Skip to main content
PMC home page
International Journal of Obesity Supplements logoLink to International Journal of Obesity Supplements
. 2015 Dec 8;5(Suppl 2):S9–S16. doi: 10.1038/ijosup.2015.13

Are participant characteristics from ISCOLE study sites comparable to the rest of their country?

A G LeBlanc 1,2, P T Katzmarzyk 3, T V Barreira 3,4, S T Broyles 3, J-P Chaput 1,2, T S Church 3, M Fogelholm 5, D M Harrington 6, G Hu 3, R Kuriyan 7, A Kurpad 7, E V Lambert 8, C Maher 9, J Maia 10, V Matsudo 11, T Olds 9, V Onywera 12, O L Sarmiento 13, M Standage 14, C Tudor-Locke 3,15, P Zhao 15, M S Tremblay 1,*; for the ISCOLE Research Group16
PMCID: PMC4850614  PMID: 27152192

Abstract

OBJECTIVES:

The International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE) provides robust, multi-national information on physical activity, diet and weight status in 9–11-year-old children around the world. The purpose of this analysis was to examine the similarities and differences between participant characteristics from ISCOLE sites and data from nationally representative surveys from ISCOLE countries (Australia, Brazil, Canada, China, Colombia, Finland, Kenya, India, Portugal, South Africa, the United Kingdom and the United States).

METHODS:

Distributions of characteristics were assessed within each ISCOLE country-level database, and compared with published data from national or regional surveys, where available. Variables of comparison were identified a priori and included body mass index (BMI), physical activity (accelerometer-determined steps per day) and screen time (child-report).

RESULTS:

Of 12 countries, data on weight status (BMI) were available in 8 countries, data on measured physical activity (steps per day) were available in 5 countries and data on self-reported screen time were available in 9 countries. The five ISCOLE countries that were part of the Health Behaviour in School-aged Children Survey (that is, Canada, Finland, Portugal, the United Kingdom (England) and the United States) also provided comparable data on self-reported physical activity. Available country-specific data often used different measurement tools or cut-points, making direct comparisons difficult. Where possible, ISCOLE data were re-analyzed to match country-level data, but this step limited between-country comparisons.

CONCLUSIONS:

From the analyses performed, the ISCOLE data do not seem to be systematically biased; however, owing to limitations in data availability, data from ISCOLE should be used with appropriate caution when planning country-level population health interventions. This work highlights the need for harmonized measurement tools around the world while accounting for culturally specific characteristics, and the need for collaboration across study centers and research groups.

Introduction

The prevalence of pediatric obesity and related lifestyle behaviors has been examined in many countries; however, the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE) represents the most up-to-date, directly measured, harmonized study of several countries ranging widely in level of human development.1 Previous multi-national studies have focused on one geographic region (for example, HELENA in Europe (Healthy Lifestyle in Europe by Nutrition in Adolescents)2), or limited their inclusion to developed, or high-income countries (for example, HBSC (Health Behaviour in School-aged Children)3, 4). Further, many previous studies have been limited by small sample sizes, indirect measurements and/or the inability to generalize their findings to larger populations. With so many small, isolated and non-representative studies, it becomes difficult to inform population health interventions, especially on a global scale.

ISCOLE aimed to recruit study sites with diverse geographical distribution drawn from low-, middle- and high-income countries. Participants were recruited in predefined locations within each country, thus the samples were not designed to be nationally representative. However, to aid in interpreting results from ISCOLE, and to make informed recommendations, it is important to understand whether ISCOLE participants are comparable to their country-level peers as a whole, or if they represent a unique sub set of the larger population. The purpose of this paper was to determine if ISCOLE data are representative of their site country. This work may be used to better understand the limitations, potential biases and generalizability of ISCOLE results and to identify current gaps in health- and obesity-related data in the participating countries.

Materials and methods

International study of childhood obesity, lifestyle and the environment

Details on participant recruitment and sampling strategy have been previously reported.1 In brief, data collection occurred from September 2011 through December 2013 with a goal of recruiting at least 500 participants, aged 9–11 years, from each study site. Sites made an effort to stratify their sample by indicators of socio-economic status to maximize variability.1 Many countries included both private and public schools, although all sites limited their data collection to large cities, and urban or suburban schools (that is, no country collected data from participants living in remote or rural areas). To ensure that ISCOLE participants represented diverse backgrounds and circumstances, study sites were chosen from diverse geographic regions around the world (that is, Europe, Africa, the Americas, South Asia and the Western-Pacific region) and across different levels of socio-economic indicators (that is, World Bank classification, Human Development Index and the Gini Index). The ISCOLE coordinating center, located at the Pennington Biomedical Research Center in Baton Rouge, LA, USA was responsible for overall administration of the study. This project was approved by the relevant research ethics boards at Pennington Biomedical Research Center, at each ISCOLE study site, and at the respective school boards. Written informed parental consent and child assent were obtained for all participants as required.

ISCOLE variables used for the present analysis were limited to those common in national surveys and surveillance systems, including body mass index (BMI), household income, physical activity and screen time. Anthropometric variables (height, weight, BMI) were collected and calculated following standard procedures and measurement tools.1 BMI was calculated and weight status was determined using various cut-points (for example, World Health Organization, US Centers for Disease Control and Prevention) to maximize comparability with nationally representative data. See Supplementary File 1 for additional details on income categories. Accelerometer-derived variables (light-, moderate- and vigorous-intensity physical activity and step counts) were obtained via a 24-h wear protocol using the waist-worn ActiGraph GT3X+ triaxial accelerometer (ActiGraph LLC, Pensacola, FL, USA) and validated data-reduction strategies.5, 6, 7 Step counts were calculated using the manufacturer's default filter. Data on behavioral characteristics (for example, self-reported physical activity and screen time) were obtained via a child-report questionnaire. For the purpose of this work, physical activity guidelines were defined as ≥60 min of daily moderate- to vigorous-intensity physical activity and sedentary behavior guidelines were defined as ⩽2 h of self-reported recreational screen time per day. These guidelines are consistent with those from many countries, including Australia,8 Canada,9, 10 the United Kingdom,11 the United States12 and the World Health Organization.13 Socio-economic status was measured via parent-reported household income. Further details on all measurement procedures and questionnaires used in ISCOLE can be found elsewhere.1

Nationally representative data

To gain access to nationally representative data, and to understand the intricacies of the data sets, the primary investigator from each ISCOLE study site was asked to provide information for their respective country. As the primary investigators for ISCOLE were chosen based on their expertise in pediatric obesity research, it was believed that they would be aware of the relevant data sources. They were asked to use their best judgement when identifying information; however, some of the available nationally representative data may be considered to be out of date (that is, >5 years old). The writing group for this paper agreed that it was more important to have comparable and well-collected data (for example, similar age group, directly measured variables) than to necessarily have the most recent data. See Supplementary File 2 for summaries of included studies.

When no data were available, cells were left blank. It is important to note that all variables included in this paper were identified a priori and deemed as the most relevant to the ISCOLE project and most likely to be captured in other studies. Therefore, some countries have a greater number of areas with no data than other countries, and this may highlight a paucity of robust data in certain areas.

Results

ISCOLE data were available for 9–11-year-old children from Australia (n=516), Brazil (n=541), Canada (n=541), China (n=537), Colombia (n=905), Finland (n=525), India (n=584), Kenya (n=563), Portugal (n=724), South Africa (n=513), the United Kingdom (n=525) and the United States (n=554). Characteristics of ISCOLE countries can be found in Table 1, and characteristics of participants from each ISCOLE site can be found in Table 2. The proportion of children overweight or obese, accelerometer-determined steps per day and the proportion of children exceeding screen time guidelines are presented in Figures 1, 2, 3, respectively, with additional information presented in Supplementary Files 3–5. Figure 4 shows the proportion of children that self-reported meeting physical activity guidelines (that is, 60 min of moderate- to vigorous-intensity physical activity per day) in ISCOLE and HBSC.

Table 1. ISCOLE country characteristics.

Country National populationa ISCOLE site location Population of ISCOLE site locationb World bank classificationc Large and/or National study
Europe
 Finland 5 442 322 Helsinki, Espoo, Vantaa 1 060 701 High income HBSC
 United Kingdom 63 705 000 Bath, North-East Somerset 177 700 High income ENERGY, HBSC,d HELENA, IDEFICS, MCS
 Portugal 10 562 178 Porto 237 584 High income EYHS, HBSC
           
Africa
 Kenya 44 354 000 Nairobi 3 138 369 Low income None available
 South Africa 52 981 991 Cape Town 3 497 097 Upper-middle income Time of Use survey
           
The Americas
 Canada 35 158 304 Ottawa 883 391 High income CHMS, HBSC
 United States 316 783 000 Baton Rouge 802 484 High income HBSC, NHANES, YRBS
 Colombia 47 262 816 Bogotá 7 674 366 Upper-middle income ENSIN
 Brazil 201 032 714 Sao Caetano do Sul 149 263 Upper-middle income BIGS
           
South Asia
 India 1 242 456 566 Bangalore 9 588 910 Lower-middle income None available
           
Western-Pacific region
 China 1 362 620 526 Tianjin 10 290 987 Upper-middle income CNNS
 Australia 23 235 207 Adelaide 1 212 982 High income AHS, ANCNPAS
Open in a new tab

Abbreviations: AHS, Australian Health Survey; ANCNPAS, Australian National Children's Nutrition and Physical Activity Survey; BIGS, Brazilian Institute of Geography and Statistics; CHMS, Canadian Health Measures Survey; CNNS, China National Nutrition Survey; ENERGY, EuropeaN Energy balance Research to prevent excessive weight Gain among Youth; ENSIN, Encuesta Nacional de la Situación Nutricional (National Survey of Nutritional Status); EYHS, European Youth Heart Study; HBSC, Health Behavior in School-aged Children; HELENA, Healthy Lifestyle in Europe by Nutrition in Adolescents; IDEFICS, identification and prevention of dietary- and lifestyle-induced health effects in children and infants; ISCOLE, International Study of Childhood Obesity, Lifestyle and the Environment; MCS, Millennium Cohort Study; NHANES, National Health and Nutrition Examination Survey; YRBS, Youth Risk Behavior Survey.

a

Population estimate accessed on October 2014.

b

Represents the population size of the city or general area where children were sampled.

c

World Bank classification represents.

d

As the UK ISCOLE site was in England, the corresponding HBSC data was taken from HBSC England site.

Table 2. ISCOLE participant characteristics.

Study site Participants (n, % boys) Age (years) (mean, s.d.) Weight status (%)a Combined annual household incomeb Approximate equivalent in US dollarsc
Europe
 Finland 536 (47.2%) 10.0 (0.4) Normal weight: 74.6 Overweight/obese: 23.7 <€20 000: 5.5% ⩾€80 000: 40.9% $26 000: 5.5% ⩾$104 000: 40.9%
 United Kingdom 525 (45.1%) 10.4 (0.5) Normal weight: 68.5 Overweight/obese: 30.3 <£10 000: 9.4% £10 000–19 999: 17.3% ⩾£90 000: 8.1% <$16 500: 9.4% $16 500–32 998: 17.3% ⩾$148 500: 8.1%
 Portugal 777 (46.1%) 10.0 (0.3) Normal weight: 52.5 Overweight/obese: 47.2 <€6000: 20.5% €6000–11 999: 30.9% ⩾€42 000: 5.5% $<7800: 20.5% $7800–15 599: 30.9% ⩾$546 000: 5.5%
           
Africa
 Kenya 563 (46.5%) 9.8 (0.7) Normal weight: 75.1 Overweight/obese: 21.1 <Ksh121 980: 23.2% Ksh⩾6 000 000: 3.5% <$1342: 23.2% ⩾$66 000: 3.5%
 South Africa 550 (40.1%) 9.8 (0.7) Normal weight: 71.2 Overweight/obese: 26.4 <R11 500: 47.8% R>500 000: 7.6% <$1081: 47.8% ⩾$47 000: 7.6%
           
The Americas
 Canada 565 (42.2%) 10.0 (0.4) Normal weight: 68.7 Overweight/obese: 30.8 <$14 999: 2.9% ⩾$140 000: 38.4% <$13 799: 2.9% ⩾$128 800: 38.4%
 United States 651 (43.2%) 9.5 (0.6) Normal weight: 58.4 Overweight/obese: 41.3 <$10 000: 20.4% ⩾$140 000: 21.6% NA
 Colombia 919 (49.4%) 10.0 (0.6) Normal weight: 75.7 Overweight/obese: 22.9 $0–1200 000: 0.7% $4 800 000–8 400 000: 29.4% ⩾$36 000 000: 8.8% $0–624: 0.7% $2496–4368: 29.4% ⩾$18 720: 8.8%
 Brazil 584 (49.1%) 10.1 (0.5) Normal weight: 52.8 Overweight/obese: 45.2 <R$6540: 3.1% R$6540–19 620: 35.1% ⩾R$85 020: 4.4% <$2943: 3.1% $2943–8829: 35.1% ⩾$38 259: 4.4%
           
South Asia
 India 620 (47.1%) 10.0 (0.6) Normal weight: 61.5 Overweight/obese: 33.7 <Rs60 000: 2.8% ⩾Rs720 000: 37.6% <$1020: 2.8% ⩾$12 240: 37.6%
           
Western-Pacific region
 China 552 (53.1%) 9.4 (0.5) Normal weight: 56.3 Overweight/obese: 41.2 <¥20 000: 18.1% ¥20 000–39 999: 18.1% ⩾¥150 000: 10.4% <$3200: 18.1% $3200–6400: 18.1% ⩾$24 000: 10.4%
 Australia 528 (46.0%) 10.3 (0.5) Normal weight: 61.4 Overweight/obese: 37.9 <$10 000: 2.1% $70 000–89 999: 17.0% ⩾$140 000: 21.7% <$9300: 2.1% $65 100–83 699: 17.0% ⩾$130 200: 21.7%
Open in a new tab

Abbreviation: ISCOLE, International Study of Childhood Obesity, Lifestyle and the Environment.

a

Weight status defined by World Health Organization cut-points.44

b

Presented as percent of participants in the lowest, highest and median income categories. If the median income category was also the lowest, or the highest, only two income categories are presented.

c

Currency conversion reflects rates from 2014.

Figure 1.

Figure 1

Open in a new tab

Proportion of children considered overweight or obese from ISCOLE study sites and their representative countries. Dark gray bars indicate data from ISCOLE participants; white bars represent country-level data. If no white bar, then country-level data are not available. Where available, data are presented for both boys and girls. Data were analyzed as per BMI cut-points available in each country as follows. World Health Organization: Portugal, Colombia, Brazil; International Obesity Task Force: Canada, Australia; Center for Disease Control and Prevention: the United States; other: the United Kingdom (British 1990 growth reference17), China (China BMI criteria, overweight ⩾19.4, obese ⩾22.2).44 Country-level data sets included: the United Kingdom: National Child Measurement Programme;17 Portugal: Plataforma Contra a Obesidade;45 Canada: Canadian Health Measures Survey;46 the United States, National Health and Nutrition Examination Survey; Colombia: Encuesta Nacional de la Situación Nutricional; Brazil: Brazilian Institute of Geography and Statistics;25 China: China Health and Nutrition Survey;44 Australia: Australian National Children's Nutrition and Physical Activity Survey.35 See Additional file 3 for additional study details for country-level data.

Figure 2.

Figure 2

Open in a new tab

Daily physical activity (steps per day) from ISCOLE study sites and their representative countries. Dark gray bars indicate data from ISCOLE participants; white bars represent country-level data. If no white bar, then country-level data were not available. Where available, data are presented for both boys and girls. Horizontal black line represents mean steps per day for all ISCOLE participants; horizontal dashed line represents target of 12 000 steps per day recommended to meet the current physical activity guidelines.47 Data were included if it was collected via pedometer or accelerometer and presented as sample mean. Country-level data sets included: Finland: Physical Activity of School-Aged Children;14 UK: Millennium Cohort Study;16 Canada: Canadian Health Measures Survey;28 the United States: National Health and Nutrition Examination Survey;48 Australia: Australian National Children's Nutrition and Physical Activity Survey.35 The Millennium Cohort Study also provided data for England: 10 147 steps per day compared with 9982 steps per day in ISCOLE. See Additional file 4 for additional study details on country-level data.

Figure 3.

Figure 3

Open in a new tab

Proportion of children exceeding the screen time guidelines (>2 h per day) from ISCOLE sites and their representative countries. Dark gray bars indicate data from ISCOLE participants; white bars represent country-level data. If no white bar, then country-level data were not available. Where available, data are presented for both boys and girls. Horizontal black like represents mean proportion of all ISCOLE participants who exceed the screen time guidelines. Country-level data sets included: Canada, Finland, the United Kingdom and the United States: Healthy Behaviors in School-aged Children;3 South Africa: Time of Use survey;23 Colombia: Instituto Colombiano de Bienestar Familiar (ICBF); China: China Health and Nutrition Survey;49 Australia: Australian Bureau of Statistics.36 See Additional file 5 for additional study details on country-level data.

Figure 4.

Figure 4

Open in a new tab

Proportion of girls (a) and boys (b) who self-reported that they engage in at least 60 min of moderate- to vigorous-intensity physical activity every day of the week. Dark gray bars indicate data from ISCOLE participants; white bars represent data that were adapted from the HBSC survey.4 MVPA, moderate- to vigorous-intensity physical activity.

Europe (Finland, Portugal and the United Kingdom)

All of the ISCOLE site countries in Europe were classified as high-income countries. They were also all part of the HBSC survey.4 On the basis of the results from HBSC, Finland provided self-reported data showing that 38% of 11-year-old boys, and 25% of 11-year-old girls met the physical activity guidelines,4 which is higher than 22% of 9–11-year-old boys and 10% of 9–11-year-old girls who self-reported that they met physical activity guidelines in ISCOLE. Additional data from Tammelin et al.,14 and the Foundation for Sport and Health Science, found that Finnish children (aged 9–10 years) accumulated 10 506 steps per day, compared with 10 485 steps per day from the ISCOLE participants. Data on screen time came from the 2010 HBSC, which reported that 61% of 11-year-old boys and 58% of 11-year-old girls self-reported that they exceeded screen time recommendations on weekdays. This was lower than in ISCOLE, which showed that 84% of boys and 74% of girls exceeded the recommendation. We were unable to find any comparable data reporting the proportion of Finnish children considered overweight or obese. Therefore, data from ISCOLE Finland are consistent with previous data for directly measured physical activity (but lower for self-reported physical activity), provide slightly higher estimates for screen time and are inconclusive with respect to weight status.

Portuguese children who were part of the HBSC study self-reported higher physical activity levels (23% of 11-year-old boys and 14% of 11-year old girls are meeting physical activity guidelines) than self-report data from ISCOLE (9% of boys and 4% of girls).4 Reports from HBSC suggest 11-year-old children are watching more television (61% of boys and 60% of girls exceed the screen time guidelines) compared with the ISCOLE participants (55.3% of boys and 43.0% of girls exceed the screen time guidelines). Regarding weight status, Portugal was able to provide comparable data for 10-year-old children from the Plataforma Contra a Obesidade.15 This showed similar proportions of children considered overweight or obese for boys (50.0% compared with 51.4% in ISCOLE) and girls (45.3 versus 42.3% in ISCOLE). Overall, participants from ISCOLE Portugal reported lower levels of physical activity, but similar levels of screen time and a similar proportion of children considered overweight or obese compared with other surveys.

Because the ISCOLE site in the United Kingdom was based in England, we aimed to obtain comparable data from England, rather than across the United Kingdom, wherever possible. Data from England came primarily from the Millennium Cohort Study (MCS), with additional data coming from HBSC. According to the MCS (children aged 7–8 years), boys averaged 10 739 steps per day and girls averaged 9699 steps per day.16 This was very similar to what was seen in ISCOLE with boys averaging 10 675 steps per day and girls averaging 9435 steps per day. Self-reported data from HBSC showed that 33% of 11-year-old boys and 20% of 11-year-old girls met the physical activity guidelines,4 whereas in ISCOLE, only 19% of boys and 11% of girls self-reported that they met the guidelines. With respect to sedentary behavior, results from the HBSC showed that 64% of 11-year-old boys and 60% of 11-year-old girls exceeded the screen time guidelines on weekdays;4 results from ISCOLE show that 73% of boys and 61% of girls exceeded the guidelines. Compared with data from 10–11-year-old children from the National Child Measurement Programme, there was a lower proportion of children considered overweight or obese in ISCOLE (33.3 versus 21.7%, respectively).17 Data from ISCOLE seem to be comparable with respect to physical activity levels, but with higher levels of screen time. However, it can be assumed that there are variations in the socio-cultural environment across the rest of the United Kingdom and it is recommended that generalizations to a wider population are made with caution.

Africa (Kenya, South Africa)

Kenya was the only study included in ISCOLE considered to be a low-income country; South Africa is considered to be an upper-middle-income country. Neither Kenya nor South Africa had any nationally representative or large data sets for comparison. Summaries of the best available evidence for each country was presented in their 2014 report card on physical activity in children and youth and included primarily early results from ISCOLE, suggesting that future studies are critically needed.18, 19

Comparisons for ISCOLE Kenya came from two systematic reviews examining obesity and physical activity transitions in sub-Saharan Africa.20, 21 One systematic review reported a range of 35–72% of children meeting the physical activity guidelines,20 which is higher than the 11.4% of children who self-reported meeting the guidelines in ISCOLE. A small study examining urban and rural Kenyan children (aged 9–12 years) found that only 13.1% of urban and no rural children used screens for >11 h per week.22 This is much lower than ISCOLE measurements which showed that 53.6% of children exceeded the screen time guidelines. This is consistent with the emerging evidence of a physical activity transition in Kenya.22 The systematic review examining the overweight and obesity transition included studies that reported proportions of overweight/obesity between 3.2–12.0%,21 which is lower than 16% found in ISCOLE. However, differences in weight status may be due, in part, to urban and rural differences. Even though there were little comparable data available in Kenya, Nairobi is an urban hub, with a population of over 3 million people, and one could assume that children from urban Nairobi may not be representative of their rural peers.

Although South Africa was not able to provide data from any nationally representative surveys, information from smaller studies suggest that 50–59% of children are meeting physical activity guidelines, and <50% of children are meeting screen time guidelines.19 These values are more promising than what was shown in ISCOLE, which found that 26.4% of children self-reported meeting physical activity guidelines, and 36.7% reported meeting screen time guidelines. Data from the 2010 Survey of Time of Use suggest that on average, children aged 10–17 years watched 3 h of television per day.23 In ISCOLE, children were asked to self-report habitual television, video game and computer use with the highest possible value being ‘≥5 h per day'. Therefore, for analysis, self-reported screen time was presented as a score, rather than total hours of screen time, as after 5 h per day we could not ascertain the participant's actual amount of screen time. Although ISCOLE can only provide an approximate value for daily screen time, it did appear to be similar to what was reported in the Time of Use survey with children averaging ~3.1 h of screen time and 2.0 h of television time per day. We were not able to find any comparable data with respect to weight status. Therefore, children from ISCOLE South Africa seem to be slightly less active than reported in other studies, but seem to engage in similar amounts of screen time.

The Americas (Brazil, Canada, Colombia and the United States)

Canada and the United States are both considered as high-income countries; whereas Colombia and Brazil are both considered as upper-middle-income countries. Canada and the United States were able to provide a significant amount of comparable and nationally representative data. Both countries are included in the HBSC and both have nationally representative surveys (CHMS: Canadian Health Measures Survey24 and NHANES: National Health and Nutrition Examination Survey (USA)).

Brazil provided comparable data for weight status for 10-year-old children from the Brazilian Institute of Geography and Statistics (BIGS),25 but was unable to provide comparable data for physical activity or screen time. BIGS reported that 33.1% of their children were considered overweight or obese, compared with a much higher 45.7% of children considered overweight or obese in ISCOLE.

Comparable data for Canada came primarily from the CHMS (with custom analysis for their 2014 Report Card26) and showed that only 5% of school-aged children met current Canadian physical activity guidelines.9, 27 However, directly measured physical activity from 6–10-year-old children in the CHMS show that ISCOLE children are less active than children from across Canada. In the CHMS boys averaged 13 217 steps per day and girls averaged 11 745 steps per day compared with 9891 steps per day for boys and 8591 steps per day for girls.28 Self-reported data from HBSC showed 31% of boys and 21% of girls met the physical activity guidelines,4 which was similar to data from ISCOLE, with 26% of boys and 18% of girls self-reporting that they met guidelines. With respect to the screen time, data came from the 2007–2009 CHMS and showed that 31% of children aged 5–11 years exceed screen time guidelines.28 This was slightly lower than what was seen in ISCOLE Canada, with 54.4% of boys and 40.2% of girls exceeding guidelines. There was a slightly lower proportion of children considered overweight or obese in ISCOLE (21.9% of boys and 21.7% of girls), compared with reports from the CHMS (31% for boys and 26% for girls). In Canada, ISCOLE participants engaged in similar amounts of self-reported physical activity (but less directly measured physical activity), greater amounts of screen time and had a lower proportion of children considered overweight or obese.

Comparable data for levels of physical activity in Colombia came from the ENSIN study (Encuesta Nacional de la Situación Nutricional (National Survey of Nutritional Status)),29 and showed that 26% of children self-reported that they met physical activity guidelines, which is higher than what was seen in ISCOLE (12% meeting guidelines). The ENSIN study also provided comparable data with respect to screen time and weight status. For screen time, ENSIN showed that fewer children (59.4% of boys and 56.3% of girls) exceeded screen time guidelines than children from ISCOLE (71.6% of boys and 60.4% of girls).30 ENSIN also reported a lower proportion of children being overweight or obese than ISCOLE (17.5% in ENSIN compared with 23.3% in ISCOLE).30 Overall, it appears that children in ISCOLE self-report engaging in less physical activity and more screen time, and are more likely to be overweight or obese than other Colombian children.

Comparable data for US physical activity levels came primarily from the 2003–2004 NHANES. Accelerometer data showed that 42.0% of children aged 6–11 years were meeting physical activity guidelines,31, 32 and on average, 10-year-old boys were accumulating 10 163 steps per day and girls were accumulating 8906 steps per day.33 This was similar to what was seen in ISCOLE, with boys accumulating only 9261 steps per day, and girls accumulating only 8078 steps per day. With respect to screen time, data from 2009–2010 NHANES showed that 47.8% of children aged 9–11 years self-reported that they exceeded screen time guidelines.34 The proportion of children exceeding guidelines was higher in ISCOLE and in 11-year-old boys and girls from HBSC4 (Figure 3). In ISCOLE, 68.3% of boys and 56.2% of girls exceeded the screen time guidelines. Compared with NHANES, we saw a similar proportion of children considered overweight or obese in ISCOLE (NHANES: boys=33.2%, girls=35.2% ISCOLE boys=32.4%, girls=35.6%). In the US sample of ISCOLE, children were less active, and were less likely to meet the screen time guidelines, but had similar weight status to children from across the country.

South Asia (India)

India is the only study site in South Asia, and is considered a lower-middle-income country. We were unable to identify any comparable data sets. Data collected in ISCOLE show that 28.4% of Indian children are meeting physical activity guidelines, 30.8% are meeting the screen time guidelines and 33.7% are considered overweight or obese.

Western-Pacific region (Australia, China)

Australia is considered a high-income country and China is considered an upper-middle-income country. Australia provided comparable data via the ANCNPAS (Australian National Children's Nutrition and Physical Activity Survey),35 the National Health Surveys36 and a summary of current evidence via their 2014 report card on physical activity in children and youth.37 China was able to provide comparable data via the China Health and Nutrition Survey (CHNS).

Comparable data from Australia came primarily from the Australian Bureau of Statistics. These data showed that 20% of children aged 5–17 years self-reported that they met the physical activity guidelines.36 This is consistent with self-reported data from ISCOLE. Australia was also able to provide information from the ANCNPAS showing higher values for number of steps per day (12 230 in ANCNPAS compared with 10 262 in ISCOLE).35 With respect to screen time, data from the Australian Bureau of Statistics showed that 70% of children aged 5–17 years exceeded the screen time guidelines. This was higher than what was seen in ISCOLE, with only 56% of children exceeding the guidelines. Data from ANCNPAS reported 28.4% of 9–11-year-old children to be overweight or obese, which is similar to 26.7% of children considered overweight or obese in ISCOLE. Overall, participants from ISCOLE accumulated similar amounts of physical activity, had lower screen time and had more favorable weight status than their Australian peers.

China provided comparable data on physical activity, screen time and weight status via the CHNS (China Health and Nutrition Survey).38, 39 For 6–11-year olds, the CHNS reported children averaged 60 min (boys) and 48 min (girls) of physical activity per day, compared with lower levels in ISCOLE (49.5 min for boys and 40.5 min for girls). Results for screen time showed that 24.9% of boys and girls exceeded the screen time guidelines, compared with 35.1% in ISCOLE. However, it is interesting to note that these data came from 2004, and in a longitudinal analysis, the proportion of children who reported >2 h of screen time per day rose from 5.8% in 1997 to 24.9% in 2004, suggesting data from present day may be more comparable. The proportion of children considered overweight or obese was much higher in ISCOLE than in the CHNS (ISCOLE boys=50.1%, ISCOLE girls=30.6% versus CHNS boys=16.6% and HBSC boys=10.9%). ISCOLE China participants accumulated slightly less physical activity and slightly more screen time than other estimates; however, the difference (in screen time at least), may be because ISCOLE data collection has occurred more recently, and may not reflect a true difference. The proportion of ISCOLE participants considered overweight or obese was much higher than previous estimates.

Discussion

This study aimed to compare data collected in ISCOLE to data collected via nationally representative studies in all ISCOLE study sites. Of the 12 ISCOLE countries, 8 provided data on weight status (BMI), 5 provided data on objectively measured physical activity (steps per day) and 9 provided data on self-reported screen time. The five ISCOLE countries that were part of the HBSC survey provided additional data for self-reported physical activity (meeting physical activity guidelines). When data were available, mean data from the ISCOLE study appears to be relatively similar to country-level data; however, this varied with data availability and quality, and there were no systematic differences across countries or variables. Few countries used the same cut-points or measurement tools in their national studies as ISCOLE when analyzing participant characteristics.

Initially, the aim of this study was to compare data from ISCOLE sites to nationally representative data in the site country, to get a crude evaluation of the potential bias in the ISCOLE sample. However, it quickly became apparent that many countries do not collect nationally representative data on physical activity, screen time, or weight status. When data were available, it was collected using different tools or with different methods. For example, we aimed to compare accelerometer-measured daily moderate- to vigorous-intensity physical activity (minutes per day); however, after scoping the literature for comparable country-level data, this was not possible. Because of the range of different models of accelerometers, and the different cut-points used to distinguish activity intensity, only two countries (the United States and Canada) were able to provide comparable data. No countries that could provide nationally representative accelerometer-measured physical activity data using common cut-points. Therefore, we opted to examine physical activity via steps per day, which can be measured using an accelerometer, or a pedometer. The added benefit of examining steps per day is that the cost of pedometers is quite low, making it more feasible to use them in large, population-based studies or in times when resources are limited.

This work has several strengths and limitations. Many countries could not provide any comparable data and therefore it was difficult for us to determine if ISCOLE participants were similar to other children. This meant that many comparisons could not be made, and in most cases we did not have access to the raw data from nationally representative surveys and could not complete any formal statistical analysis. This leaves our comparisons open to interpretation, and our judgement on similarities (or differences) in the data open to criticism. However, we believe this also highlights an important limitation of making global comparisons from many unconnected (but nationally representative) studies and the error of collapsing data that has used inconsistencies in measurement tools, different cut-points, across different research centers and countries, and the unavailability of raw data. Most of the studies included in this work used different methodologies, and different cut-points to assess similar health variables. For example, BMI is a common measure of adiposity in children, yet there are four different cut-points used, all of which provide different prevalence of overweight/obesity.40 ISCOLE recruitment was also limited to urban and suburban schools, and therefore we cannot account for populations living in rural areas. Previous work has consistently shown differences in obesity and lifestyle habits between these two groups, and that these differences may be most important in developing areas.41, 42, 43 The results of this work provide a valuable summary when developing future research programs and can help inform public health interventions. Another strength of this work is related to the rigor of the ISCOLE data collection and management procedures.1 The ISCOLE framework and coordinating center ensured all study sites, and all ISCOLE researchers, completed mandatory training for all aspects of the study.

Conclusions

This manuscript was designed as a methodological and ecological comparison study that may be used to provide evidence of the potential bias from each ISCOLE country sample, facilitating future intra- and inter-country comparisons. Because of the limited availability of country-level data, it is suggested that ISCOLE data be used with appropriate caution when planning countrywide population health interventions. However, for many countries ISCOLE currently provides the most up-to-date, most robust and sometimes the only data on obesity and physical activity in children. This work has identified the paucity of comparable country data around the world and highlighted the importance of large, multi-national studies like ISCOLE. Moving forward, we recommend that researchers harmonize procedures for data collection and analysis. It is important to use the momentum, and collaborations that were built in ISCOLE to inform public health interventions, as well as other cross-sectional, surveillance surveys.

Acknowledgments

We thank the ISCOLE External Advisory Board and the ISCOLE participants and their families who made this study possible. A membership list of the ISCOLE Research Group and External Advisory Board is included in Katzmarzyk et al. (this issue). ISCOLE was funded by The Coca-Cola Company. MF has received a research grant from Fazer Finland. RK has received a research grant from Abbott Nutrition Research and Development.

MF has received an honorarium for speaking for Merck. AK has been a member of the Advisory Boards of Dupont and McCain Foods. VM is a member of the Scientific Advisory Board of Actigraph and has received an honorarium for speaking for The Coca-Cola Company. TO has received an honorarium for speaking for The Coca-Cola Company. The remaining authors declare no conflict of interest.

Footnotes

Supplementary Information accompanies this paper on International Journal of Obesity Supplements website (http://www.nature.com/ijosup)

Supplementary Material

Supplementary Information
Click here for additional data file. (26.5KB, doc)
Supplementary Table 1
Click here for additional data file. (17.6KB, docx)
Supplementary Table 2
Click here for additional data file. (23.3KB, docx)
Supplementary Table 3
Click here for additional data file. (25.2KB, docx)
Supplementary Table 4
Click here for additional data file. (27.2KB, docx)
Supplementary Table 5
Click here for additional data file. (25.8KB, docx)

References

  1. Katzmarzyk PT, Barreira TV, Broyles ST, Champagne CM, Chaput J-P, Fogelholm M et al. The International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE): design and methods. BMC Public Health 2013; 13: 900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Healthy Lifestyle in Europe by Nutrition in Adolescence. Available at http://www.helenastudy.com/ (accessed on 11 February 2015).
  3. Currie C, Gabhainn SN, Godeau E, Roberts RS, Currie O, Picket W et al. Inequalities in Young People's Health: Health Behaviour in School-aged Children. International Report from the 2005–2006 Survey, Health Policy for Children and Adolescents, HBSC International Report 2008.
  4. Currie C, Zanottie C, Morgan A, Currie D, de Looze M, Roberts C et al. Social determinants of health and well-being among young people. WHO Regional Office for Europe. World Health Organization: Copenhagen, 2012. [Google Scholar]
  5. Barreira TV, Schuna JM, Mire EF, Katzmarzyk PT, Chaput J-P, Leduc G et al. Identifying children's nocturnal sleep using 24-h waist accelerometry. Med Sci Sports Exerc 2015; 47: 937–943. [DOI] [PubMed] [Google Scholar]
  6. Trost SG, Loprinzi PD, Moore R, Pfeiffer KA. Comparison of Accelerometer cut points for predicting activity intensity in youth. Med Sci Sports Exerc 2011; 43: 1360–1368. [DOI] [PubMed] [Google Scholar]
  7. Colley RC, Connor Gorber S, Tremblay MS. Quality control and data reduction procedures for accelerometry-derived measures of physical activity. Health Rep 2010; 21: 63–69. [PubMed] [Google Scholar]
  8. Okely AD, Salmon J, Trost SG, Hinkley T. Discussion Paper for the Development of Physical Activity Recommendations for Children Under 5 Years. Department of Health and Ageing, Australian Government: Australia, 2008. [Google Scholar]
  9. Tremblay MS, Warburton DER, Janssen I, Paterson DH, Latimer AE, Rhodes RE et al. New Canadian Physical Activity Guidelines. Appl Physiol Nutr Metab 2011; 36: 36–46. [DOI] [PubMed] [Google Scholar]
  10. Tremblay MS, LeBlanc AG, Janssen I, Kho ME, Hicks A, Murumets K et al. Canadian Sedentary Behaviour Guidelines for Children and Youth. Appl Physiol Nutr Metab 2011; 36: 59–64. [DOI] [PubMed] [Google Scholar]
  11. Bull FC. Expert Working Group. Physical activity guidelines in the UK: review and recommendations. School of Sport, Exercise, and Health Sciences, Loughborough University: Leichestershire, UK, 2010. [Google Scholar]
  12. Pate R, Pratt M. Physical activity and public health: A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA 1995; 273: 402–407. [DOI] [PubMed] [Google Scholar]
  13. World Health Organization. Global recommendations on physical activity for health 2010. Available at http://www.who.int/dietphysicalactivity/publications/9789241599979/en/index.html (accessed on 28 February 2013). [PubMed]
  14. Tammelin T, Laine K, Turpeinen S. Oppilaiden fyysinen aktiivisuus. (Physical Activity of School-aged children). Liikunnan ja kansanterveyden edistämissäätiö LIKES (Sport and Health LIKES): Jyväskylä, 2013. [Google Scholar]
  15. Plataforma Contra a Obesidade. Available at http://www.plataformacontraaobesidade.dgs.pt/PresentationLayer/homepage_institucional.aspx?menuid=113 (accessed on 19 February 2015).
  16. Griffiths LJ, Cortina-Borja M, Sera F, Pouliou T, Geraci M, Rich C et al. How active are our children? Findings from the Millennium Cohort Study. BMJ Open 2013; 3: e002893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lifestyle Statistics TeamNational Child Measurement Programme: England, 2012/2013 school year. Health and Social Care Information Centre, Public Health England, 2013. [Google Scholar]
  18. Wachira LJM, Muthuri SK, Tremblay MS, Onywera VO. Results from Kenya's 2014 report card on the physical activity and body weight of children and youth. J Phys Act Health 2014; 11: S69–S73. [DOI] [PubMed] [Google Scholar]
  19. Draper C, Basset S, de Villiers A, Lambert EV. Writing Group HAKSA. Results from South Africa's 2014 report card on physical activity for children and youth. J Phys Act Health 2014; 11: S98–S104. [DOI] [PubMed] [Google Scholar]
  20. Muthuri SK, Wachira L-JM, Leblanc AG, Francis CE, Sampson M, Onywera VO et al. Temporal trends and correlates of physical activity, sedentary behaviour, and physical fitness among school-aged children in Sub-Saharan Africa: a systematic review. Int J Environ Res Public Health 2014; 11: 3327–3359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Muthuri SK, Francis CE, Wachira L-JM, LeBlanc AG, Sampson M, Onywera VO et al. Evidence of an overweight/obesity transition among school-aged children and youth in Sub-Saharan Africa: a systematic review. PLoS One 2014; 9: e92846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Onywera VO, Adamo KB, Sheel AW, Waudo JN, Boit MK, Tremblay M. Emerging evidence of the physical activity transition in Kenya. J Phys Act Health 2012; 9: 554–562. [DOI] [PubMed] [Google Scholar]
  23. Statistics South AfricaA Time of Use. Statistics South Africa: Pretoria, South Africa, 2013. [Google Scholar]
  24. Giroux S. Canadian Health Measures Survey: sampling strategy overview. Health Rep 2006; 18: 31–36. [PubMed] [Google Scholar]
  25. Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística). Available at http://www.ibge.gov.br/english/ (accessed on 18 February 2015).
  26. Healthy Active Living and Obesity Research Group, Gray CE, Barnes JD, Cowie Bonne J, Cameron C, Chaput JP et al. Results from Canada's 2014 report card on physical activity for children and youth. J Phys Act Health 2014; 11: S26–S32. [DOI] [PubMed] [Google Scholar]
  27. Active Healthy Kids Canada. Are we driving our kids to unhealthy habits? The 2013 Active Healthy Kids Canada Report Card on Physical Activity for children and youth 2013. [DOI] [PMC free article] [PubMed]
  28. Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS. Physical activity of Canadian children and youth: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep 2011; 22: 15–23. [PubMed] [Google Scholar]
  29. Instituto Colombiano de Bienestar Familiar ICBFEncusesta Nacional de la Situacion Nutricional en Colombia ENSIN. Insituto Colombiano de Bienestar Familiar, Ed: Bogota, 2005. [Google Scholar]
  30. Gomez LF, Parra DC, Lobelo F, Samper B, Moreno J, Jacoby E et al. Television viewing and its association with overweight in Colombian children: results from the 2005 National Nutrition Survey: A cross sectional study. Int J Behav Nutr Phys Act 2007; 4: 41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Troiano RP, Berrigan D, Dodd KW, Mâsse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008; 40: 181–188. [DOI] [PubMed] [Google Scholar]
  32. Dentro KN, Beals K, Crouter SE, Eisenmann JC, McKenzie TL, Pate RR et al. Results from the United States' 2014 report card on physical activity for children and youth. J Phys Act Health 2014; 11: S105–S112. [DOI] [PubMed] [Google Scholar]
  33. Barreira TV, Schuna JM, Mire EF, Broyles ST, Katzmarzyk PT, Johnson WD et al. Normative steps/day and peak cadence values for United States children and adolescents: National Health and Nutrition Examination Survey 2005–2006. J Pediatr 2015; 166: 139–143 e3. [DOI] [PubMed] [Google Scholar]
  34. Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents. Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents: summary report. Pediatrics 2011; 128: S213–S256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Department of Health and AgeingAustralian National Children's Nutrition and Physical Activity Survey (ANCNPAS): main findings. Commonwealth Scientific and Industrial Research Organisation (CSIRO) and the University of South Australia: Canberra, Australia, 2008. [Google Scholar]
  36. Australian Bureau of Statistics. Australian Health Survey: Physical Activity, 2011–12. 2013. Available at http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/4364.0.55.001main+features12011-12 (accessed on 10 October 2014).
  37. University of South Australia, Schranz N, Olds T, Cliff D, Davern M, Engelen L et al. Results from Australia's 2014 report card on physical activity for children and youth. J Phys Act Health 2014; 11: S21–S25. [DOI] [PubMed] [Google Scholar]
  38. Zhang J, Seo D-C, Kolbe L, Middlestadt S, Zhao W. Associated trends in sedentary behavior and BMI among chinese school children and adolescents in seven diverse Chinese provinces. Int J Behav Med 2011; 19: 342–350. [DOI] [PubMed] [Google Scholar]
  39. Yao N. Weight status in Chinese children: maternal perceptions and child self-assessments. World J Pediatr 2012; 8: 129–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shields M, Tremblay MS. Canadian childhood obesity estimates based on WHO, IOTF and CDC cut-points. Int J Pediatr Obes 2010; 5: 265–273. [DOI] [PubMed] [Google Scholar]
  41. Joens-Matre RR, Welk GJ, Calabro MA, Russell DW, Nicklay E, Hensley LD. Rural-urban differences in physical activity, physical fitness, and overweight prevalence of children. J Rural Health 2008; 24: 49–54. [DOI] [PubMed] [Google Scholar]
  42. Plotnikoff RC, Mayhew A, Birkett N, Loucaides CA, G Fodor. Age, gender, and urban–rural differences in the correlates of physical activity. Prev Med 2004; 39: 1115–1125. [DOI] [PubMed] [Google Scholar]
  43. Onywera VO Adiposity and physical activity among children in countries at different stages of the physical activity transition: Canada, Mexico and Kenya 2013. Available at http://reference.sabinet.co.za/sa_epublication_article/ajpherd_v19_n1_a12 (accessed on 25 April 2015).
  44. He W, Li Q, Yang M, Jiao J, Ma X, Zhou Y et al. Lower BMI cutoffs to define overweight and obesity in China. Obesity 2015; 23: 684–691. [DOI] [PubMed] [Google Scholar]
  45. Sardinha LB, Santos R, Vale S, Silva AM, Ferreira JP, Raimundo AM et al. Prevalence of overweight and obesity among Portuguese youth: a study in a representative sample of 10-18-year-old children and adolescents. Int J Pediatr Obes 2011; 6: e124–e128. [DOI] [PubMed] [Google Scholar]
  46. Tremblay M, Shields M, Laviolette M, Craig C, Janssen I, Connor Gorber S. Fitness of Canadian children and youth: results from the 2007-2009 Canadian Health Measures Survey. Health Rep 2010; 21: 7–20. [PubMed] [Google Scholar]
  47. Colley RC, Janssen I, Tremblay MS. Daily step target to measure adherence to physical activity guidelines in children. Med Sci Sports Exerc 44: 977–982. [DOI] [PubMed] [Google Scholar]
  48. Tudor-Locke C, Johnson WD, Katzmarzyk PT. Accelerometer-determined steps per day in US children and youth. Med Sci Sports Exerc 2010; 42: 2244–2250. [DOI] [PubMed] [Google Scholar]
  49. Cao H, Qian Q, Weng T, Yuan C, Sun Y, Wang H et al. Screen time, physical activity and mental health among urban adolescents in China. Prev Med 2011; 53: 316–320. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
Click here for additional data file. (26.5KB, doc)
Supplementary Table 1
Click here for additional data file. (17.6KB, docx)
Supplementary Table 2
Click here for additional data file. (23.3KB, docx)
Supplementary Table 3
Click here for additional data file. (25.2KB, docx)
Supplementary Table 4
Click here for additional data file. (27.2KB, docx)
Supplementary Table 5
Click here for additional data file. (25.8KB, docx)

Articles from International Journal of Obesity Supplements are provided here courtesy of Nature Publishing Group

RESOURCES