Teens Engaged in Exercise and Nutrition (TEEN): A RCT of a Family-Based Distance Intervention for Overweight Rural Adolescents: Teens Engaged in Exercise and Nutrition (TEEN): A RCT of a Family-Based Distance Intervention for Overweight Rural Adolescents TEEN Health Group


Slide2:


Slide3:


The Problem: The Problem Health risks as children and adults early morbidity and mortality cardiovascular risk factors link with adult obesity Over one quarter of rural children overweight/obese, compared to 10-25% in the national population Rural children w/ 54.7% increased risk of obesity when compared to urban children (McMurray, Harrell, Bangdiwala andamp; Deng, 1999) Access to medical and behavioral healthcare is limited in rural areas


A Novel Solution: A Novel Solution Efficacious treatments for childhood obesity already exist Jelalian andamp; Saelens, 1999 State-of-the-art Delivery System Web TV PDA technology with satellite links for feedback Immediate, individualized treatment— 'electronic therapist' Close the access gap for rural adolescents


Slide6: Family Web-TV


Slide7: Life Coach © TEEN Health


Why Kansas?: Why Kansas? Collaborative partners with KDHE and KSDE HIGH NEED: 15% children/adolescents overweight; 3 in 5 adults are overweight $$$ Obesity-related medical expenditures in Kansas total $657 million, 5.5 % of the state’s annual health care bill (KDHE 2004) Mandated annual BMI assessment by school 98% access to internet technology in the home (KAN-ED)


OBJECTIVE: OBJECTIVE To reduce BMI of rural overweight adolescents using a family-based, multi-component distance intervention


Research Context: Research Context Previous research has established probably efficacious interventions for children and adolescents Adaptation of Epstein’s comprehensive family-based intervention for exercise and diet in overweight and obese children tested in three small randomized trials Studies link increased fruit and vegetable consumption with decreased overall caloric intake Our pilot studies used 10 focus groups to establish the acceptability and feasibility of the current intervention


Research Context : Research Context Life Coach© Development SBIR Phase I/II studies from NHLBI conducted by TeamSix, Inc. Enhanced PDA featuring: Pedometer Ecological momentary assessment (EMA) Feedback component Satellite link Feasibility tested in populations 12 years and older Preliminary efficacy data in adults Increases Kcal expended Decreases Kcal consumed Decreases weight


Primary Question and Hypothesis: Primary Question and Hypothesis Will the TEEN technology-based weight management program be efficacious in reducing BMI in overweight rural adolescents (grades 7-10)? Hypothesis: The TEEN treatment will produce a greater reduction in BMI over time as compared to control


Secondary Questions and Hypotheses: Secondary Questions and Hypotheses Will reduction in BMI in the treatment group be mediated by changes in activity, diet, and parent involvement? Hypothesis 2A: Increase in activity level (LifeCoach) will mediate the effects of treatment on reduction in BMI. Hypothesis 2B: Increase in self-reported consumption of healthy foods (e.g., fruits and vegetables) will mediate the effects of treatment on reduction in BMI. Hypothesis 2C: Increase in parental involvement will mediate the effects of treatment on reduction in BMI.


Secondary Questions and Hypotheses: Secondary Questions and Hypotheses Will the intervention result in an improved quality of life? Hypothesis 3: Adolescents in the treatment group will show greater improvement in quality of life (PEDS-QL) as compared to control.


Subgroup Analyses: Subgroup Analyses Assess potential moderating role of age, gender, race/ethnicity and baseline BMI on treatment gains. Hypothesis 4: No differential change in BMI between groups by age, gender, race/ethnicity, and baseline BMI


OVERVIEW OF STUDY: OVERVIEW OF STUDY 2 group parallel design Clustered randomization Assessors are blind to group assignment ITT Intervention duration=6 months Study duration=18 months total


RANDOMIZATION WITHIN STRATA (MS VS. HS): RANDOMIZATION WITHIN STRATA (MS VS. HS) Unit of randomization: school 'Clustered randomized trial' Keep arms spatially separate All eligible grade-level children within school are assigned to same arm Reduces treatment contamination Equal treatment of equals within the small communities Nested data


POPULATION SIZE: POPULATION SIZE 279 rural schools (84 middle, 20 junior, 175 high school), 47.5% in rural areas 46,642 rural students in 7-10th grade 15% overweight on avg., approx. 7,000 potential population 10 of 70 students per school are overweight (2 grades sampled/school; average grade size=35 students)


POWER CALCULATIONutilizing Optimal Design software: POWER CALCULATION utilizing Optimal Design software There is a 90% chance of a statistically significant result w/ 8 (of 10) participating children/school under the following conditions (p=.05, 2-tail):


SAMPLE SIZE: SAMPLE SIZE Control/Treatment Arms 26 HS (260 students) 26 JH/MS (260 students) TOTAL 104 schools (52 per arm) 1,040 students in grades 7-10


Slide21:


Recruitment : Recruitment Through collaboration with KDHE, KDE, school districts, and all principals (see letters of agreement), contact potential families Principal’s office Mails a letter to each identified household Each mailing includes a consent form No informed consent received within 3 weeks, second mailing Follow-up recruitment calls from schools for non-response


Screening: Screening Location – schools Staff - LVNs


Slide24: Epstein’s Components Caregivers and Child: Weight control / prevention Traffic light diet with focus on increasing fruit and vegetables Developing healthy eating environment: situational / contingencies Behavior change techniques Maintenance of behavior changes Caregiver Only: Education on reinforcement of target child behaviors Intervention emphasizes simple changes: Eating more fresh fruits vegetables per day Reducing sedentary behaviors Increasing activity through low intensity 'lifestyle' changes. WebTV Family Intervention


WebTV Family Intervention: WebTV Family Intervention Typical mode of administration is through a workbook and 12 individual and family face-to-face sessions. Adapted to technology-based administration for WebTV for rural access. Adapted each lesson into an 'edutainment' based approach to modeling health behavior through 12 bi-weekly 20-minute 'shows' on the WebTV. This translates into behavioral goals of 5 fresh fruits/veg per day and 10,000 steps/day.


Life Coach© Assessment: Life Coach© Assessment Child/Parents in both groups Pedometer (Assessment) Daily upload to server via satellite Ecological Momentary Assessment (EMA) (Assessment) Randomly selects 4 days/week Activity/Exercise checklist Food intake checklist Red/Yellow/Green foods Barriers to exercise checklist Barriers to healthy food intake checklist


Life Coach© Intervention: Life Coach© Intervention Feedback Weekly/PRN Via Life Coach©, web, email, mail Summaries (actual vs. goal) Tailored suggestions for change Parent gets embedded suggestions based on child’s assessments Informational Resources Event driven/PRN Kcal for foods Kcal expenditure for certain exercise/activity Solutions for barriers Request for resources, video modules


Control Condition: Control Condition Written notification on child’s baseline BMI Description of health risks and treatment options Encouragement to see PCP/healthcare provider Get Life Coach© without feedback


Treatment Fidelity:Delivery: Treatment Fidelity: Delivery Date/Time stamp of observation of video modules (both parent and child) 'Feedback' suggestions viewed (Life Coach©, web) Life Coach© worn (andgt;100 steps/day)


Treatment Fidelity:Receipt: Treatment Fidelity: Receipt Scavenger Hunt Videos/Feedback have embedded words/objects that must be identified after video/feedback Post-video content quiz (videos)


Treatment Fidelity:Enactment: Treatment Fidelity: Enactment Daily pedometer readings Change in daily activity and food monitoring from EMA Changes in self-report assessment of activity/food measures Parent report of child’s behavior


Efforts to Increase Treatment Compliance: Efforts to Increase Treatment Compliance Remuneration for 0, 6, 12, 18 month assessments ($5.00/$10.00/$20.00/$40.00 worth of gift certificates to vendors) Keep Life Coach© upon completion Points given on scavenger hunt items Prizes/Gift certificates available based on number of points per week/month


Slide33:


Slide34: Primary Data Analysis H1: A multi-level growth curve model will be fit to compare whether average rate of change over time in BMI differs across treatment and control, and whether mean BMI differs at end of treatment and follow-up across groups


Slide35: Secondary Data Analysis H3: A multi-level growth curve model will be fit to compare whether average rate of change over time in QOL differs across treatment and control, and whether mean QOL differs at end of treatment and follow-up across groups H4 Subgroup: Test whether age, gender, race/ethnicity covariates interact with the main effect of treatment on intercept (mean BMI at time t) and BMI growth rate (average change in BMI over time). Examine the correlation between intercept (initial BMI) and slope (change in BMI).


Secondary Data Analyses (cont): Secondary Data Analyses (cont) H2a-c: Conduct separate mediational analyses (Baron andamp; Kenny, 1986) for physical activity, food selection, and parental involvement


DSMB Issues: DSMB Issues Interim analyses (2) After 333 and 666 adolescents randomized and completed/drop-out Stopping rules O’brien-Flemming for efficacy / harm Stochastic Curtailment for futility


DSMB Issues (cont): DSMB Issues (cont) SAE/AE A Serious Adverse Event (SAE) is any adverse experience occurring during the study Death, disability, serious medical illness Hospitalization In the absence of medical intervention would have led to any of the above An Adverse Event (AE) is a lesser adverse experience occurring during the study (e.g., minor injury, underweight, depression) SAE/AEs will be reported verbally and in writing to the PI. All SAEs will be reported to the IRB andamp; DSMB w/in 24 hours


Conclusion: Conclusion Limitations Costs associated with technology Long-term effects? Innovative combination of technology and family to overcome access barriers Rural setting but portable to other populations with limited access Anticipated cost effectiveness Reduced medical costs Relatively low-cost


Acknowledgements: Acknowledgements Thanks to all our mentors, consultants, and peer consultants that made TEEN possible! Thanks in advance to our future funders and reviewers!!


References: References McMurray RG, Harrell JS, Bangdiwala SI, Deng S. Cardiovascular disease risk factors and obesity of rural and urban elementary school children. Journal of Rural Health. 1999;15(4):365-374. Epstein LH, Myers MD, Raynor HA, Saelens BE. Treatment of pediatric obesity. Pediatrics. 1998;101(Supplement):554-570. Troiano, R. andamp; Flegal, K. (1998). Overweight children and adolescents: Description, epidemiology, and demographics. Pediatrics, 101, 497-504. Kromeyer, K., Zellner, K., Jaeger, U. andamp; Hoyer, H. (1999). Prevalence of overweight and obesity among school children in Jena (Germany). International Journal of Obesity and Related Metabolic Disorders, 23, 1143-50. Rios, M., Fluiters, E. andamp; Perez, M. (1999). Prevalence of childhood overweight in northwestern Spain: A comparative study of two periods with a ten year interval. International Journal of Obesity and Related Metabolic Disorders, 23, 1095-8. Strauss, R. andamp; Pollack, H. (2001). Epidemic increase in childhood overweight, 1986-1998. Journal of the American Medical Association, 286, 2845-8. DiPietro, L., Mossberg, H. andamp; Stunkard, A. (1994). A 40-year history of overweight children in Stockholm: Life-time overweight, morbidity, and mortality. International Journal of Obesity and Related Metabolic Disorders, 18, 585-90. Freedman, D., Dietz, W., Srinivasan, S. andamp; Berenson, G. (1999). The relation of overweight to cardiovascular risk factors among children and adolescents: The bogalusa heart study. Pediatrics, 103, 1175-82. Freedman, D., Kettel Khan, L., Dietz, W., Srinivasan, S. andamp; Berenson, G. (2001). Relationship of childhood obesity to coronary heart disease risk factors in adulthood: The bogalusa heart study. Pediatrics, 108, 712-8. Must, A., Jacques, P., Dallal, G., Bajema, C., Dietz, W. (1992). Long-term morbidity and mortality of overweight adolescents: A follow-up of the harvard growth study of 1922 to 1935. New England Journal of Medicine, 327, 1350-5.