Can the Weight Reduction Program Improve Obese Thai Adolescents ’ Body Mass Index and Autonomous Motivation ?

Introduction: Overweight and obesity has impacted both the physiological and psychological health of children and adolescents. Evidence of cross-cultural studies has showed mixed results regarding weight reduction program (WRPs) based on self-determination theory (SDT) and social cognitive theory (SCT) affecting students with overweight and obesity. This study examined the effectiveness of a WRP on body mass index (BMI) and autonomous motivation in overweight and obese Thai adolescents.Material and Methods: An experimental study with intervention and control groups was employed. Participants were 304 Thai obese 7th- to 12th-grade students, ranging in age from 12 to 19 years. The intervention group was assigned to a program based on SDT and SCT, consisting of (1) a 12-week weight control phase and (2) a 12-week follow-up phase. The control group was assigned to a general health education program. The data were collected via psychometric questionnaires and anthropometric data.Results: There were no statistical differences in adjusted means of BMI and autonomous motivation of physical activity and dietary intake occurred among participants in the two groups. Compared to those in the control group, participants in the intervention group showed a 0.4 smaller increase in BMI and a 0.4 greater decrease in autonomous motivation of dietary intake.Conclusion: The WRP based on SDT and SCT showed no apparent effectiveness in improving BMI or autonomous motivation among these participants. Future studies should incorporate program that effectively harmonize with core school teaching activities and schedules.


Introduction
C hildhood and adolescent overweight and obesity have increased rapidly in both developed and developing countries 1,2,3,4 .Obesity rates in children and adolescents in Thailand were 5.8%, 6.7%, and 9.7% in 1995, 2001, and 2008-2009, respectively 5 .Childhood and adolescence obesity has both short-and long-term physical and psychosocial impacts 6,7 .It can progress into adult obesity 7,8 resulting in an increased rate of chronic disease in adulthood 7 , Schools offer important settings for implementing weight reduction programs (WRPs) as they can reach a large number of youths 9,10,11,12 .WRPs help prevent and treat obesity in adolescents 13 .Motivation is a strong predictor of weight reduction in children and adolescents in WRPs 13 .Previous school-based interventions demonstrated that youth who participate in enjoyable and engaging activities to enhance autonomous motivation experience more positive results in body mass index (BMI), healthy diet, and physical activity 10,14,15 .Few studies have evaluated autonomous motivation for physical activity and dietary intakes in school-age (particularly secondary school) students, but their fi ndings are appropriate for Western contexts 14,15,16,17 .
Social cognitive theory (SCT) 9,10,11 and selfdetermination theory (SDT) 17,18 have been widely used in school-based WRPs and in Western contexts.Grounded SCT 19 came from the triadic reciprocal determinism evident in interaction patterns among personality, behaviors, and environments.A person's ability to originate and direct actions for given purposes affecting his or her environment and behaviors is called personal agency and termed self-effi cacy 20 .Self-effi cacy refers to a person's belief in his or her ability to achieve behaviors and is refl ected in the capacity to control his or her motivation, behavior, and environment.Self-regulation is directly affected by self-effi cacy and a correlate of activity.Self-regulation involves individual processes to attain goals and behaviors 19 , thereby enhancing skill development and motivational activities.Empirical studies of WRP based on SCT found positive outcomes, such as reducing BMI and participating in activities 9,10,11 .
SCT's concept is rather similar to self-determined motivation from SDT, which refers to people's capacity to refl ect upon and engage in tasks of their choosing, not because of external rewards or pressures 21 .Few studies have applied SCT to school interventions in Thailand; those that have showed inconsistent impacts on weight measures and psychometric variables 22,23,24 .Results do not provide clear guidance for decreasing excessive weight through school-based programs 22,23,24 .
SDT could provide a useful framework for understanding individuals' task participation.SDT has been applied in studying health behavior interventions (e.g., physical activity and healthy diet), with some positive outcomes 17,18,25 .SDT identifi ed two types of motivation-autonomous motivation and controlled motivation 21,26 through six steps of continual motivation development: amotivation, external regulation, introjected regulation, identifi ed regulation, integrated regulation, and intrinsic motivation.External through integrated regulation (the second to fi fth types) rely on extrinsic motivation or a higher hierarchy development of controlled motivation 21,27 whereas intrinsic motivation occurs through one's inner self or fundamental psychological needs-namely, autonomy, relatedness, and competence 27 .Autonomy is the need to experience individual actions resulting from self-endorsement and perceived choice; competence is the need to experience one's capability to control an environment and predict outcomes; and relatedness is the need to experience satisfaction through involvement with one's connectedness 21 .
WRP combines SDT and SCT to increase the intervention's effectiveness 15 .Previous studies incorporating both theories have demonstrated stronger relationships among self-determined motivation of physical activity, healthy diet, and maintained weight loss 14,15 .However, no studies have yet applied both SDT and SCT to WRP for overweight and obese Thai secondary school students.It is hypothesized that, by participating in WRP based on SDT and SCT, these students will improve their BMI and autonomous motivation for physical activity and dietary intake.This study measures the effectiveness of WRP on BMI and autonomous motivation among such students.

Materials and Methods
This experimental study involved a pretest, posttest, follow-up with a control group for evaluating the effectiveness of school-based physical activity and dietary plans.Measurements were tested at weeks 0, 12, and 24.The design and enrollment of this study followed Consolidated Standards of Reporting Trials (CONSORT) guidelines 28 , as shown in Figure 1.
The fi rst step was a purposive sampling of two largest secondary schools in two districts with similar characteristics (e.g., number of students, health promotion school policy, and willingness to participate in the study) in the Secondary Educational Service Area Offi ce 37 in the Nan Province in Thailand.One school was then randomly assigned into the intervention while the other into the control groups.In the third step, one classroom was simple randomly selected each for the 7 th -to 12 th -grade classrooms in each school.Students with the BMI for age and sex more than or equal to the 85 th percentile when using WHO growth standard reference [WHO, 2007]  29 and volunteer to participate with the allowance of the parents were recruited into the study.Those with the existing musculoskeletal and cardiovascular diseases were however excluded.Sample size was calculated by a formula for the two populations' means and relying on the data about the mean BMI for age and sex from Duangchan's study 22 .The possibility of 20% dropout was also compensated, resulting in 160 students per group.However, 10 and 6 students in control and intervention groups who were later detected to have normal BMI were excluded after baseline assessments.A total of 304 participants were then included in this study.These participants were mostly males (intervention group: n=154, 81 males, 52.6%; control group: n=150, 79 males, 52.7%) and aged between 12 to 19 years.The detailed data are presented in Figure 1  The WRP used in this study was a school-based curriculum developed by the authors that combined health education and physical exercise and used SCT and SDT for the program's scope.The program's contents and activities were multidisciplinary in nature and were contributed by a nurse, two health promotion experts, a psychologist, a dietitian, an exercise expert, and a school teacher.The school teacher was a health education and physical exercise teacher who cooperated on lesson plan and content development, and was the sole teacher during eight sessions.The program included two phases.The fi rst 12-week weight-control phase consisted of eight teaching sessions (overweight and obesity, dietary, physical activity, weight control, and negative emotional management) led by the school teacher (40 to 60 minutes each), once a week for the fi rst four sessions and once every two weeks for the last four sessions.In addition, the 30-minute free-choice physical activity sessions were offered three times a week, while participants engaged in their favorite exercises and physical activity as a group or on their own during exercise times.Moreover, they were asked to take notes on their daily physical activity and dietary intakes.The second 12-week follow-up phase, there was no structured activity except for some stimulation techniques such as individual and group meetings, supports, monitoring, good communication, and group counseling.Examples of the WRP component details are presented in Table 1.

Measures
1. Outcomes Measures: The primary outcome was BMI, which was used to evaluate the effectiveness of this program.The secondary outcomes were the autonomous motivation of physical activity and dietary intakes.

Anthropometric
Measures: Heights were measured to the nearest millimeter using a portable stadiometer.Weights (in kilograms) were measured using a portable scale; participants were weighed without shoes and in light clothing.BMI was calculated, and the WHO growth standard reference (WHO 2007) 29 cut-points were determined as normal (BMI for age and sex < 85 th percentiles), overweight (BMI for age and sex ³ 85 th percentiles to < 95 th percentiles), and obesity (BMI for age and sex ³ 95 th percentiles).Statistical analysis: Descriptive statistics were used to describe participants' characteristics.Mean differences and adjusted mean differences of the dependent variables among the intervention and the control group at 12 th and 24 th weeks were compared using an independent t-test and simple linear regression analysis, respectively.Controlling confounders were done using a multiple linear regression analysis.The cut-off point for statistical signifi cance was p value <.05.

Results
Participants: Ultimately, 304 Thai overweight and obese secondary school students were recruited from the intervention and the control schools.Only daily school pocket money (Baht/day) differed between the two groups at the baseline.Data are presented in Table 2. BMI: In the 12 th week of the program, the intervention group showed a signifi cantly increase in BMI 0.4 kg/m 2 (95% CI = 0.1 to 0.7, p =.017) compared to the control group; in the 24 th week of the program, there was no signifi cant difference in BMI between the groups (p >.05), after variables were adjusted at baseline.Data are presented in Table 3.

Autonomous motivation of physical activity (RAI):
In the 12 th and 24 th weeks of the program, only the intervention group had signifi cantly increased in the amotivation score by 0.4 points (95% CI = 0.2 to 0.5, p <.001) and 0.2 points (95% CI = 0.0 to 0.4, p =.021) when compared to the control group, but the two groups were not signifi cantly different for scores of RAI, external regulation, introjected regulation, identifi ed regulation, or intrinsic regulation (p >.05) after variables were adjusted at baseline.Data are presented in Table 4.

Autonomous motivation of dietary intakes (RAMI):
In the 12 th week of the program, the intervention group's RAMI score was signifi cantly decreased by 0.4 points (95% CI = -0.6 to -0.1, p =.003), while its amotivation scores signifi cantly increased by 0.4 points (95% CI = 0.1 to 0.6, p =.004) and the externally controlled motivation scores signifi cantly increased by 0.3 points (95% CI = 0.1 to 0.5, p =.010).There were no signifi cant differences in the autonomous motivation score compared to the control group (p >.05).All variables were adjusted at baseline.Data are presented in Table 5.
Similarly, in the 24 th week of the program, the intervention group's RAMI score signifi cantly decreased by 0.3 points (95% CI = -0.5 to 0.0, p =.046), while its amotivation scores signifi cantly increased by 0.3 points (95% CI = 0.0 to 0.5, p =.046) and the externally controlled motivation scores signifi cantly increased by 0.3 points (95% CI = 0.0 to 0.5, p =.025).There were no signifi cant differences in the autonomous motivation score compared to the control group (p >0.05).All variables were adjusted at baseline.Data are presented in Table 5.    a Comparison mean difference of physical activity motivation change (the intervention -control groups) at weeks 12and 24 by using simple linear regression; b Adjusted for knowledge, school level, Identifi ed regulation, Intrinsic regulation, RAI and that motivation subscales at baseline (week 0) by using multiple linear regression.95% CI = 95% confi dence interval.a Comparison mean difference of weight reduction (the intervention -control groups) at weeks 12 and 24 by using simple linear regression; b Adjusted parents' physical characteristics, daily school pocket money,physical activity behaviors, eating behavior, triceps skinfold thickness, subscapular skinfold thickness and BMI at baseline (week 0) by using multiple linear regression.95% CI = 95% confi dence interval.

Table 1: Examples of program component details
.063 -0.3 (-0.5, 0.0) 0.046* Note.Comparison of mean dietary intake motivation change at weeks 12 and 24 by using independent t-test.a Comparison mean difference of dietary intake motivation at weeks 12 and 24 by using simple linear regression; b Adjusted for knowledge, school level, Identifi ed regulation, Intrinsic regulation, RAI and that motivation subscales at baseline (week 0) by using multiple linear regression.95% CI = 95% confi dence interval.

Discussion
A multi component WRP based on SCT and SDT was developed to evaluate its effects on BMI and autonomous motivation toward physical activity and dietary intakes among overweight and obese Thai students.A multiple linear regression found no statistical differences in adjusted means of BMI and autonomous motivation of physical activity and dietary intake in either participant group.The intervention group showed a 0.4 smaller increase in BMI than the control group, which is similar to the PRALIMAP project's 0.11 lower increase 34 .
Meanwhile, the intervention group showed a 0.4 greater decrease in autonomous motivation of dietary intakes (RAMI) than the control group, but no difference in autonomous motivation of physical activity (RAI).Increased amotivation, one form of motivation calculated for RAMI and RAI composite scores, was evident for both autonomous motivation of dietary intakes and physical activity.Previous research has indicated that programs might pressure youth to become physically active 35 ; this might have infl uenced the RAMI and RAI scores in the current study.This study also found that students did not continuously self-record for physical activity and dietary intakes in their individual notebooks due to the low levels of self-regulation and self-control.A previous study described the correlation among amotivation, self-regulation and self-control, and the inability to lose weight in the short term 36 .Therefore, using external rewards and praise (e.g., a colorful cartoon bundle and compliment sticker) twice during the program (in the intervention group) to create motivation might not be enough to make students with a high level of amotivation change their behavior and reduce their weight.No signifi cant differences emerged in the externally controlled motivations before and after the experiment, meaning behaviors can be maintained in the short term, but not the long term 37 , Externally controlled motivation is a controlled form of motivation created by reward 38 .This study found that the intervention and control groups showed no difference in eating based on externally controlled motivation.
Moreover, participants in this study lacked the readiness to change, and many lacked motivation for exercise and food intake.High amotivation scores were J. Nepal Paediatr.Soc.related to low autonomous motivation.At the 12 th and 24 th weeks, the intervention group had an exercise amotivation score (≥ 2 points, total 4 points) of 74.5% and 73.3% and an eating amotivation score (≥ 3 points, total 6 points) of 87.7% and 86.1%, respectively.Interestingly, the results indicated that BMI was related to autonomous motivation.The intervention group's BMI was signifi cantly lower at weeks 12 and 24 [27.7 (±3.67), p = 0.001 and 27.4 (±3.72), p <0.0001] than the baseline (data not shown), inferring that they were still obese.Individuals' autonomous motivation of physical activity and dietary intakes did not increase in the obesity group.Hwang and Kim 39 found that overweight and obese adolescents had lower scores on intrinsic motivation but higher scores on demotivation and extrinsic motivation than normal weight adolescents.The school had many extra-school learning opportunities; participants who joined the school's and program's activities at the same time might have experienced stress and coped poorly, resulting in no increase in the autonomous motivation of physical activity and dietary intakes.
The WRP showed no effect on BMI and autonomous motivation of physical activity and dietary intakes.Various explanations may account for this result.First, the adherence rate was low.Adherence and completion of the intervention are correlated with positive outcomes 40 .This study's low adherence rate stemmed from unplanned changes in school schedules or extra learning and teaching activities (e.g., national standardized test preparation and religious and traditional activities).During the initial weight control phase, 23.1% to 62.8% (n = 78) of middle school students joined the 8-period classroom education project, compared to 13.2% to 56.6% (n = 76) of high school students; 10.3%to 75.6% of middle school students joined physical activity 3 times a week (through week 12) compared to 6.6% to 26.3% of high school students.During the follow-up phase (2 times), 41.0%to 48.7% of middle school students and 30.3% to 18.4% of high school students participated (data not shown).
Second, some parts of the motivational questionnaires showed lower reliability.Instruments for measuring autonomous motivation of physical activity and dietary intakes of Thai adolescents are quite limited.Cross-cultural motivation tools might be less sensitive with Thai youths.Third, WRPs can help children and adolescents succeed in losing weight over 6 to 12 months 13 .Such programs require a minimum of 24 weeks (6 months).The intervention group's lacked the readiness to change and motivation for exercise and food intake were a different fi nding from previous studies, which found that longer WRPs improved autonomous motivation.Verloigneet et al 16 .found that 177 obese participants (mean age 15.1 years) who joined a 10-month exercise and dietary program showed increasing intrinsic motivation.Saavedra et al. 17 found that participants in an exercise-only program and exercise and dietary program for 6 months to 3 years showed increased levels of intrinsic motivation and exercise; such participation infl uenced the persistence of internal motivation when engaged in the program.This fi nding is consistent with SDT's assertion that participating in activities for a long period contributes to intrinsic motivation 26,41 .The currently studied 24-week research program might not have been long enough to make changes in students' dietary self-determination motivation.
Finally, the internal and external school environments (i.e., food shops) were less controlled.These changing environments make this experimental design less appropriate for fi eld study.The intervention group was exposed to unhealthy foods more often than the control group through a 24-hour convenience store near the intervention school's entrance and a temporary market selling unhealthy foods near the school.These stores were not accessible from the control school.A greater availability of convenience stores leads to higher BMI in adolescents 42 .
Despite the study's strengths (i.e., the teacher leading this program participated in the program and lesson plan development processes, taught all eight sessions himself, and specialized in teaching middle and high school students), several limitations exist.For example, the experimental study results might not be generalizable to both public and private secondary schools in Thailand.However, when implementing the program, many extra school activities and schools did not plan to change school teaching periods, thereby affecting an important reduction in participation program activities.In addition, it was diffi cult to combine middle and high school students into one classroom due to different schedules.

Conclusions
The WRP based on SDT and SCT provided insuffi cient evidence of effects on BMI and autonomous motivation among overweight and obese secondary school students.Future studies should incorporate program activities with core teaching activities.They should also consider using sensitive autonomous motivation tools for Thai youths, taking long period of the program and creating supportive environments for weight loss and weight control in schools.

2 (
BREQ-2) 30 : The BREQ-2 measures an individual's motivation toward exercise.Its 19 items use a rating scale (0 = not true for me; 4 = very true for me) for 5 factors of motivation: amotivation, external regulation, introjected regulation, identifi ed regulation, and intrinsic motivation.In scoring the BREQ-2 16,31 , the total mean of the 5 subscales is calculated using a 5-point motivation subscale separating each type of motivation.The Relative Autonomy Index (RAI) was used to evaluate the degree of relative autonomy given that the 5 motivation types are located along the self-determination continuum.The RAI is calculated by weighting each motivation subscale and summing the weighted scores: (amotivation multiplied by -3) + (external regulation multiplied by -2) + (introjected regulation multiplied by -1) + (identifi ed regulation multiplied by 2) + (intrinsic regulation multiplied by 3).The minimum RAI score is -24, and the maximum score is +20.Higher positive RAI scores indicate higher autonomous motivation whereas lower negative scores indicate lower autonomous motivation.The RAI is the composite score of relative autonomy.The BREQ-2 was validated on Thai adolescents.The reliability of each factor was 0.57, 0.35, 0.70, 0.55, and 0.83, respectively, and overall it was 0.77; details have been reported elsewhere 32 .J. Nepal Paediatr.Soc.

3. 2
Dietary Self-Regulation (DSR) questionnaire 33 : The DSR measures the degree of a person's motivation for maintaining a healthy diet that is autonomous or self-determined.Its 15 items are measured using a 7-point rating scale (1 = not true at all; 7 = very true) for 3 factors of motivation: autonomous motivation, externally controlled motivation, and amotivation.When scoring the DSR 33 , items in each of the 3 subscales are averaged to create separate scores for autonomous motivation, controlled motivation, and amotivation.These three subscale scores can be used separately.The Relative Autonomous Motivation Index (RAMI) can be calculated by subtracting the average of the externally controlled reasons from the average of the autonomous reasons.The minimum RAMI score is 1, and the maximum score is 7. Higher positive RAMI scores indicate a higher autonomous motivation of eating.For the DSR validated on Thai adolescents, the reliability of each factor was 0.67, 0.83, and 0.81, respectively, and overall was .82;details have been reported elsewhere 32 .

-
and obesity -autonomous motivation (SDT) -self-effi cacy (SCT) -self-regulation (SCT) -observational learning (SCT) -education -Introduction to weight reduction program -Group relationship activities -Teaching by using symbolic model, students brain storming and set individual weight goal -Students were assigned for weekly weight recording, long 1 month for 6 months and offered their interesting exercise and physical activity Teaching by using question emotional arousal and slide presentation -Students were assigned for calculated their own BMI and physical activity daily self-record, long 1 month for 6 months and offered their interesting exercise and physical activity for 2 nd to 6 diet -autonomous motivation (SDT) -self-effi cacy (SCT) -self-regulation (SCT) -education -Teaching by using question emotional aro usal and slide presentation -Students were assigned for dietary consuming daily self-record, long 1 month for 6 months 40 physical activity 3 days a week -Students did exercise and physical activity by choosing from their offering 30

Fig 1 :
Fig 1: Study fl ow and Table 2.They were recruited upon individual and parental consent.The ethics of this study were approved by the Institutional Review Board of the Faculty of Medicine at Chulalongkorn University (IRB No. 253/56).The study ID of Thai Clinical Trials Registry (TCTR) was TCTR20130515001.The study began in July 2013 and fi nished in February 2014.

Table 2 :
Participants' baseline characteristics (n=304) Note.All topics taught by train teacher.Physical activity 3 days a week was on Monday, Wednesday and Friday.

Table 1 cont...Table 4 :
Scores of physical activity motivation before and after program of the intervention and the control groups (n=304) Comparison of mean changed in physical activity motivation at weeks 12 and 24 by using independent t-test.

Table 3 :
BMI change before and after program for the intervention and the control groups (n=304) Comparison of mean weight reduction at weeks 12 and 24 by using independent t-test.

Table 5 :
Scores of dietary intakes motivation before and after the program for intervention and control groups (n=304)