Mobile Health Intervention to Identify Early Responders to Treatment in Adolescent Obesity

Pilot Evaluation of a Mobile Health Intervention to Identify Early Responders to Treatment in Adolescent Obesity as a Triage Approach

National Institute of Education, Singapore, National Dental Centre, Singapore, Duke-NUS Graduate Medical School

Background: The Expert Committee on the Assessment, Prevention and Treatment of Child and Adolescent Overweight and Obesity recommends a staged based approach to the management of adolescents with overweight and obesity from Stage 1-4 with increasing intensity of management in higher stages. Mobile health application is an attractive community based treatment for adolescent obesity due to its wide penetration and convenience. Early weight loss has been found to be the strongest predictor of good long term outcome in obesity. However there is currently no known study that use early weight loss as a predictor factor for a stepped up approach using a mobile health application. Clinical significance: The current study use a mobile health intervention to identify participants with early weight loss in a stepped up approach. Primary objective will be to examine the proportion of patients triaged to the low risk Weight Management Clinics (WMC) after brief intervention by a nurse coordination and completion of 4 sessions of Kurbo Program over a 12 month recruitment period. Secondary objectives will be to examine changes in BMI z-score, metabolic profile, examine program feasibility and fidelity and explore other predictors of poor response to program. Methodology: Children aged 13-17 years old with BMI percentile of above 90th percentile, who are referred to the WMC, will receive a brief intervention by the WMC nurse coordinator followed by introduction to Kurbo program, a multifunctional mobile application, for more detailed dietary and physical activity recommendations and implementation of behavioural changes. Patients that are able to engage with Kurbo intervention and showed a decrease in BMI percentile over 4 sessions of Kurbo will be offered the low risk weight management clinic. At baseline, month 3 and month 6, the patient's weight and height, body fat composition, waist circumference and blood pressure will be measured as per usual standard protocol. Questionnaires to assess eating, quality of life and dietary recall will be administered as part of the research. Accelerometers will also be fitted to assess physical activity. At baseline and month 6, metabolic blood tests (HbA1C, fasting lipid panel, oral glucose tolerance test, fasting insulin level and liver function test) were collected after a minimum 8 hour fasting period together with bloods for aromatic amino acid, branch chain amino acid and long chain acylcarnitines . Current low risk WMC patients will be offered 2 monthly follow up with optional dietician and exercise physiologists counselling and exercise sessions. The high risk WMC patients will be routinely offered the standard high risk follow up protocol consisting of weekly follow up with the multidisciplinary team for 4 weeks followed by 2 weekly appointments for 2 months and monthly appointment thereafter based on clinical response.

Participants will undergo Kurbo program, a Mobile health application, for more detailed dietary and physical activity recommendations and implementation of behavioural changes. The patient's progress will be reviewed by the nurse clinician at one month post intervention to determine whether the BMI percentile has shown a reduction through the Kurbo Program. Patients that declined Kurbo intervention, has a BMI of more than 99th percentile or continue to have increase in their BMI percentile in Kurbo program, will be offered the high risk weight management clinic appointment for a more detailed multidisciplinary evaluation for targeted intervention. Patients that are able to engage with Kurbo intervention and showed a decrease in BMI percentile over 4 sessions of Kurbo will be offered the low risk weight management clinic (WMC). There will be a month 3 and month 6 visit for study measurements in this study.

The Kurbo program is a multifunctional mobile application that aid adolescents and their families to learn healthy eating habits and weight management through the use of a mobile application with dietary self-monitoring and weekly interactive coaching sessions. Using the Kurbo app, adolescents track their food and exercise, as well as learn about healthy behaviours through games and videos. The Kurbo coaches check in with adolescents for 15 minutes once a week via video, phone or text over a 12 weeks period.

Examine the proportion of patients triaged to the low risk weight management clinics after brief intervention by a nurse coordination and completion of 4 sessions of Kurbo Program.

Kurbo Program Feasibility, Percentage of Participants Who Completed at Least One Health Coaching Session

To examine the feasibility of using Kurbo program as a waitlist intervention for the target population as measured by quantifying the percentage of patients offered the program who agreed to enrol and engaged in Kurbo coaching session

To examine program fidelity as measured by the percentage of patients who complete all of the 12 sessions of online coaching by Kurbo coaches, as part of Kurbo programme.

To examine the rate of attrition from weight management clinic after implementation of the waitlist intervention and new model of care in weight management clinic.

To examine changes in treatment outcomes as a function of program fidelity as measured by those who complete more online coaching show greater improvements in BMI z-score. Body mass index (BMI) will be calculated as kg/m2 and BMI-z score calculated using the L, M, S parameters published by the Centre for Disease Control and Prevention. BMI z-score are measures of relative weight adjusted for child age and sex. Reduction of more than 0.25 in childhood obesity has been found to be clinically significant for reduction of cardiovascular risk factors.

To examine changes in treatment outcomes as a function of program fidelity as measured by those who complete more online coaching show greater improvements in BMI z-score. Body mass index (BMI) will be calculated as kg/m2 and BMI-z score calculated using the L, M, S parameters published by the Centre for Disease Control and Prevention. BMI z-score are measures of relative weight adjusted for child age and sex. Reduction of more than 0.25 in childhood obesity has been found to be clinically significant for reduction of cardiovascular risk factors.

To examine the effect of the waitlist intervention and WMC intervention on waist to height ratio at baseline and 6 months. Waist and height will be measured in cm and ratio of more than 0.5 is indicative of higher cardiometabolic risk. Reduction of the waist height ratio has been shown to be beneficial in cardiometabolic outcomes.

To examine the effect of the waitlist intervention and WMC intervention on waist to height ratio at baseline and 3 months. Waist and height will be measured in cm and ratio of more than 0.5 is indicative of higher cardiometabolic risk. Reduction of the waist height ratio has been shown to be beneficial in cardiometabolic outcomes.

Change in body fat percentage from Baseline to 6 months. Body fat percentage was assessed using a bioimpedance analysis.

Change in body fat percentage from Baseline to 3 months. Body fat percentage was assessed using a bioimpedance analysis.

To examine the effects of waitlist intervention on nutrition using a three day food diary at baseline and 6 months to assess total caloric intake and number of servings of fruits and vegetables.

To examine the effects of waitlist intervention on nutrition using a three day food diary at baseline and 3 months to assess total caloric intake and number of servings of fruits and vegetables.

To examine the effects of waitlist intervention on nutrition using a three day food diary at baseline and 6 months to assess number of servings of vegetables per day.

To examine the effects of waitlist intervention on nutrition using a three day food diary at baseline and 3 months to assess number of servings of vegetables per day.

To examine the effects of waitlist intervention on physical activity using results from accelerometer to assess time spent on sedentary and moderate to vigorous physical activity at baseline and 6 months. The Actigraph data were processed using the Actilife 6 software. The Puyau cut-off point of 3200 counts per minute (cpm) was used to estimate time spent in moderate-to-vigorous physical activity (MVPA).

To examine the effects of waitlist intervention on physical activity using results from accelerometer to assess time spent on sedentary and moderate to vigorous physical activity at baseline and 3 months. The Actigraph data were processed using the Actilife 6 software. The Puyau cut-off point of 3200 counts per minute (cpm) was used to estimate time spent in moderate-to-vigorous physical activity (MVPA).

To examine the effects of waitlist intervention on quality of life at baseline and 6 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4) will be administered as a comprehensive and multi-dimensional construct that includes physical, emotional, and social functioning to assess quality of life in the adolescents. The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life at baseline and 3 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4) will be administered as a comprehensive and multi-dimensional construct that includes physical, emotional, and social functioning to assess quality of life in the adolescents. The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, physical domain, at baseline and 6 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, physical domain, at baseline and 3 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, emotional domain, at baseline and 6 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, emotional domain, at baseline and 3 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, school domain, at baseline and 6 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, school domain, at baseline and 3 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, psychosocial domain, at baseline and 6 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

To examine the effects of waitlist intervention on quality of life, psychosocial domain, at baseline and 3 months using Pediatric Quality of Life Inventory (PedsQL; UK version 4). The self-report scale had a five-point response scale (0 = never a problem, 1 = almost never, 2 = sometimes, 3 = often and 4 = almost always). Items are linearly transformed to a 0-100 scale, so that higher scores indicate better quality of life.

Examine changes in insulin resistance index at baseline and 6 months. Insulin resistance index is calculated according to the formula: fasting insulin (microU/L) x fasting glucose (nmol/L)/22.5

Examine changes in 120 minute glucose result changes in oral glucose tolerance test at baseline and month 6 measured in nmol/L.

Examine changes in fasting lipids, high density lipoprotein cholesterol (HDL), at month 6 measured in mg/dL

Examine changes in fasting lipids, low density lipoprotein cholesterol (LDL), at month 6 measured in mg/dL

To examine the effects of waitlist intervention on disordered eating at baseline and 6 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on disordered eating at baseline and 3 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on External Eating dimension on disordered eating at baseline and 6 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on External Eating dimension on disordered eating at baseline and 3 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on Parental Pressure to Eat dimension on disordered eating at baseline and 6 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on Parental Pressure to Eat dimension on disordered eating at baseline and 3 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on Emotional Eating dimension on disordered eating at baseline and 6 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

To examine the effects of waitlist intervention on Emotional Eating dimension on disordered eating at baseline and 3 months. Psychological dimensions of eating behaviors were determined using a validated self-reporting Eating Pattern Inventory for Children (EPI-C). The 20-item questionnaire assessed four dimensions ((dietary restraint, external eating, parental pressure to eat, and emotional eating). Responses to each item were listed on a 4-point Likert scale (1 = not at all, 2 = sometimes, 3 = mostly, 4 = always). Higher scores are indicative of greater dietary restraint.

Examine changes in branched chain amino acid(BCAA) at baseline and month 6. Metabolites to be analysed include BCAA which is a combination of isoleucine, leucine, valine. All metabolites concentrations will be reported in micro mol per litre

Examine changes in aromatic amino acid (AAA) at baseline and month 6. Metabolites to be analysed include AAA which is a combination of phenylalanine and tyrosine. All metabolites concentrations will be reported in micro mol per litre.

Inclusion Criteria: Overweight as defined by BMI percentile of above 90th percentile Age 13-17 years old in the year of referral Ability to provide informed consent Adolescents with a phone which is able to download the application to be used on their devices Exclusion Criteria: Patients with secondary causes of obesity especially genetic syndromes e.g. Trisomy 21, Prader-Willi Currently participating in a weight management program Unable to understand and speak English sufficiently to give informed consent and complete the research assessments.