Novel Approach to Stimulant Induced Weight Suppression and Its Impact on Growth

The primary endpoint will be change in z-height at month 30 which is study endpoint. Measured as a zscore with more negative units reflecting smaller incremental height gain. Z units used to account for differences between groups in gender and age with both impact height at a fixed time.

Time frame: month 30 or last assessment point

Population: includes all with 2+ growth assessments from the behavior therapy and ER stimulant arms. Doesn't include adaptive randomization arms (drug holiday, cal supplement, monitoring) as they didn't exist until 2nd randomization. See outcome #10 for change in zht from beginning to end of second randomization.

sum of score on 10 item IOWA Conners with range from 0 to 30 and higher values indicating more symptoms. Collected at end point or last assessment point.

Time frame: at month 30 or last collected assessment point

Population: those assigned to either Behavior therapy or ER stimulant with at least one post baseline assessment of ADHD symptoms. Second randomization arms (drug holiday, cal supplement, monitoring) not included as only relevant outcomes are ht, wt and BMI. All subjects in these arms are either in behavior therapy or er stimulant arm from 1st randomization.

sum of items on 10 item IOWA Conners with range from 0-30 and larger values indicating greater symptoms. Collected at endpoint or last assessment point.

Time frame: month 30 or last assessment point

Population: those assigned to either Behavior therapy or ER stimulant with at least one post baseline assessment of ADHD symptoms. Second randomization arms (drug holiday, cal supplement, monitoring) not included as only relevant outcomes are ht, wt and BMI. All subjects in these arms are either in behavior therapy or er stimulant arm from 1st randomization.

difference in height z score from entry into weight recovery phase to exit from weight recovery phase (exact duration varied by participant). Randomization could not occur before month 6 (equaling a 24 month duration) but could start as late as month 29 (equaling a 1 month duration) of treatment based on the pattern of zBMI change by the individual participant. Z scores used to account for differences in age and gender. More negative values reflecting less incremental height gain.

Time frame: between 1 month and 24 months

Population: all participants prescribed an ER stimulant who were also went through the second randomization to one of three weight recovery interventions. One monitoring participant never prescribed med was excluded.First randomization arms not included as not all of those participants had a second randomization as it was adaptively based on change in zBMI.

difference in weight z score from entry into weight recovery phase to exit from weight recovery phase (exact duration varied by participant). Randomization could not occur before month 6 (equaling a 24 month duration) but could start as late as month 29 (equaling a 1 month duration) of treatment based on the pattern of zBMI change by the individual participant. Z scores used to account for differences in age and gender. Larger values reflect a greater incremental weight gain.

Time frame: 1 to 24 months duration

Population: all participants prescribed an ER stimulant who were also went through the second randomization to one of three weight recovery interventions.First randomization arms not included as not all of those participants had a second randomization as it was adaptively based on change in zBMI.

BMI will be calculated at endpoint (month 30). Difference between baseline and endpoint (month 30 or last assessment point if did not finish study). Measured as a zscore with more negative units reflecting less BMI gain. Z units used to account for differences between groups in gender and age with both impact BMI at a fixed time.

Time frame: baseline to month 30 or last assessment point

Population: includes all with 2+ growth assessments from the behavior therapy and ER stimulant arms. Doesn't include adaptive randomization arms (drug holiday, caloric supplement, monitoring) as they did not exist until 2nd randomization. See outcome #12 for change in zBMI from beginning to end of second randomization.

difference in BMI z score from entry into weight recovery phase to exit from weight recovery phase (exact duration varied by participant). Randomization could not occur before month 6 (equaling a 24 month duration) but could start as late as month 29 (equaling a 1 month duration) of treatment based on the pattern of zBMI change by the individual participant. Z scores used to account for differences in age and gender. Larger values reflecting a greater incremental BMI gain.

Time frame: between 1 month and 24 months

Population: all participants prescribed an ER stimulant who were also went through the second randomization to one of three weight recovery interventions. First randomization arms not included as not all of those participants had a second randomization as it was adaptively based on change in zBMI.

in addition to the primary outcome of height at month 30, change in z-height from baseline to study month 6 post is also reported. Subjects who were still moderately impaired after 6 months in their initial treatment arm were allowed to cross over and receive the treatments in the other arm so prior to month 6 no participants randomized to behavior arm were prescribed study medication. This outcome includes all participants with 2+ growth assessments from the behavior therapy and ER stimulant arms. Doesn't include adaptive randomization arms (drug holiday, cal supplement, monitoring) as they didn't exist until 2nd randomization which did not occur until after this assessment period was over. Height converted to z score to account for differences in age and gender. More negative values reflecting smaller incremental height gain. If participant dropped out prior to month 6, then the last assessment point was used.

Time frame: baseline to month 6

Population: participants with at height measurement at month 6

difference between baseline and endpoint (month 30 or last assessment point if did not finish study). Measured as a zscore with more negative units reflecting lesser weight gain. Z units used to account for differences between groups in gender and age with both impact weight at a fixed time.

Time frame: baseline to month 30 or to last assessment point

Population: includes all with 2+ growth assessments from the behavior therapy and ER stimulant arms. Doesn't include adaptive randomization arms (drug holiday, caloric supplement, monitoring) as they did not exist until 2nd randomization. See outcome #11 for change in zwt from beginning to end of second randomization.

% of study days that study ADHD medication was taken when prescribed to be taken; behavior group could be prescribed medication if moderately impaired still after month 6. Once prescribed, all medication was prescribed to be taken 7 days a week except for in the drug holiday weight recovery arm.

Time frame: denominator is number of days in study for which study med was prescribed

Population: any participants with at least one dose of med prescribed. The second randomization arms are not included as all participants in those arms are derived from these two groups and this assessment period includes the entire duration of the second randomization. Also, the second randomization addressees weight gain, not ADHD treatment.

Raw number of behavior therapy sessions attended; participants could cross over to other treatment arm if moderately impaired after 6 months in initial randomly assigned arm

Time frame: months 0 through 30

Population: those with at least one follow up assessment. The second randomization arms are not included as all participants in those arms are derived from these two groups and this assessment period includes the entire duration of the second randomization. Also, the second randomization addressees weight gain, not ADHD treatment.

percent of days caloric supplement were taken versus prescribed in caloric supplement arm

Time frame: from entry to exit of caloric supplement arm

Population: those assigned to caloric supplement group

measures change in BMI z score from baseline to last assessment with participants grouped based on actual medication usage versus randomly assigned group since participants were allowed to cross treatment arms after 6 months and not all participants assigned to medication used it consistently. The rarely med group (n=44) used med \<12.5% of the study duration (with most using not at all). The consistent med group (N=38 used med for at least 87.5% of their time in the study with most using the entire time). The inconsistent med group (N=111, 27.5% used medication 45% of the time in the study. The other 37 participants did not have one year of growth data so were excluded from this analysis. Z scores used to account for differences in age and gender between groups. Higher values represent a larger BMI

Time frame: baseline to month 30 or last assessment point

Population: Includes all participants with at least one year of growth data as goal was to assess impact of extended treatment on growth. These same outcomes are reported elsewhere for the first randomization arms of Behavior Therapy and Med as well as the second randomization arms of cal supplement, drug holiday and monitoring (see outcomes 3 and 12).

difference in height z score from entry into weight recovery phase to exit from that phase (exact duration varied by participant). Randomization could not occur before month 6 (so max of 24 month duration) but could start as late as month 29 (equaling a 1 month duration) based on the pattern of zBMI change. In this post hoc analysis we grouped participants by what they did (caloric supplementation, drug holiday or monitoring) not what they were randomly assigned to. The most common change was from drug holiday to monitoring for participants who were not using medication on weekends before assignment to drug holiday (family stopped weekend med by own accord prior to 2nd randomization) so assignment to drug holiday did not alter actual frequency of use as was designed to.Therefore they were reclassified as monitoring as frequency of med use did not change. Z score used to account for differences in age and gender between groups. Higher values reflect greater incremental BMI increase.

Time frame: between 1 month and 24 months

Population: all participants prescribed an ER stimulant and also assigned to one of the weight recovery treatments. Arms from first randomization are not included as not all of those participants progressed to the second randomization which was done adaptively based on change in zBMI. Those arms are reported on for outcome 15.

measures change in height z score from baseline to last assessment with participants grouped based on actual medication usage versus randomly assigned group since participants were allowed to cross treatment arms after 6 months and not all participants assigned to medication used it consistently. The rarely med group (n=44) used med \<12.5% of the study duration (with most using not at all). The consistent med group (N=38 used) med for at least 87.5% of their time in the study with most using the entire time. The inconsistent med group (N=111), used medication between 12.5 to 87.5 of the time (mean time on med was 45% of the time in the study) The other 37 participants did not have one year of growth data so were excluded from this analysis. Z scores used to account for differences in age and gender between groups. More negative values reflecting a smaller incremental height gain.

Time frame: baseline to month 30 or last assessment point

Population: Includes all participants with at least one year of growth data as goal was to assess impact of extended treatment on growth. These same outcomes are reported elsewhere for the first randomization arms of Behavior Therapy and Med as well as the second randomization arms of cal supplement, drug holiday and monitoring (see outcomes 1 and 10).

measures change in weight z score from baseline to last assessment with participants grouped based on actual medication usage versus randomly assigned group since participants were allowed to cross treatment arms after 6 months and not all participants assigned to medication used it consistently. The rarely med group (n=44) used med \<12.5% of the study duration (with most using not at all). The consistent med group (N=38 used med for at least 87.5% of their time in the study with most using the entire time). The inconsistent med group (N=111, 27.5% used medication 45% of the time in the study. The other 37 participants did not have one year of growth data so were excluded from this analysis. Z scores used to account for differences in age and gender between groups. Higher values represent a greater incremental weight gain.

Time frame: baseline to month 30 or last assessment point

Population: Includes all participants with at least one year of growth data as goal was to assess impact of extended treatment on growth. These same outcomes are reported elsewhere for the first randomization arms of Behavior Therapy and Med as well as the second randomization arms of cal supplement, drug holiday and monitoring (see outcomes 2 and 11).

difference in height z score from entry into weight recovery phase to exit from that phase (exact duration varied by participant). Randomization could not occur before month 6 (so max of 24 month duration) but could start as late as month 29 (equaling a 1 month duration) based on the pattern of zBMI change. In this post hoc analysis we grouped participants by what they did (caloric supplementation, drug holiday or monitoring) not what they were randomly assigned to. The most common change was from drug holiday to monitoring for participants who were not using medication on weekends before assignment to drug holiday (family stopped weekend med by own accord prior to 2nd randomization) so assignment to drug holiday did not alter actual frequency of use as was designed to.Therefore they were reclassified as monitoring as frequency of med use did not change. Z score used to account for differences in age and gender between groups. Higher values reflect greater incremental weight gain.

Time frame: between 1 month and 24 months

Population: participants using an ER stimualnt and also randomized to one of the weight recovery treatments. Arms from first randomization are not included as not all of those participants progressed to the second randomization which was done adaptively based on change in zBMI. Those arms are reported on for outcome 14.

difference in height z score from entry into weight recovery phase to exit from that phase (exact duration varied by participant). Randomization could not occur before month 6 (so max of 24 month duration) but could start as late as month 29 (equaling a 1 month duration) based on the pattern of zBMI change. In this post hoc analysis we grouped participants by what they did (caloric supplementation, drug holiday or monitoring) not what they were randomly assigned to. The most common change was from drug holiday to monitoring for participants who were not using medication on weekends before assignment to drug holiday (family stopped weekend med by own accord prior to 2nd randomization) so assignment to drug holiday did not alter actual frequency of use as was designed to.Therefore they were reclassified as monitoring as frequency of med use did not change. Z score used to account for differences in age and gender between groups. Larger values reflect greater height change.

Time frame: between 1 month and 24 months

Population: all participants using an ER stimulant and were also assigned to a weight recovery arm. Arms from first randomization are not included as not all of those participants progressed to the second randomization which was done adaptively based on change in zBMI. Those arms are reported on for outcome 13.