SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Objective: To evaluate demographic and psychosocial predictors of attrition and weight loss in a behaviorally based adolescent weight control trial.

Methods and Procedures: Adolescents (N = 76) aged 13–16 years and 20–80% overweight (M = 60.56%, s.d. = 15.17%) received standard group-based behavioral treatment as part of a randomized trial comparing different activity interventions for overweight adolescents. Anthropometric and psychosocial measures were obtained at baseline and after the 16-week intervention.

Results: Higher parent (P < 0.01) and adolescent BMI (P < 0.05) at baseline, as well as ethnic minority status (P < 0.05) were significantly associated with attrition in univariate analyses. Parent BMI remained the only significant predictor of attrition in multivariate analyses. BMI change for completers (N = 62) was highly variable, ranging from −6.09 to +1.62 BMI units. Male gender (P < 0.01) was a significant predictor of reduction in BMI, whereas not being from an ethnic minority group (P < 0.05) and attendance at group sessions (P = 0.05) were associated with ≥5% absolute weight loss in multivariate analyses. Absolute weight loss during the first 4 weeks of the program was strongly associated with weight loss (pr = 0.44, P < 0.001) during the remainder of the intervention. Psychosocial variables were unrelated to attrition or treatment outcome.

Discussion: These findings highlight the potential importance of attending to parental BMI in efforts to retain adolescent participants in treatment, as well as the need to develop weight control interventions that are more effective for ethnic minority youth.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

The past two decades have seen an unprecedented increase in the prevalence of pediatric overweight (1,2). Recent survey data for the United States indicate that ∼31% of all children and adolescents between 6 and 19 years of age are either at risk of overweight (BMI > 85th% and <95th%) or overweight (BMI ≥ 95th%), (1). Family-based behavioral weight control programs have demonstrated efficacy in the treatment of overweight children between 8 and 12 years of age, with some promise regarding long-term outcomes (3,4); however, findings with adolescents are less consistent.

Reviews of randomized treatment studies conducted with adolescents in outpatient settings indicate that weight losses typically range from 1 to 4 kg (5,6,7,8). However, some studies have produced much larger weight losses (9,10). Significant variability has also been observed within studies. For example, adolescents in a family-based behavioral treatment combined with placebo demonstrated a mean weight loss of 3.2 kg, but the associated s.d. was 6.1 kg (11). Likewise, in an adolescent trial of two different dietary interventions, weight losses of 9.9 ± 9.3 kg were reported for adolescents prescribed a low-carbohydrate diet and 4.1 ± 4.9 kg for adolescents prescribed a low-fat diet (12). These findings suggest that there is significant variability in weight loss for adolescents receiving the same standardized intervention.

A second dimension of importance in weight control interventions with pediatric populations is attrition. Identifying variables associated with attrition has direct implications for enhancing retention efforts for pediatric weight control programs. Findings suggest retention with pediatric samples may be more challenging than with adults. For example, Zeller and colleagues (13) reported a completion rate of 45% for a clinic-based interdisciplinary weight management program. A variety of predictors of attrition and treatment responses have been identified in standardized weight control trials with adults (14), including initial weight, previous attempts to diet, binge eating, and depression, with inconsistent results (15). Dimensions with demonstrated association to weight loss success in adults are ethnicity (16), initial body weight (17), psychological variables such as higher self-efficacy (14,18), and treatment-related variables such as attendance and initial weight loss (17).

In contrast, very few studies conducted with pediatric samples have evaluated predictors of attrition or response to weight control intervention. Receipt of Medicaid, child age, and black race were associated with attrition from a clinic-based pediatric weight management program (13), but were not examined with regard to weight loss. Higher child baseline BMI and degree of parental obesity were associated with less successful weight outcomes in one study of pediatric weight interventions (19). In contrast, a recent study demonstrates that decrease in parent weight as reflected by z-BMI change is predictive of child z-BMI change after a standardized weight control intervention targeting modification to both child and parents' eating and activity patterns (20). In a study of overweight adolescents (11), neither maternal baseline BMI nor maternal education was related to outcome.

With regard to psychosocial variables, lower self-concept and greater endorsement of depressive symptomatology at baseline were associated with attrition from a pediatric weight management program (13), and social support for changes in diet and physical activity has been associated with weight loss outcome. In a 10-year follow-up of children treated for obesity, retrospective report of social support for eating and exercise, including good eating encouragement from family, was a significant predictor of long-term weight loss (21). The purpose of this study was to evaluate predictors of attrition and treatment response in adolescents receiving a standard behavioral weight control intervention as part of a randomized controlled trial. Consistent with the findings of previous studies involving children and adults, we predicted that adolescent ethnic minority status, higher parental baseline BMI, lower adolescent-reported social support for diet and exercise, and lower adolescent self-concept would be predictive of both attrition and less successful outcome in an adolescent weight control study. Similar to previous research in the area (13), self-concept was defined as a multidimensional construct consisting of adolescents' evaluations of their competence across a number of domains (22). We further hypothesized that initial treatment engagement, as reflected by attendance and initial weight loss, would be associated with greater weight loss during the course of the intervention.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Participants

Participants included in this sample are 76 adolescents who were recruited for an adolescent weight control trial (23). The majority of participants were recruited from advertisements in local newspapers between January 2000 and January 2002. One hundred and forty-five adolescents between 13 and 16 years of age were evaluated for participation. Of these, 21 (16%) were ineligible because they met exclusion criteria of major psychiatric disorder during evaluation, were taking medications that might impact weight loss, had medical comorbidities that would prevent them from participating in the diet and physical activity prescription, or were already enrolled in counseling or a weight loss program. An additional 22 (17%) were not interested following evaluation. A total of 89 adolescents and their parents were randomized to one of three treatment conditions. Two of the conditions involved the same standardized group based behavioral weight control intervention, while the third involved standard care (four meetings with a nutritionist). There were a number of differences between the group interventions and standard care, including the format for treatment delivery (individual vs. group), extent of parental involvement, use of behavioral strategies, and specific dietary and physical activity prescriptions. To decrease variability attributable to treatment differences, our emphasis will be on the 76 participants who were randomized to the two behavioral treatment arms. Two of these participants were excluded from analyses; one was placed on Ritalin during the course of the intervention, which potentially impacted rate of weight loss, the second never completed baseline measures of weight and height.

Procedures

Institutional review boards at Rhode Island Hospital and The Miriam Hospital approved the protocol. Written informed consent was obtained from parents and assent was obtained from adolescents. Participants were enrolled into the study by one of two research assistants. Participants were randomly assigned to one of two group treatment conditions: cognitive behavioral treatment with peer-based adventure therapy (CBT+PEAT) or cognitive behavioral treatment with aerobic exercise (CBT+EXER). Measures of height and weight, self-concept, and social support were obtained at baseline, with weight and height repeated at completion of the 16-week intervention.

Intervention

The same standardized cognitive behavioral weight loss intervention was delivered to participants in both treatment conditions. The intervention was group-based and included 16 weekly sessions, with parents and adolescents attending separate concurrent meetings. Nutrition intervention included a balanced deficit diet (1,400–1,600 calories) based on the dietary exchange system. Physical activity prescription included gradual increase to a minimum of 30 min a day of aerobic activity for 5 days a week. Behavioral topics included self-monitoring, motivation for weight loss, goal setting, implementation of stimulus control strategies, use of parent-teen contracts to support nutrition and physical activity goals, social influences on diet and exercise, the relationship between stress and eating, and relapse prevention. Parallel meetings were conducted for parents of participating teens. Parent sessions focused on similar nutrition and behavioral content, as well as guidance regarding family level implementation of eating and activity changes. Groups differed in the weekly on-site activity sessions that accompanied the standard behavioral program. Adolescents randomized to CBT+EXER participated in supervised traditional cardiovascular physical activity sessions, while those randomized to CBT+PEAT participated in peer activities based on the principles of Outward Bound and geared to facilitate group cohesion and self-efficacy. The intervention components are described in more detail elsewhere (23).

Measures

Demographic variables. Basic demographic variables were collected from adolescents and parents, including age, gender, and race/ethnicity. Owing to the small number of participants from individual ethnic minority groups, non-Hispanic white adolescents were compared to those from ethnic minorities. Parent education and occupation were also collected to compute Hollingshead socioeconomic status (SES) (24).

Anthropometric variables. Weight and height were obtained for each adolescent and participating parent at baseline and at completion of the 16-week intervention. Weight was obtained on a balance beam scale in street clothes without shoes. Height was obtained with a stadiometer. Height and weight were used to calculate BMI (kg/m2), as well as percent overweight with reference to age and gender-specific BMI.

The Self-Perception Profile for Adolescents. Self-Perception Profile for Adolescents (SPPA) is a 45-item, self-report measure of adolescents' perceptions of competence in eight specific domains: scholastic competence, social acceptance, athletic competence, physical appearance, job competence, romantic appeal, behavioral conduct, and close friendship, as well as a scale assessing global self-worth (22). Internal consistency for the individual subscale scores range from 0.74 to 0.92 and factor analysis indicates identification of a unique factor for each of the eight subscales (22). For purposes of this study, adolescents' perceptions of physical appearance, athletic competence, and global self-worth were examined. Average subscale scores range from 1 to 4, with higher scores being associated with greater self-worth and mean values for subscales typically at 2.9 (s.d. 0.50–0.75) (22). The self perception profile has been used in previous studies of overweight children (25) as well as in examining the relationship between self-concept and BMI (26) and body image (27).

The Social Support for Specific Diet and Exercise Behavior Scale. This is a 57-item measure assessing specific support from family and friends regarding eating habits and exercise habits (28). The factors measured are friend support for eating habits (positive and negative comments), family support for eating habits (encouragement and sabotage), friend support for exercise (exercising together), and family support for exercise (participation and involvement). Sallis et al. (28) reported high test-retest reliabilities (correlations 0.55–0.86) for each of the factors, and high internal consistency of each factor (α = 0.61–0.91). As this scale was normed on college students, minor modifications in language were made to the original scale for use with the current sample of adolescents. Given that the measure was developed for an older age group, we evaluated the internal consistency of the scale in the current sample. Internal consistency coefficients were high for both family and friend support of exercise, α = 0.92 and 0.89, respectively. In contrast, the internal consistency coefficients for family (α = 0.39) and friend (α = 0.43) support of diet were low. As a result, the scales assessing social support for diet were not included in subsequent analyses.

Statistical analysis

All data analyses were conducted using the Statistical Package for the Social Sciences, Version 11.0 (SPSS 11.0). Preliminary univariate analyses were conducted evaluating adolescent and parent BMI, SES, adolescent age, race/ethnicity, baseline measures of global and physical self-concept and social support, and treatment attendance. Treatment condition was covaried in all univariate analyses. Variables showing trends for, or significant associations with, dependent variables were then entered into separate regression analyses to predict attrition from active weight loss treatment, change in adolescent BMI, and ≥5% weight loss. We elected to evaluate a measure of clinically significant weight loss (5%), in addition to BMI, because weight loss was a goal for this sample of adolescents. In each analysis, treatment condition was controlled for by entering it into the first step of the regression. Finally, a partial correlation (controlling for treatment condition) was run to determine the association between initial change in weight (over the first 4 weeks of treatment) and later weight loss change (from week 5 to 16).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Baseline characteristics

The sample was primarily non-Hispanic white (78.4%) and female (71.6%). The mean age of adolescents at baseline was 14.50 years (s.d. = 0.93). Mean BMI for adolescents was 32.51 (s.d. = 3.09) and 32.46 (s.d. = 8.46) for parents. The majority of participants were characterized as being of middle SES according to Hollingshead, ranging from 13.5 to 63.5.

Baseline levels of global self-concept ranged from 1.20 to 4.00 (M = 2.79, s.d. = 0.71) and physical self-concept ranged from 1.00 to 3.40 (M = 1.87, s.d. = 0.66). Adolescent ratings of social support for exercise were somewhat higher for family, 2.69 ± 0.92 than for friends, 1.97 ± 0.82.

Predictors of attrition

Of the 74 adolescents and their parents whose baseline measures were included, 13 (17.6%) did not complete the 4-month active treatment trial, as indicated by failure to complete the end of treatment evaluation. Univariate analyses showed that baseline parent BMI (F =10.38, P < 0.01) and teen BMI (F = 5.20, P < 0.05) each was associated with attrition at 4 months, with heavier parents and heavier teens being more likely to drop out of treatment. Furthermore, adolescent race/ethnicity (F = 5.97, P < 0.05) was associated with attrition at 4 months. Gender, SES, measures of social support, and measures of physical, athletic, and global self-concept were not associated with attrition.

To determine the relative likelihood of dropping out of treatment, adolescent race/ethnicity, and adolescent and parent BMI were entered into a logistic regression after controlling for treatment condition. As can be seen in Table 1, parent BMI remained the only significant predictor of attrition, with adolescents of heavier parents being 4.6 times more likely to drop out of treatment than adolescents whose parents weighed less.

Table 1.  Logistic regression predicting attrition from the 16-week weight loss trial (N = 74)
inline image

Predictors of weight loss

For those participants who completed the 4-month active treatment phase (N = 62), analyses were run to determine predictors of change in adolescent BMI as well as predictors of ≥5% absolute weight loss.

Variability in treatment outcome. Figure 1 highlights the large individual variability in treatment outcome for change in adolescent BMI during treatment. Change in BMI ranged from −6.09 to +1.62 with a mean change of −1.75 (1.58) BMI units.

image

Figure 1. Variability in BMI change among adolescents participating in the 16-week behavioral weight loss program.

Download figure to PowerPoint

Change in adolescent BMI. Male gender (F = 4.49, P < 0.05) was significantly associated with greater reduction in BMI. There was a trend for attendance at group treatment sessions (pr = −0.24, P = 0.06) to be associated with change in BMI over the course of the 16-week treatment, with higher attendance being associated with greater decrease in BMI. Ethnicity was not significantly associated with reduction in BMI (F = 2.67, P = 0.11). In addition, SES, dimensions of self-concept, and social support were not associated with change in adolescent BMI.

Hierarchical regression analyses were performed to examine the proportion of variance in changes in adolescents' BMI that could be explained by adolescent gender and treatment attendance after controlling for treatment condition in the first step. As can be seen in Table 2, the complete model accounted for 18% of the variance in BMI at 4 months, with both male gender and greater attendance significantly associated with greater reduction in adolescent BMI.

Table 2.  Hierarchical linear regression predicting change in adolescent BMI
inline image

Predictors of ≥ 5% weight loss. Teen race/ethnicity (F = 5.76, P < 0.05) significantly predicted ≥5% weight loss at 4 months. Furthermore, there was a trend for attendance at group treatment sessions (F = 3.91, P = 0.05) to be significantly associated with ≥5% weight loss, with better attendance related to greater weight loss. Again, SES, self-concept, and social support were not associated with ≥5% weight loss.

To determine the unique contribution of independent variables in predicting ≥5% weight loss, teen race/ethnicity, and treatment attendance were entered into a logistic regression after controlling for treatment condition (see Table 3). Adolescents who were non-Hispanic white were 9.7 times as likely as ethnic minority adolescents to achieve ≥5% weight loss.

Table 3.  Logistic regression predicting ≥5% weight loss
inline image

Initial weight loss success as a predictor of later weight loss. In addition to the use of baseline variables to predict changes in BMI across treatment, data were broken down into early (change in weight from baseline to week 4 in program) vs. later (change in weight from week 5 to the fourth month, end-of-treatment assessment) absolute weight loss. Partial correlation controlling for treatment condition showed that absolute weight loss within the first few weeks of the program significantly predicted subsequent weight loss (pr = 0.44, P < 0.001).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Few studies have examined predictors of attrition or weight loss outcome in adolescent weight control programs. We found that higher parent BMI was the only significant predictor of attrition during the 4-month program. Decreases in BMI were greater in males than in females and non-Hispanic Whites were more likely to lose ≥5% of body weight. Of particular note is the fact that both attendance at treatment sessions and weight loss in the initial weeks of treatment were highly related to overall outcome in this weight loss program.

Approximately 18% of the adolescents in this study failed to complete the initial treatment program. The fact that parental BMI was the only predictor of attrition observed in multivariate analysis points to the potential importance of parental engagement with programs involving their adolescents. This is consistent with studies conducted with school age children demonstrating the critical role of parents (29). Data regarding the role of parental involvement in adolescent weight loss efforts are inconclusive (4); however, previous research has not addressed the importance of parent involvement to attrition. One potential approach to enhance retention in adolescent weight control interventions may be to offer simultaneous weight control treatment to parents, as has been done commonly with school age children (20,21). In the current trial, parents were involved as supports to their adolescents, but not targeted for weight loss themselves. More actively engaging parents in a weight loss protocol geared toward themselves, as well as their teen, may help with improving motivation to continue program involvement.

With regard to participant characteristics, male gender was predictive of reduction in BMI. Some adult studies have found greater weight loss for men, but these tend to be accounted for by heavier weight at the outset. (17) It is possible that male gender was important to reduction in BMI in this study because of the greater likelihood of linear growth in boys of this age group. This is consistent with the finding that gender was not predictive of greater absolute weight loss, as measured by ≥5% loss, in our sample.

Adolescent ethnic minority status was related to attrition in univariate analyses and to achieving a predetermined degree of weight loss in this study. This finding is partially consistent with studies documenting that African-American participants are less likely to remain in pediatric weight management programs (13,30). In addition, consistent with adult treatment studies (16), ethnicity was associated with less likelihood of losing a preset amount of weight in this study. On average, adolescents from ethnic minorities lost ∼2.1 kg, whereas those who were non-Hispanic white lost ∼4.3 kg. These findings highlight the importance of developing interventions for adolescents from diverse ethnic and racial backgrounds. While efforts were made to individualize both dietary and physical activity prescriptions and to be responsive to existing cultural traditions and family resources—i.e., modifying family recipes to be lower in calories, developing physical activity programs that could be conducted within the home setting—additional efforts that enhance weight loss in ethnic minority participants are needed. As noted by Kumanyika and Grier (31), understanding cultural differences in parental attitudes regarding weight and recognizing environmental variables that may impact a family's ability to implement dietary and physical activity recommendations are critical.

Contrary to expectation, adolescent baseline self-concept and social support for diet and exercise were not related to either attrition or weight loss in this adolescent weight control trial. This is consonant with previous reports indicating lack of a consistent relationship between improvement in self-concept and weight loss in pediatric weight control trials (32). One potential positive implication is that baseline level of self-concept may not be a barrier to participation in an adolescent weight control program.

Consistent with studies conducted with adults (17), the treatment related variables of attendance and initial weight loss were related to some measures of weight loss. It is of interest to note that the same association holds for adolescent weight loss efforts. There are a number of mechanisms through which attendance may positively impact reduction in BMI. One possibility is that adolescents who attend more sessions get a greater “dose” of the intervention, leading to better outcome. Alternatively, attendance may serve as a proxy for a global construct such as investment in the treatment program.

Weight loss during the initial 4 weeks of the intervention was also highly related to weight loss during the subsequent weeks. Initial weight loss may serve as a reinforcer, providing motivation for adolescents to continue with changes made to diet and physical activity. To the extent that adolescents' initial efforts meet with success, they may be more inclined to continue. Alternatively this variable, like attendance, may be a behavioral indicator of those individuals who are most motivated to lose weight. One direct implication for treatment is the importance of enhancing initial weight loss to support better overall outcome. Furthermore, it may be important to provide additional encouragement to adolescents who struggle at the beginning of the intervention to help maintain participation in program components.

This study should be considered in light of a number of limitations. While the percentage of enrolled adolescents who completed treatment was acceptable, the attrition rate from initial response to final recruitment was fairly high, potentially biasing the sample. Measures of adherence to diet and physical activity prescription, as well as self-monitoring were not included, making it difficult to explore the mechanisms through which variables such as attendance and initial weight loss may have impacted reduction in BMI. The scope of psychological evaluation targeting adolescents was limited, and did not include measures of depressive symptomatology, locus of control, and self-efficacy, which may have been related to both attrition and weight outcome. In addition, owing to the limited heterogeneity of the sample, we were only able to evaluate race/ethnicity at the most global level. Hence, the findings must be interpreted with considerable caution. Overall, the sample size was small and lack of adequate statistical power may have impacted study findings.

In conclusion, our findings point to the need to attend to the role of parents in adolescent weight loss as well as the potential importance of addressing the specific needs of ethnic minority populations when designing and implementing adolescent weight loss treatments. Given the limited sample size, global definition of ethnic minority status, and relatively narrow scope of psychological evaluation, a number of directions are suggested for future research. While the current findings suggest the potential importance of minority status in impacting both attrition and weight outcomes, the global definitions employed preclude any definitive conclusions. Of most importance is the need for studies with larger sample sizes to explore adequately the efficacy of weight control interventions with Latino and African-American youth. This research will be critical in determining the need for alternative interventions for adolescents from diverse ethnic backgrounds. Future studies should also include expanded assessment of both adolescent and parent psychosocial functioning to better understand the importance of a broad spectrum of dimensions to both attrition and weight loss. While one study did not demonstrate a relationship between maternal distress and attrition from a weight control program (13), higher maternal distress has been associated with more negative weight outcomes for school age children (33). Recent research also demonstrates a relationship between parental distress and quality of life in overweight children (34). These studies point to the importance of psychological distress in parents of overweight children and suggest that identifying the relevance of these for attrition and adolescent weight outcomes is an area of high priority.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

This study was supported by grant R01 HL65132 from the National Institutes of Health/National Heart, Lung, and Blood Institute (to E.J.).

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES
  • 1
    Hedley AA, Ogden CD, Johnson CL et al. Prevalence and trends in overweight among US children and adolescents, 1999–2000. JAMA 2004; 288: 17281732.
  • 2
    Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999–2000. JAMA 2002; 288: 17281731.
  • 3
    Epstein LH. Development of evidence-based treatments for pediatric obesity. In: Kazdin AE, Weisz JR (eds). Evidence-Based Psychotherapies for Children and Adolescents. The Guilford Press: New York, NY, 2003, pp 374388.
  • 4
    Jelalian E, Saelens BE. Empirically supported treatments in pediatric psychology: pediatric obesity. J Pediatr Psychol 1999; 24: 223248.
  • 5
    Emes C, Velde B, Moreau M, Murdoch DD, Trussell R. An activity based weight control program. Adapt Phys Activ Q 1990; 7: 314324.
  • 6
    Ikeda JP, Fujii M, Fong KA, Hanson M. Two approaches to adolescent weight loss. J Nutr Educ 1982; 14: 9092.
  • 7
    Mellin LM, Slinkard LA, Irwin CE Jr. Adolescent obesity intervention: validation of the SHAPEDOWN program. J Am Diet Assoc 1987; 87: 333338.
  • 8
    Wadden TA, Stunkard AJ, Rich L et al. Obesity in black adolescent girls: a controlled clinical trial of treatment by diet, behavior modification, and parental support. Pediatrics 1990; 85: 345352.
  • 9
    Coates TJ, Jeffery RW, Slinkard LA, Killen JD, Danaher BG. Frequency of contact and monetary reward in weight loss, lipid change, and blood pressure reduction with adolescents. Behav Ther 1982; 13: 175185.
  • 10
    Brownell KD, Kelman JH, Stunkard AJ. Treatment of obese children with and without their mothers: changes in weight and blood pressure. Pediatrics 1983; 71: 515523.
  • 11
    Berkowitz RI, Wadden TA, Tershakovec AM, Conquist JL. Behavior therapy and sibutramine for the treatment of adolescent obesity: a randomized controlled trial. JAMA 2003; 298: 18051812.
  • 12
    Sonkike SB, Copperman N, Jacobson MS. Effects of low-carbohydrate diet on weight loss and cardiovascular risk factor in overweight adolescents. J Pediatr 2003; 142: 225227.
  • 13
    Zeller M, Kirk S, Claytor R et al. Predictors of attrition from a pediatric weight management program. J Pediatr 2004; 144: 466470.
  • 14
    Wadden TA, Letizia KA. Predictors of attrition and weight loss in patients treated by moderate and severe caloric restriction. In: Wadden TA, Van Itallie TB (eds). Treatment of the Seriously Obese Patient. The Guilford Press: New York, NY, 1992, pp 383410.
  • 15
    Teixeira PJ, Going SB, Houtkooper LB et al. Pretreatment predictors of attrition and successful weight management in women. Int J Obes 2004; 28: 11241133.
  • 16
    Kumanyika SK, Espelan MA, Bahnson JL et al. Ethnic comparison of weight loss in the trial of nonpharmacologic interventions in the elderly. Obes Res 2002; 10: 96106.
  • 17
    Wing RR, Phelan S. Behavioral treatment of obesity: strategies to improve outcome and predictors of success. In: Eckel RH (ed). Obesity: Mechanisms and Clinical Management. Lippincott, Williams, & Wilkins: Philadelphia, PA, 2003, pp 415435.
  • 18
    Wamsteker EW, Geenen R, Iestra J et al. Obesity-related beliefs predict weight loss after an 8-week low-calorie diet. J Am Diet Assoc 2005; 105: 441444.
  • 19
    Eliakim A, Friedland O, Kowen G, Wolach B, Nemet D. Parental obesity and higher pre-intervention BMI reduce the likelihood of a multidisciplinary childhood obesity program to succeed—a clinical observation. J Pediatr Endocrinol Metab 2004; 17: 10551061.
  • 20
    Wrotniak BH, Epstein LH, Paluch RA, Roemmich JN. Parent weight change as a predictor of child weight change in family-based behavioral obesity treatment. Arch Pediatr Adolesc Med 2004; 158: 342347.
  • 21
    Epstein LH, Valoski A, Wing RR, McCurley J. Ten-year outcomes of behavioral family-based treatment for childhood obesity. Health Psychol 1994; 13: 373383.
  • 22
    Harter S. Manual for the Self-Perception Profile for Adolescents. University of Denver Department of Psychology: Denver, CO, 1988.
  • 23
    Jelalian E, Mehlenbeck R, Richardson E, Birmaher V, Wing RR. “Adventure therapy” combined with cognitive behavioral treatment for overweight adolescents. Int J Obes (Lond) 2006; 30: 3139.
  • 24
    Hollingshead AB. Two Factor Index of Social Status. Yale University: New Haven, CT, 1975.
  • 25
    Franklin J, Denyer G, Steinbeck KS, Caterson ID, Hill AJ. Obesity and risk of low self-esteem: a statewide survey of Australian children. Pediatrics 2006; 118: 24812487.
  • 26
    Biro FM, Striegel-Moore RH, Franko DL, Padgett J, Bean JA. Self-esteem in adolescent females. J Adolesc Health 2006; 39: 501507.
  • 27
    Canpolat BI, Orsel S, Akdemir A, Ozbay MH. The relationship between dieting and body image, body ideal, self-perception, and body mass index in Turkish adolescents. Int J Eat Disord 2005; 37: 150155.
  • 28
    Sallis JF, Grossman RM, Pinski RB, Patterson TL, Nader PR. The development of scales to measure social support for diet and exercise behaviors. Prev Med 1987; 16: 737750.
  • 29
    Golan M, Crow S. Parents are key players in the prevention and treatment of weight-related problems. Nutr Rev 2004; 62: 3950.
  • 30
    Tershakovec AM, Kuppler K. Race/ethnicity/ethnicity, insurance type, and follow-up in a pediatric weight management program. Obes Res 2003; 11: 1720.
  • 31
    Kumanyika SK, Grier S. Targeting interventions for ethnic minority and low-income populations. Future Child 2006; 16: 187207.
  • 32
    Bosch J, Stradmeijer M, Seidell J. Psychosocial characteristics of obese children/youngsters and their families: implications for preventive and curative interventions. Paient Educ Couns 2004; 55: 353362.
  • 33
    Epstein LH, Wisniewski L, Weng R. (1994c). Child and parent psychological problems influence child weight control. Obes Res 1994; 2: 509515.
  • 34
    Janicke DM, Marciel KK, Ingerski LM et al. Impact of psychosocial factors on quality of life in overweight youth. Obesity (Silver Spring) 2007; 15: 17991807.