Leandro Araujo, and Victor Matsudo. CELAFISCS. Tiago V. Barreira, Catrine Tudor-Locke, and Peter Katzmarzyk. Pennington Biomedical Research Center.
Pediatric Exercise Science, 2015, 27, 380 -389 http://dx.doi.org/10.1123/pes.2014-0150 © 2015 Human Kinetics, Inc.
ORIGINAL RESEARCH
Moderate-to-Vigorous Physical Activity and Sedentary Behavior: Independent Associations With Body Composition Variables in Brazilian Children Gerson Luis de Moraes Ferrari, Luis Carlos Oliveira, Timoteo Leandro Araujo, and Victor Matsudo CELAFISCS
Tiago V. Barreira, Catrine Tudor-Locke, and Peter Katzmarzyk Pennington Biomedical Research Center This study aimed to analyze the independent associations of accelerometer-determined sedentary behavior, physical activity, and steps/day with body composition variables in Brazilian children. 485 children wore accelerometers for 7 days. Variables included time in sedentary behavior and different physical activity intensities (light, moderate, vigorous, or moderate-to-vigorous) and steps/day. Body fat percentage was measured using a bioelectrical impedance scale, and BMI was calculated. Children spent 55.7% of the awake portion of the day in sedentary behavior, 37.6% in light physical activity, 4.6% in moderate physical activity, and 1.9% in vigorous physical activity. Moderate-to-vigorous physical activity and steps/day were negatively associated with body composition (BMI and body fat percentage) variables, independent of sex and sedentary behavior. Beta values were higher for vigorous physical activity than moderate physical activity. Vigorous physical activity was negatively associated with BMI (β-.1425) and body fat percentage (β-.3082; p < .0001). In boys, there were significant negative associations between moderate, vigorous, and moderate-to-vigorous physical activity and steps/day with body composition, and in girls, there was only a negative association with vigorous physical activity, independent of sedentary behavior. Moderate-to-vigorous physical activity and steps/day (in boys), but especially vigorous physical activity (in boys and girls), are associated with body composition, independent of sedentary behavior. Sedentary behavior was not related with any of the body composition variables once adjusted for moderate-to-vigorous physical activity. Keywords: children, physical activity, accelerometer, sedentary behavior, anthropometric, bioelectric impedance The increasing prevalence of overweight and obesity in children around the world is a recognized critical public health concern that has spurred regional and local governments to consider different strategies to reduce excess weight in the population (20,24). Recent systematic reviews concluded that high levels of regular physical activity likely protect against obesity in children and adolescents, and high levels of sedentary behavior likely promote obesity (15,18). Although public health guidelines recommend that children and adolescents accumulate at least 60 min of Ferrari, Oliveria, Araujo, and Matsudo are with the CELAFISCS, Sao Caetano do Sul, Brazil. Barreira, Tudor-Locke, and Katzmarzyk are with the Pennington Biomedical Research Center, Baton Rouge, LA. Address author correspondence to Gerson Luis de Moraes Ferrari at gersonferrari08@yahoo. com.br.
380
moderate-to-vigorous physical activity daily (26,35). Recent evidence suggests that the majority of children fail to meet this target (29). The 2009 Brazilian National Adolescent Schoolbased Health Survey (Pesquisa Nacional de Saúde do Escolar) reported that one in three (33.5%) children was overweight, and 16.6% of boys and 11.8% of girls were obese (27). Further, only 27.9% of Brazilian boys and 13.1% of girls self-report achieving moderate-to-vigorous physical activity recommendations (29). Detailed and objective measures of time spent in sedentary behavior, different physical activity intensities (including light, moderate, vigorous, and moderate-to-vigorous physical activity), and steps/day (as an indicator of overall physical activity volume) would help identify which aspects of human movement behaviors are most related to overweight and obesity in children. The objective assessment of children’s sedentary behavior, physical activity intensities, and steps/day using
Physical Activity and Anthropometric Variables 381
accelerometers has become more common practice in research originating from developed countries (15,40). In contrast, there are relatively few studies that have used this technology to study children in developing countries like Brazil. Thus, the purpose of this article is to analyze the independent association of accelerometer-determined sedentary behavior, a complete range of physical activity intensities, and steps/day with body composition variables among Brazilian children averaging 10 years of age.
Methods Study Sample This study was part of the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE), a multinational cross-sectional study conducted at sites from twelve countries (Australia, Brazil, Canada, China, Colombia, Finland, India, Kenya, Portugal, South Africa, United Kingdom, and United States) (19). The analysis herein focuses only on the data collected in the city of São Caetano do Sul, located in the state of São Paulo, Brazil, with a land area of 15.3 km2 and a subtropical climate. The population of the municipality in 2013 consisted of 149,263 inhabitants, including 1,557 children (812 boys and 745 girls) 10 years of age (16). The city is characterized as a service economy (16) and has the best Human Development Index in Brazil (0.86) according to the United Nations Program for Development (28). There is variability in socioeconomic status between schools in the region of São Caetano do Sul. Schools were recruited proportional to the distribution of public and private school attendance. Public schools represent the lower and lower-middle socioeconomic strata, while private schools reflect the middle and upper-middle class. A complete list of public and private schools enrolling 5th grade students in São Caetano do Sul was assembled. Public and private schools were sampled separately and schools were selected from each list at a ratio of 4 (public) to 1 (private). All schools were placed in random order within each stratum and each school was approached according to the random order established within each stratum. This 80% public to 20% private schools ratio was purposely implemented to maximize SES distribution. If a school refused to participate in the study, it was replaced by the next school on the list. Twenty schools were sampled (16 public and 4 private) to generate a sample of 25–30 children from each school with a stipulation that each sex comprise 50% of the selected sample, resulting in a minimum enrollment of 500 5th grade children. Given that the primary sampling strategy was based on schools, International Study of Childhood Obesity, Lifestyle and the Environment data collection was conducted during the school year (from March 2012 to April 2013). Children were eligible to participate in International Study of Childhood Obesity, Lifestyle and the
Environment if they: (a) were 9–11 years of age; (b) regularly enrolled in a school in São Caetano do Sul system; and (c) did not have clinical or functional limitations preventing daily physical activity. Before participating, children and at least one of their parents/legal guardians were asked to sign the Instrument of Consent according to Resolution 196/96 of Brazil’s National Health Council. The overarching International Study of Childhood Obesity, Lifestyle and the Environment protocol was approved by the Pennington Biomedical Research Center Institutional Review Board and Federal University of São Paulo, Brazil.
Accelerometry The Actigraph GT3×+ accelerometer (ActiGraph, Ft. Walton Beach, United States) was used to objectively monitor sedentary behavior, the complete range of incremental intensities of physical activity (including light, moderate, vigorous and moderate-to-vigorous physical activity), and steps/day. The accelerometer was worn at the waist on an elasticized belt, on the right midaxillary line. The participants were encouraged to wear the accelerometer 24 hr/day for at least 7 days (plus an initial familiarization day and the morning of the final day), including 2 weekend days. The minimal amount of accelerometer data that was considered acceptable for analytical purposes was 4 days (including at least one weekend day) with at least 10 hr/day of waking wear time (10,36). Due to the 24-hr wear protocol, the nocturnal sleep time was identified and excluded from any analysis of sedentary behavior, intensities of physical activity and steps/ day. The nocturnal sleep duration was estimated from the accelerometry data using a published automated algorithm (4,37). Briefly, the algorithm estimates the sleep period using a series of steps which identify accelerometer-determined markers for sleep onset and offset, as well as extended episodes of wakefulness separating the sleep period into distinct sleep episodes with multiple sleep onsets and offsets (4). After exclusion of the nocturnal sleep period time, waking nonwear time was defined as any sequence of at least 20 consecutive minutes of zero activity counts. Following the final day of data collection, staff went into the school and retrieved the accelerometers. The research team verified the data for completeness (4 days of at least 10 hr wear time) using the most recent version of the ActiLife software (version 5.6 or higher; ActiGraph, Pensacola, FL) available at the time. Data were collected at a sampling rate of 80 Hz, downloaded in 1-s epochs, and were aggregated to 15-s epochs for analysis (11). We chose to use the accelerometer activity count cut-points established by Evenson et al. (11) for 15-s epochs. The cut-points capture the sporadic nature of children’s activity and provide the best classification accuracy among the currently available cut-points for physical activity in children (36). Specifically, sedentary
PES Vol. 27, No. 3, 2015
382 Ferrari et al.
behavior was defined as time accumulated at ≤25 activity counts/15 s, ≥26–573 activity counts/15 s for light physical activity, ≥574–1002 activity counts/15 s for MPA, ≥1003 activity counts/15 s for VPA, and ≥574 activity counts/15 s for moderate-to-vigorous physical activity. Further, we included overall steps/day as an easily interpretable measure of total “volume” of physical activity over the entire day.
Body Size and Composition The body size and composition variables included were body height, body weight, BMI, and body fat percentage. A battery of anthropometric measurements was conducted according to standardized procedures. Body height was measured without shoes using a Seca 213 portable stadiometer (Hamburg, Germany), with the participant’s head in the Frankfort Plane (21). Body weight and body fat percentage were measured using a portable Tanita SC-240 body composition analyzer (Arlington Heights, IL) after all outer clothing, heavy pocket items, and shoes and socks were removed (5). Two measurements were obtained, and the average was used in analysis (a third measurement was obtained if the first two measurements were more than 0.5 kg or 2.0% apart, for body weight and body fat percentage, respectively). BMI (kg/m2) was calculated from measured height and weight.
Statistical Analysis Descriptive statistics included means, standard deviations, or frequencies as appropriate. A KolmogorovSmirnov test was applied to evaluate the data distribution. Differences between sexes were analyzed using a t test for independent samples and chi-square tests were used for categorical data. A Pearson correlation was used to assess the association of sedentary behavior, different physical activity intensities, and steps/day with body size and composition variables (12). Multilevel linear regression models were used to examine the independent associations between sedentary behavior, different physical activity intensities, and steps/ day with body composition (BMI and body fat percentage) variables. Our first model (model 1) was adjusted for sex and school (to allow for clustering at the school level). The second model (model 2) was additionally adjusted for sex, school, and moderate-to-vigorous physical activity. We also adjusted for moderate-to-vigorous physical activity (min/day) to examine whether the associations with time spent in sedentary behavior were independent of time spent moderate-to-vigorous physical activity. The third model (model 3) was adjusted for sex, school, and sedentary behavior. We additionally adjusted for sedentary behavior (min/day) to examine whether the associations with time spent in each of the different physical activity intensities was independent of time spent in sedentary behavior. Statistical Analysis System (SAS, version 9.3) was used for data analyses and p < .05 was adopted as the significance level (32).
Results The analysis sample included 485 children (238 boys, 247 girls). Table 1 presents the accelerometer-derived data and body composition variables. For the total sample (boys and girls combined), the most accelerometer-determined waking wear time was spent in sedentary behavior (55.7%), followed by light physical activity (37.6%) and moderate-to-vigorous physical activity (2.8%). Boys wore the accelerometer for a significantly longer waking wear time than girls; however, there was no significant difference in the number of valid days of monitoring obtained between boys or girls. Girls accumulated significantly more minutes in sedentary behavior than boys; however, no differences were observed in light physical activity between boys and girls. Boys were consistently higher than girls with regards to time spent in both moderate physical activity and VPA. On average, boys accumulated significantly more minutes in moderate-to-vigorous physical activity (23 min/day) than girls. Boys also accumulated 1,850 steps/day more than girls (p < .001). There were no significant differences between boys and girls for body height, body weight, and BMI. Boys had lower body fat percentage values than girls (21.3% and 24.7%, respectively) and this difference was statistically significant. Table 2 presents the results of the correlation analysis describing associations between the different accelerometer-derived variables and body composition (BMI and body fat percentage) variables by sex. In boys, sedentary behavior was only weakly and positively associated with body fat percentage; however, there were no significant associations between light physical activity and the body composition variables. On the other hand, there were weak and moderate negative associations between moderate physical activity, VPA, and moderate-to-vigorous physical activity and both of the body composition variables. Steps/day was moderately and negatively associated with both of the body composition variables. In girls, sedentary behavior was not associated with either of the body composition variables. Light physical activity and moderate physical activity were not significantly associated with either of the body composition variables in girls; however, VPA was weakly and negatively associated with each of the body composition variables. Moderate-to-vigorous physical activity and steps/day were not associated with either of the body composition variables in girls (Table 2). For boys and girls combined, Table 3 presents the results of the regression analysis describing the association between each of the accelerometer-derived variables and the body composition variables. There were no significant associations between sedentary behavior or light physical activity and any of the body composition variables when adjusted for sex, school, and moderateto-vigorous physical activity. Moderate physical activity, VPA and moderate-to-vigorous physical activity were
PES Vol. 27, No. 3, 2015
Table 1 Descriptive Analysis (Mean and SD) of Accelerometer-Derived Data and Body Composition Variables of 10-Year-Old Brazilian Children: The International Study of Childhood Obesity, Lifestyle and the Environment Variables Age (years)
Both (n = 485)
Boys (n = 238)
Girls (n = 247)
p-value
10.1 (0.5)
10.1 (0.5)
10.1 (0.5)
.9712
Waking wear time (min/day)
897 (50)
904 (50)
890 (50)
