Journal of Physical Activity and Health, 2015, 12, 604  -609 http://dx.doi.org/10.1123/jpah.2014-0024 © 2015 Human Kinetics, Inc.

ORIGINAL RESEARCH

Colombian Children With Overweight and Obesity Need Additional Motivational Support at School to Perform Health-Enhancing Physical Activity Patricia Olaya-Contreras, Myriam Bastidas, and Daniel Arvidsson Aims: The aim of this study is to investigate associations of screen-time and physical activity (PA) with self-efficacy for PA, intrinsic motivation to PA and health-related quality of life (HRQoL) in Colombian schoolchildren from socioeconomically disadvantaged neighborhoods, and to compare these variables among children with normal-weight (NW), overweight (OW) and obesity (OB). Methods: In 678 schoolchildren (age 10–14 years) screen-time (TV, video games, computer) and number of days being physically active ≥ 60 minutes were self-reported. Multi-item scales were used to assess self-efficacy to PA and intrinsic motivation to PA. The KIDSCREEN-27 was used to assess HRQoL. Results: Screen-time was associated with HRQoL in the school/learning environment dimension. Number of days being physically active was associated with self-efficacy for PA, intrinsic motivation for PA and with HRQoL concerning physical well-being, autonomy/parent relation and social support/peers. Group differences were found for days being physically active (OW = 2.8 and OB = 2.7 vs. NW = 3.3) but not for screen-time (NW = 5.0, OW = 4.7 and OB = 5.7 hrs·d-1). OW and OB scored lower on intrinsic motivation to PA than NW (OW = 19.2 and OB = 17.9 versus NW = 20.1). All 3 groups differed in physical well-being scores (NW = 50.3, OW = 48.1, OB = 40.6, P < .001). Conclusions: Schoolchildren with overweight and obesity from socioeconomically disadvantaged neighborhoods need additional motivational support to perform health-enhancing PA to experience higher physical well-being. Keywords: intrinsic motivation, self-efficacy, health-related quality of life, sedentary behavior

South America is undergoing rapid socioeconomic and demographic transformations accompanied by changes in lifestyle behaviors and disease profiles.1 There has been a dramatic increase in overweight and obesity among children, causing detrimental metabolic alterations, which increase the risk of cardiovascular disease, diabetes and cancer.2,3 One determinant to becoming overweight and obese is a sedentary lifestyle, where outdoor play has been replaced by indoor entertainment such as television viewing, video games and internet.3 In Colombia, a low level of physical activity (PA) in children has been reported.4,5 Colombia is a country with large socioeconomic inequalities in health-promoting resources,6 which may include access to playgrounds, exercise spaces/facilities, green-areas as well as to adults/leaders for motivational support to participation in PA. It is therefore postulated that children from socioeconomically disadvantaged neighborhoods are less likely to perform health-enhancing PA compared with their more privileged peers. They are frequently exposed to violence and abuse with negative cognitive and behavioral impacts, and outdoor activities in their neighborhoods are often considered unsafe.6 National surveys show that the prevalence of overweight and obesity increased from 13.9% to 17.5% in children 5 to 17 years old between 2005 and 2010.7 Although the prevalence of overweight and obesity is still higher in the more socioeconomically privileged children,7,8 there are indications of an incipient rapid increase among the less privileged.2,9,10 A higher increase in overweight and obesity has been shown in Hispanic children and children from low-income households in the United States.11 Olaya-Contreras is with the Faculty of Nursing, University of Antioquia, Medellín, Antioquia, Colombia. Bastidas is with the Dept of Pediatrics, University of Antioquia, Medellín, Antioquia, Colombia. Arvidsson (daniel. [email protected]) is with the Dept of Clinical Sciences, Lund University, Malmö, Sweden. 604

Physical activity and physical inactivity are considered as 2 distinct behaviors with independent influences on health. 12 Screen-time (TV, video games, computer) has often been used as an indicator of physical inactivity,13 and it has shown to be higher in overweight and obese children compared with normal-weight children.14–16 The same has been found in Colombian children,7 although some studies showed no difference.8,17 Physical activity has also been associated with indices of adiposity.15,16,18 Studies of Colombian children show inconsistent results concerning the association of PA with adiposity.4,8,17 Therefore, physical activity behaviors in Colombian children with overweight and obesity need further investigations to clarify their role and relation to adiposity. Several mediators to children´s PA have been identified in the literature.19–21 Some of them concern the cognitive and behavioral processes by which individuals consolidate their self-conception and alter their behaviors or the environment to reach outcomes matched with their self-perception and goals (ie, self-regulation).21 These mediators are conceptualized within social cognitive theories and may be of interest to study in overweight and obese children that live in socioeconomically disadvantaged neighborhoods. Two of these mediators are self-efficacy and motivation. Self-efficacy is conceptualized within the Social Cognitive Theory and describes the confidence in being physically active in specific situations.20,21 It is one of the most investigated mediators in relation to PA.19–21 Motivation is central in the Self-Determination Theory.22,23 This theory distinguishes between different types of motivation, ranging from amotivation at the one end, to autonomous (or intrinsic) motivation at the other end. A child with intrinsic motivation is described as having high sense of autonomy, competence and relatedness to other (ie, joyful participation in physical activities).23 Higher self-efficacy and intrinsic motivation have both been associated with more PA in children.19–21,23,24

Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 5

Physical Activity and Obesity in Colombian Schoolchildren   605

The KIDSCREEN project investigates health-related quality of life (HRQoL) in children, which is a multidimensional concept including physical and psychological well-being. There has been an increased interest in measuring HRQoL in children with overweight and obesity. The European KIDSCREEN Survey found the HRQoL for many of its dimensions to be lower in these groups compared with children with normal weight.25,26 There is little data on social cognitive mediators and HRQoL in relation to PA and obesity in children from developing countries such as Colombia.20 Access to this data may support decisions on intervention programs in these countries. We hypothesize that self-efficacy for PA, intrinsic motivation to PA and HRQoL may be potential predictors of screen-time as indicator for physical inactivity and PA. Furthermore, self-efficacy for PA, intrinsic motivation to PA and HRQoL is lower in children from socioeconomically disadvantaged neighborhoods, especially in those with overweight and obesity. Therefore, more screen-time and less PA may be observed in Colombian children with overweight and obesity from socioeconomically disadvantaged children compared with those with normal weight. The first aim of the current study was to investigate associations of screen-time and PA with self-efficacy for PA, intrinsic motivation to PA and HRQoL in Colombian schoolchildren from socioeconomically disadvantaged neighborhoods. The second aim was to compare these variables among children with normal weight, overweight and obesity.

Methods Study Design and Participants This is a cross-sectional study performed in children 10 to 14 years old in Medellin, Colombia. Participants were recruited from grade 6, 7, and 8 in one of the largest public schools in Medellin. This school receives about 7000 children with the majority from socioeconomically disadvantaged neighborhoods in the urban areas of Medellin (strata 1 to 3). At the time of recruitment the grades 6, 7, and 8 comprised of 2000 children. A total sample size was calculated with a confidence level of 95% and a design effect of 1.0%. A representative subsample from each grade was recruited using a randomized stratified sampling (EPIDAT-4, 2012). The samples consisted of 163 students from grade 6 (10–11 years), 229 from grade 7 (12–13 years) and 262 students from grade 8 (14 years), for a total of 654 students. A final sample of 678 children was recruited to account for possible drop-outs. Informed consent was obtained from the children and their parents or caregivers. Ethical clearance to conduct the study was provided by the Human Subject Ethics, Research and Biosecurity Board committees of the University of Antioquia (ACTA No CEI-FE 2013–05; Faculty of Nursing, University of Antioquia). The measurements were performed in the school under the supervision of the school nurse and the study researchers, from January 2013 to June 2013.

Body Mass Index Categories Body height was measured to the nearest millimeter using a stadiometer and body weight to the nearest 100 g using a calibrated digital scale. The children wore light clothes and no shoes during the measurements. Body mass index (BMI) was calculated as weight (kg) / height (m)2. The WHO Reference 2007 was used to derive age and gender specific BMI z-scores and to divide the children

into 4 BMI categories: underweight (z-score < –1.0), normal-weight (z-score = –1.0 to +1.0), overweight (z-score > +1.0 to < +2.0), and obese (z-score > +2.0).27–29 We used the cut-off for underweight corresponding to grade 1 thinness in adults (z-score = –1.0).29 The WHO AnthroPlus software version 3.2.2 was used to derive the BMI z-scores (WHO, Geneva, Switzerland). For the present investigation, underweight children were excluded from the analyses as their nutritional and energy level may limit their ability to perform normal PA (N = 75).

Screen-Time and Physical Activity Preexisting and newly created PA and sedentary behavior questions were gathered and culturally adapted in the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE).30 This battery of questions was also applied to the children in the current study. In this study, screen-time and days being physically active were the target variables for the analyses as indicators of physical inactivity and PA. For the questions about time spent watching TV and playing videogames/using computer not for school-work purposes during the latest week, the children had to choose between the answer alternatives 0, < 1, 1, 2, 3, 4, and 5 hours or more. The < 1 hour alternative was given the value 0.5 for the statistical analyses. Total screen-time (hours·d-1) was created by the sum of the time spent watching TV and playing videogames/ using computer not for school-work. Physical activity was assessed from the question asking about how many days in the past week the children were physically active for a period of at least 60 minutes and at an intensity that increased their heart rate and made them breathe hard.

Self-Efficacy and Intrinsic Motivation The ISCOLE incorporated culturally adapted questions to assess self-efficacy for PA and intrinsic motivation to PA.30 Self-efficacy for PA was assessed from 8 items with statements such as “I can be physically active during my free time on most days” that are rated on a 5-point scale from “Total disagreement” (1) to “Total agreement” (5). This 8 items self-efficacy assessment method presents satisfying psychometric properties.31 The self-efficacy score was calculated by the sum of the ratings from the 8 items. Intrinsic motivation was assessed from 5 items with statements such as “I take part in exercise because other people say I should” or “It is important to me to exercise regularly” and rated on a 5-point scale from “Never true for me” (1) to “Very true for me” (5). The 5 items represent the different types of motivation described in the Self-determination Theory and is a shortened adaption from the original 15 items previously evaluated for its psychometric properties.32 The score for intrinsic motivation was calculated by the sum of the rating from the 5 items, considering reversed ratings for some of the items.

Health-Related Quality of Life The KIDSCREEN-27 was developed to assess HRQoL in 5 dimensions: physical well-being, psychological well-being, autonomy/ parent relation, social support/peers, and school environment. It has been shown to be valid to assess HRQoL in European as well as Colombian children and adolescents.33,34 The items included in each dimension are rated on a 5-point scale to assess frequency (never-seldom-quite often-very often-always) or the intensity (not at all-slightly-moderately-very-extremely) of a certain behavior, feeling or attitude referring to the last week. Scores are created for

JPAH Vol. 12, No. 5, 2015

606  Olaya-Contreras, Bastidas, and Arvidsson

each dimension following a standardized calculation procedure established by the KIDSCREEN project, where higher scores indicate higher HRQoL.33

Statistics

Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 5

Linear regression was used for association analyses, and one-way analysis of variance (ANOVA) with post hoc tests (Tukey´s Test) was used for group comparisons except for gender distribution where a chi-square test was used. The variables studied in the regression analyses and in the ANOVA showed sufficient normality according to distribution of frequencies and residuals, and did not required transformations. Age and gender were considered covariates in the association analyses and group comparisons. Significance was considered at P < .05. All statistical analyses were performed with SPSS Statistics version 20.0 (IBM Corporation, NY).

Results In the whole sample recruited (N = 678, including underweight children), the prevalence of overweight was 19% and obesity was 4%. The 3 BMI categories included in the analyses (normal-weight, overweight and obesity, N = 603) differed in age and gender distribution, as well as in body weight and BMI z-score (Table 1).

For the whole group, higher HRQoL in the school environment dimension was statistically significant associated with less screen-time (Table 2). The other HRQoL dimensions were not associated with screen-time. In contrast, higher self-efficacy for PA, intrinsic motivation to PA, and HRQoL for the dimensions physical well-being, autonomy/parent relation, and social support/ peers were significantly associated with more days being physically active (Table 2). Significant group differences were found for days being physically active but not for screen-time (Table 3). Children with overweight and with obesity both reported less days being physically active than the normal-weight group, but the difference was statistically significant only for the overweight group. Furthermore, children with overweight and with obesity both scored lower on intrinsic motivation to PA compared with normal-weight children, while there was no statistically significant group difference in self-efficacy for PA (Table 3). The intrinsic motivation score decreased across BMI categories, although this pattern did not reach statistical significance. The score for each of the 5 HRQoL dimensions decreased across BMI categories (Table 3). However, a statistically significant group differences were found only in the physical well-being dimension. This score differed between all the 3 BMI categories, with the lowest score found in the children with obesity.

Table 1  Characteristics of Study Participants by BMI Category (N = 603) BMI category Normal weight, N = 451 (66%)

Overweight, N = 127 (19%)

Obese, N = 25 (4%)

P

13,1 (0,9)a

12,9 (0.9)b

12.4 (1.0)c