised protocols aligned with WHO growth references should be adopted region-wide so that prevalence estimates are comparable. The use of different growth references (predominantly WHO and CDC) across studies is also an important source of heterogeneity. While these systems are broadly comparable for many children, they are not identical and may generate different prevalence estimates, limiting direct comparisons between studies and reinforcing the need for a consistent regional surveillance standard. Better data will enable policymakers to identify regions or demographic groups with increasing obesity and respond with targeted interventions. It will also allow for the evaluation of policy impacts. Importantly, surveillance data must be fed back into policy. Governments and stakeholders need to make use of the evidence to adapt strategies. Without robust, standardised, and routinely reported data, policymakers are limited in their ability to monitor trends, evaluate interventions, and adapt strategies to address childhood obesity effectively. Limitations This review has several limitations. First, the data sources were heterogeneous. We included both published studies and national survey reports, which used varying definitions of overweight and obesity and covered different age ranges. This heterogeneity complicates direct comparisons between studies. For example, some studies reported overweight, including obesity as a combined metric, while others reported obesity alone, and age groupings were not uniform across all reports. We attempted to standardise categories where possible, but residual inconsistencies remain. Harmonisation algorithms that convert prevalence estimates between different growth references (e.g., WHO, IOTF, and CDC) may improve comparability across studies in future reviews, although their application requires appropriate age- and sex-specific data and further validation.55 Our findings highlighted the paucity of surveys and non-standardised methodology in PICTs. There is a risk of measurement bias if one country’s survey measured children at a younger age (yielding lower BMI on average) while another measured older children. We mitigated this by stratifying results by age, where data allowed, but comparisons should still be made cautiously. Second, the coverage of PICTs in the literature is uneven. There is likely publication bias and data gap bias in the countries that have performed studies; often, the ones with more severe obesity problems or more resources appear frequently in the literature, whereas others have little to no data. For instance, we found multiple studies from Samoa, Fiji, and Guam, but none from some smaller or less resourced nations. Some community surveys or unpublished data may exist that we could not obtain, especially if not in English. Our search included grey literature, but we might still have missed fragmented data, such as a local survey of one island’s schools. As such, our review might over-represent the situation in certain PICTs and under-represent others. The overall regional estimates should therefore be interpreted as indicative rather than definitive. Third, there is the issue of temporality—some data are relatively old. A few national surveys date back 10–15 years. Given the rapid increase in obesity, those might underestimate the current prevalence. Conversely, some high-prevalence reports, like very high adolescent obesity in certain micro-states, could reflect unique circumstances or small sample sizes. We tried to account for study quality, but the overall evidence base in the Pacific is limited. Additionally, our review did not include children above 12 or adolescents in detail, though we referenced some adolescent data for context. Their exclusion means we did not capture the full trajectory into adolescence, which is when obesity rates become even higher—but that was outside our scope. Finally, as with any systematic review, our conclusions depend on the quality of the underlying studies. Many studies had modest sample sizes or were geographically limited to one province or island group within a country. This raises the possibility of sampling bias. There is also the concern of language and indexing bias. Relevant studies published in French for territories like New Caledonia or those not indexed in common databases might have been overlooked. These limitations largely stem from the need for more and better data—a finding in itself that points to improving surveillance and research capacity as a priority. Conclusion Our systematic review confirms an urgent public health challenge, as PICTs are confronted with an epidemic of childhood overweight that varies by sociodemographic factors yet is pervasive across the region. Compared with global trends, Pacific children are some of the most affected in the world by excess weight. Pacific governments and communities will need to implement bold, multi-faceted interventions that improve food environments, encourage active lifestyles, and strengthen health and nutrition systems. Enhanced surveillance and research are also critical to monitor progress and tailor strategies. Childhood obesity poses a serious threat to the health and well-being of Pacific peoples, but with concerted action—aligning with global recommendations and anchored in Pacific contexts—there is an opportunity to turn the tide. The findings of this review provide an evidence base to inform such actions, underscoring that the time to act on childhood overweight in PICTs is now, before the current generation carries this heavy burden into adulthood and further entrenches the NCD crisis in the region. During the preparation of this work the author(s) used Chat GPT and Claude AI in order to revise text that had been written by the authors for clarity, brevity, grammar and consistency. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article. Contributors SR registered the review with PROSPERO. AO, SV, RE and SR screened titles and abstracts, performed data extraction, and independently assessed methodological quality and risk of bias for each study. OG provided interpretation and advice on articles and databases in French. AO drafted the initial version of the MS. All authors provided feedback on this and on subsequent versions. All authors approved the final version. Declaration of interests RN is an associate editor for Nutrition Reviews; she received payment from ILSI for manuscript edit. AO is a member of the Research Advisory Committee for Health and Wellbeing Queensland, he has received consulting fees from World Bank and honorarium for a presentation at the Capital Institute of Pediatrics in Beijing. The rest of the authors declare no conflict of interest.