Design and Methods of a Randomized Web-based Physical Activity Intervention among Children with Cancer: A Report from the Children’s Oncology Group

Abstract

BACKGROUND:

Promoting physical activity soon after treatment for childhood cancer may benefit health because sedentary lifestyle during curative therapy may perpetuate physical and emotional complications. The primary goals of this study are to evaluate the effects of a 6-month web-based, rewards-based physical activity intervention on fitness, biomarkers of cardiometabolic health, inflammation, adipokine status, quality of life and school attendance, and determine if effect of intervention on markers of cardiometabolic health is mediated by changes in fitness. The primary outcome of interest is fitness (physiological cost index, six-minute walk test) measured at end of intervention.

METHODS:

This ongoing study is a two-arm, prospective, randomized design with accrual goals of 192 children for intervention and control groups. Children ≥8 years and <16 years of age, not meeting recommended levels of physical activity, who completed therapy within the past 12 months are eligible. Both groups receive: 1) educational materials encouraging physical activity, 2) activity monitor, 3) access to web-based interface designed to motivate physical activity, 4) rewards based on physical activity levels, and 5) access to their activity data on the web-interface. Those randomized to intervention: 1) can view others’ activity and interact with other participants, and 2) receive rewards based on physical activity levels throughout the intervention (vs. at the end of the intervention for control group).

CONCLUSION:

Unique, scalable, and portable physical activity interventions that motivate young survivors are needed. This study will inform future web-based physical activity interventions for children with cancer by demonstrating effects of rewards and social interaction.

CLINICAL TRIALS

ClinicalTrials.gov Identifier: NCT03223753; COG Identifier: ALTE1631

INTRODUCTION

Improvements in treatment have resulted in five-year survival rates of over 84% for children diagnosed with cancer [1]. However, cancer therapy is not benign. Survivors are at increased risk for treatment-related chronic health problems, such as obesity, cardiomyopathy, peripheral neuropathy, muscle weakness and low bone mineral density [2]. While these conditions are associated with treatment, and not completely avoidable [2], condition severity and progression may be remediated by behaviors that improve fitness, including regular physical activity [3]. Elevated levels of inflammatory biomarkers, adipokines, and early evidence of vascular disease after treatment [4] and into survivorship [5, 6], are present in some children, and may be positively impacted by physical activity.

Childhood cancer survivors are more likely to report an inactive lifestyle than peers (23% vs. 14%) [7], likely because acquired physical impairments interfere with ability and confidence to initiate and engage in regular exercise [8] [9, 10], and because their parents may be overprotectvie and discourage physical activity [11, 12]. However, evidence suggests that children perceive physical activity as a both a coping mechanism for symptoms after treatment [13], and as a mechanism to combat feelings of isolation due to treatment [10]. Data collected from survivors engaging in a virtual physical activity intervention with a social engagement component (via Facebook) points to increased levels of physical activity in participants who engaged in the social forum versus those who did not [14]. Therefore, interventions that address immediate barriers to engagement and promote participation in regular physical activity to recommended levels (60 minutes a day of moderate to vigorous physical activity (MVPA) [15]) may provide children a mechanism to help with coping and social interaction.

In-person exercise interventions demonstrate that children with cancer and young survivors have improved function and fitness in response to exercise [16]. While immediately effective, these interventions are not scalable as they are resource intensive, increase survivor/family burden, and not accessible to survivors who live far from intervention delivery sites. Completely remote interventions have also been published, but in smaller samples with a focus on acute lymphoblastic leukemia [17, 18]. Creative approaches are needed to develop interventions that increase physical activity and fitness in childhood cancer survivors of varying diagnoses that are cost-effective, easily generalizable, and accessible. A 24-week pilot study that randomized 78 childhood cancer survivors (any diagnosis) ages 11-15 years to an activity monitor in combination with either 1) education with access to an interactive website that rewarded physical activity (intervention), or 2) education alone, demonstrated increased weekly MVPA in the intervention group and decline in MVPA in the control group [7]. While preliminary results are encouraging, overall increase in physical activity was not substantial (4.7±119.9 mins post-intervention [19]), indicating need to enhance the intervention to encourage more activity.

COG ALL Stars, described here, uses data from this preliminary trial as its foundation and aims to increase physical activity and fitness in childhood cancer survivors by evaluating the added impact of virtual social support to the intervention that features an activity monitor, physical activity incentives, and interactive website. The study leverages resources of the Children’s Oncology Group (COG), supported by the National Cancer Institute (NCI). COG is the world’s largest children’s cancer research entity, with over 200 cooperating children’s hospitals (101 participating institutions in this study). The study aims are to compare the effects of a 6-month long, rewards-and web-based physical activity intervention that includes structured social interaction between participants to the same web-based physical activity intervention that does not include structured social interaction on 1) fitness, 2) markers of cardiometabolic health, 3) physical activity, quality of life (QoL), fatigue, and school attendance; and 4) to determine if the effect of a rewards-based, socially interactive, and web-based physical activity intervention on markers of cardiometabolic health is mediated by changes in fitness. The hypotheses are: 1) children randomized to the intervention will demonstrate lower fitness during a six minute walk test than children randomized to the control condition at the end of the intervention, as well as six months and one year after the intervention is completed; 2) children randomized to the intervention will have lower blood pressure, body mass index, waist to height ratio, fasting insulin, glucose, and lipid levels and more favorable inflammatory biomarker and adipokine profiles than children randomized to the control condition at the end of the intervention, as well as six months and one year after the intervention is completed; 3) children randomized to the intervention will report more physical activity, better quality of life, less fatigue, and fewer missed school days than children randomized to the control condition at the end of the intervention, as well as six months and one year after the intervention is completed; 4) the effect of the intervention on biomarkers of cardiometabolic health will be mediated by changes in fitness as assessed by the six minute walk test.

METHODS

Study Design and Participants

This is a 6-month long, two arm, prospective, randomized web-based physical activity intervention among children immediately following treatment for cancer (ClinicalTrials.gov Identifier: NCT03223753; COG Identifier: ALTE1631). Children 8-15 years of age, treated at any COG institution, who have completed treatment for their cancer in the past 12 months are randomized to either an activity tracker plus structured social interaction and rewards-based website/mobile application (web/app) tracker interface or an activity tracker plus a web/app tracker interface where there is data presentation without social interaction or in-intervention rewards. Outcomes are evaluated at enrollment, at the end of intervention, and 6- and 12- months post-intervention. The primary time point for analysis is the end of intervention. Participants are followed for 18 months, including intervention time. This study is approved by all clinical site Institutional Review Boards or Research Ethics Boards, NCI Cancer Trials Support Unit Regulatory Office, and Central Institutional Review Board.

Participant Eligibility and Recruitment

The study will enroll 384 children in the given age range from COG institutions. Each child will remain enrolled in the study for approximately 18 months. Enrollment is facilitated by the study team who communicate with COG institutions via regular email and phone calls to identify and track potentially eligible participants.

Study inclusion and exclusion criteria are outlined in Table 1. Because data indicates that children have an average decrease of 38 minutes per year on weekdays and 41 minutes per year on weekends from ages 9-15 years, this intervention includes children as young as age 8 to prevent the typical onset of decline in physical activity. Children with significant concurrent disease, illness, or psychiatric disorder or social issue that would compromise safety in participation are excluded, as in previous exercise interventions [20-22]. Based on the racial distribution of childhood cancer in the United States, we expect the sample to be 80% Caucasian, 11% Black, and 8.0% Asian/Pacific Islander. Twenty percent of the total sample is estimated to be Hispanic. (Estimates derived from the Surveillance Epidemiology and End Results Program.) Children with diagnoses of leukemia/lymphoma, bone tumor, central nervous system tumor, and other solid tumors are included. Combined mail, phone, and in-clinic approach is used to recruit individual participants. This study opened for recruitment on 08/14/2017; 189 participants have been enrolled.

Table 1.

Study Inclusion and Exclusion Criteria

Inclusion Criteria	Exclusion Criteria
Patient must be ≥8 years of age and <16 years of age at time of enrollment.	Patients with previous Hematopoietic Stem Cell Transplant.
All cancer cases with an ICD-O histologic behavior code of two “2” (carcinoma in situ) or three “3” (malignant), in remission.	Patients with significant concurrent disease, illness, psychiatric disorder or social issue, as determined by enrolling institution or physician, that would compromise patient safety or compliance with protocol therapy, or interfere with consent, study participation, follow up, or interpretation of study results.
Patient must have completed curative therapy (surgery and/or radiation and/or chemotherapy) within the past 12 months at a COG institution.	Female patients who are pregnant are not eligible. Women of childbearing potential require a negative pregnancy test.
Patients must have a performance status corresponding to ECOG scores of 0, 1, or 2.	Female patient who is postmenarcheal and has not agreed to use an effective contraceptive method (including abstinence) for duration of study participation.
At the time of consent, patient or parent/guardian reports less than current Centers for Disease Control and Prevention (CDC) physical activity guidelines for children (420 minutes of moderate to vigorous physical activity over the last week [15]).	Patients with a cognitive, motor, visual, or auditory impairment that prevents computer use (e.g. unresolved posterior fossa syndrome) are not eligible.
Patient and at least one parent/guardian are able to read and write English Spanish, and/or French. At least 1 parent/guardian must be able to read and write English Spanish, and/or French in order to assist the patient with using their physical activity tracking device account.	Patients who are unable to walk independently are not eligible.

Consent and Randomization

Study personnel inform the participant/parent or guardian that they will be randomly placed into an intervention or control group and that there is a chance they will not receive the intervention. When the participant/parent or guardian agrees to participate, study personnel will ensure that the potential participant and parent/guardian understand what the study involves, study measures, and potential risks and benefits. Consent and assent (if applicable) are signed in accordance with all local, federal, and institutional guidelines. Participants are then enrolled and randomized to the intervention or the control group. Patients are stratified by age groups (8-11 and 12-15 years), diagnosis group (leukemia/lymphoma, bone tumor, central nervous system tumor, other solid tumor) and gender and randomized 1:1 between the two arms.

Study Materials

All participants receive wearable activity monitors and educational materials. The patient is given instructions on set-up and use of the monitor during the baseline visit by study personnel. The participant is instructed to wear the activity monitor (MovBand5, Engage Moves, LLC) on the wrist every day during waking hours for the duration of the 6-month intervention. Reinforcement for device wear is provided the week prior to each of the three post-baseline study visits via text, email, or phone call. Other than encouraging weekly data download (sync), no specific instructions are given regarding how frequently participantts use/access the website/mobile application. Participants who find it difficult to wear the device continuously are encouraged to wear it for a minimum of two hours per day during periods when they expect to be active. The paper-based educational handouts (supplement) contain examples of types of physical activity, tips for exercising with cancer, and encouragement to reach Centers for Disease Control and Prevention physical activity guidelines for children [15], and are provided at the beginning of the study and via mail three and five months into the study. Handouts for children younger than 12 years old at enrollment contain additional physical activity information for parents. Study staff encourage participants to increase physical activity levels to recommended levels at initial visit; however, during the study, participants and/or their parents are free to choose how little or how much they engage in physical activity.

Intervention

Rewards-based interventions for physical activity are supported by the Triadic Neural Systems Model of motivated behavior [23], as cognition and affective development require interaction of neural systems that mediate threat, reward, and cognitive control. Neuroimaging indicates that rewards [24-26] and threat systems [27-29] follow a curvilinear developmental trajectory, most sensitive during older childhood and adolescence. Control processes mature linearly with age [30, 31], with an unbalanced system common among older children and adolescents whose physiology initially favors reward seeking behaviors. As success with behavior increases, children and adolescents are less motivated by extrinsic rewards, and new behaviors are maintained [32]. This intervention leverages rewards-based systems of motivation, as well as social interaction as a system of motivation for the intervention group (Figure 1). Selection of social interaction as a motivator is based on data demonstrating that social influence and support is associated with individual physical activity levels for children and adolescents [33, 34]. Social interaction is of high importance among children treated for or who have survived cancer report deficits in social functioning, likely due to disruption of social development during diagnosis and treatment phases of the childhood cancer experience [35, 36].

Figure 1. Theoretical Targets of Intervention.

Dashed lines representing indirect effects; solid lines representing direct effects

All participants earn rewards by being physically active and using the website and app (Engage Moves). When control and intervention participants sync their device and/or log in to the rewards website, they can see their physical activity data as moves, steps, distance, and calories in numeric and graphic formats. The activity monitor also displays activity data and heart rate (via photoplethysmography). Control participants get information about their physical activity and can take heart rate but cannot access to other participants’ physical activity data or activity challenges. Intervention participants get both individual activity data and can see how their activity compares to others. Intervention participants give each other “high fives” for progress and compete in activity challenges created periodically by study staff. The visualization component of platform, managed administratively by the study team, allows control participants to use the website for self-monitoring without the social interaction component. Study staff can view participants’ activity on the website, including data syncing and social interaction frequencies to create compliance reports. Privacy is protected as participants are not allowed to use any part of their legal name to sign up and the website/app does not have any identifying information, such as sex, age, or location.

The more active a participant is, the more points they accumulate. Points are earned for activity, which translates to credits towards incentives. Participants earn about 250 points for every one minute of MVPA, about 4 to 6 “moves”. Points are also earned for wearing and syncing the device. Five-dollar ($5.00) gift cards are awarded every five credits to the intervention group. One credit is earned for wearing the device for five days, for syncing the device five days in a row, and for every 125,000 points. Fifteen thousand points per day is the challenge all participants are given by study staff to meet(250 points per one-minute MVPA, 60 minutes per day to meet recommended guidelines). Only the intervention group receives detail about how to earn credits corresponding to gift cards. The control group also earns gift cards, but is not aware that they will get them, receiving them all the end of the 6-month intervention. Table 2 summarizes differences in group conditions. Participants from both groups will have continued access to the website after intervention. However, if the intervention is effective, control participants will be given full access to the website when the study is over.

Table 2.

Intervention vs. Control Group

Intervention Materials	Intervention Group	Control Group
Educational Handouts	X	X
Physical Activity Tracking Device	X	X
Website/App	Full access (see/interact with others)	Limited access (only see own progress/unable to interact with others)
Gift Cards/Prizes	During intervention	After 6-month intervention

Study Outcomes

Outcome collection includes fitness, markers of cardiometabolic health, physical activity, QoL, fatigue, and school attendance at enrollment, end of intervention, 6-months post-intervention, and 12-months post-intervention (Table 3). All visits for outcome collection will be scheduled to match the child’s follow-up visits with their treating institution.

Table 3.

Outcome Collection Timeline

Measure	Instrument	Time point				Detectable difference	Reference
		E	EI	6M	12M
Fitness	PCI from six-minute walk test	X	X	X	X	0.052 b/m	Geiger [37]
Blood pressure	Blood pressure cuff (mmHg)	X	X	X	X	4 mmHg	Going [47]
BMI percentile	Height with a wall mounted stadiometer (m), weight with an electronic scale (kg)	X	X	X	X	5.4%	Going [47]
Waist to height ratio	Waist circumference (cm), height (cm)	X	X	X	X	3.8%	Kuba [43]
Total cholesterol	Fasting* blood draw	X	X	X	X	16.1 mg/dL	Going [47]
High density lipoprotein	Fasting* blood draw	X	X	X	X	6.6 mg/dL	Going [47]
Low density lipoprotein	Fasting* blood draw	X	X	X	X	13.9 mg/dL	Going [47]
Triglycerides	Fasting* blood draw	X	X	X	X	28.1 mg/dL	Going [47]
Homeostatic Model Assessment for Insulin Resistance	Insulin and glucose - fasting* blood draw	X	X	X	X	0.2	Lee [46]
High sensitivity C-reactive protein	Fasting* blood draw	X	X	X	X	0.79 mg/L	Dowd [48]
Interleukin-6	Fasting* blood draw	X	X	X	X	1.02 pg/mL	Mazur [4]
Tumor necrosis factor alpha	Fasting* blood draw	X	X	X	X	0.69 pg/mL	Mazur [4]
Leptin	Fasting* blood draw	X	X	X	X	0.17 μg/L	Garcia-Mayor [49]
Health Related QoL	PedsQL Core Modules	X	X	X	X	8.6	Varni [59]
Fatigue	PedsQL Multidimensional Fatigue Scale	X	X	X	X	7.1	Varni [54]
School Attendance	Parent Report	X	X	X	X	2.7	(SD 5 days/month)

E= Enrollment; EI= End of intervention; 6M= 6 months post-intervention; 12M= 12 months post-intervention

Fitness is evaluated with the physiologic cost index (PCI) from the six-minute walk test. Children are asked to walk as fast as possible for six minutes on a level surface. Distance is recorded in meters [37]. Heart rate is monitored continuously. Physiologic cost index is calculated as maximum heart rate during walking minus resting heart rate divided by meters walked (beat per meter (b/m)) [38]. The PCI evaluates energy cost of walking and has little variation by age or sex. [39, 40]. Among healthy children, PCI is inversely correlated with oxygen uptake and has a mean value of 0.4±0.125 b/m [41, 42].

Anthropometric measurements include height by wall-mounted stadiometer (without shoes), weight by electronic scale, and waist circumference Trained clinic staff measure waist girth (cm) with a non-stretching anthropometric tape directly over the skin midway between the anterior superior iliac spine and lower rib margin. Measurements are recorded to the nearest 0.1 cm. Waist to height ratio has known associations with cardiometabolic risk in children [43] and is calculated by dividing waist circumference by height. Reference values are available [44]. Body mass index (BMI) percentile is calculated by dividing weight in kilograms by height in meters squared. Reference values from the Centers for Disease Control and Prevention are used to convert values into percentiles [16].

Resting blood pressure is taken after the participant has been sitting upright in a chair with both feet on the floor for five minutes. The right arm is supported at the level of the heart. One to two minutes after the first reading, a duplicate reading is taken on the same arm to ensure accuracy [45].

Ten (10) mL of blood is collected by venipuncture after an 8-hour fast and sent to a central laboratory for testing. Glucose levels are assessed using spectrophotometrics [46]. Insulin levels are assessed using electrochemilumiscent immunoassays (ECLIA) [46]. Lipid panels are obtained using colorimetrics [47]. High sensitivity C-reactive protein and tumor necrosis factor-alpha are measured via quantitative immunoturbidimetry [48] and quantitative multiplex bead assays [4], respectively. Both interleukin-6 and leptin levels are assessed by enzyme-linked immunosorbent assays (ELISA) [4, 49].

During the study, physical activity is continuously recorded by the tri-axial accelerometer technology in MovBand, and raw data is automatically transferred to the web-based platform when the device is synced with a mobile device. Activity is recorded in “counts” per 15 second epoch [50-52], and count data are converted into daily minutes of MVPA using established metrics [53]. Three days of wear will constitute valid wear time.

Quality of life and fatigue measures include the PedsQL™ 4.0 Generic Core Scale and 18-item PedsQL™ Multidimensional Fatigue Scale [54]. The PedsQL instrument is reliable and valid in children and adults with cancer [55, 56], and includes physical, emotional, social scale, and school functioning scales, a total summary score, and physical and psychosocial health summary scores. Scores range from 0-100 [55-59]. The 18-item PedsQL™ Multidimensional Fatigue Scale [57] has three subscales: (1) general fatigue (six items), (2) sleep/rest fatigue (six items), and (3) cognitive fatigue (six items). The format, instructions, Likert response scale, and scoring method are identical to PedsQL™ 4.0 Generic Core Scales, with higher scores indicating better health-related quality of life (fewer problems or symptoms).

Participants and/or parents are also be asked to indicate number of missed days of school (SD 5 days/month).

Sample Size Calculations

The study will enroll 384 children in the given age range from COG institutions. To evaluate our primary outcome, 192 evaluable participants are needed in each of the intervention and control groups, and allows for death, relapse, or drop out (~20%). With 154 children per group, power is 85%, with alpha of 0.01, to detect a 13% difference between groups in fitness (PCI). This sample size also gives power to detect differences noted in Table 2 for measures of cardiometabolic health, inflammation, leptin, physical activity, fatigue, QoL, and missed school days.

Data Management and Analysis

After initial registration, enrollment, and randomization, data entry is manually entered by the recruiting site. This data is submitted to the study coordinating center at St. Jude Children’s Research Hospital (SJCRH). Data collected during clinic assessments is filled out on case report forms and sent to SJCRH. The data received is scanned or hand-entered into a secure database.

Statistical Design and Data Analysis

This study is powered to detect a 13% difference in PCI between intervention and control groups. This corresponds to the decrease in peak oxygen uptake seen among children after cancer treatment when compared to peers, and represents the difference between climbing or not climbing stairs or between running and walking [60]. The analysis for the primary aim will be conducted by comparing PCI differences between groups in at the end of the intervention using intent to treat analysis of variance that controls for sex, age, and diagnosis. General linear mixed models will be utilized to evaluate effects of group assignment on changes in PCI over time to account for repeated measures on individual children, and for potential random effects such as original treating institution. A similar approach will be used to evaluate outcomes at 6- and 12-months to evaluate differences between groups and effects of group assignment on changes over time in outcomes of interest.

Because we will have data on daily MVPA over the course of the 24-week trial, we will use longitudinal methods as described by Fitzmaurice et al [61] to reduce data for analysis and assess impact of group assignment on levels of MVPA levels in the two groups over time.

For the fourth study aim, we will use a causal inference approach to evaluate whether group assignment results in an improved outcome in the presence of a lower PCI (evaluated in aim one) for both 6- and 12- months post intervention. This requires decomposing the averaged total effect into direct effects of group assignment and indirect effects, the mediation effects of fitness, on cardiometabolic outcomes. We plan to use a principal stratification method as described by Lynch [62] that classifies participants into strata based on their performance on PCI (mediator variable) for each group of the randomized intervention.

Missing and Incomplete Data

We anticipate missing data for participants who attend visits at <1%. When a participant initiates sync through the app, recorded activity data automatically transfers to the website, and is accessible to study personnel. Activity data not uploaded through syncing may be lost. Therefore, missing data are possible. To determine bias because of missing data and drop-out, we will do sensitivity analyses of outcome data of all enrollees, constructing and comparing models that do and do not carry forward baseline data. We will also consider using probability weighting to assess bias in our results; comparing baseline characteristics of those who remain on study and those who drop out. We will include variables where there are differences in a logistic regression model to determine probability of participation and use this variable to construct an inverse probability weight for each participant. Analyses with and without weighting will be compared to see if drop-out has an appreciable influence on our study results.

DISCUSSION

Evidence indicates that simply offering portable or remote interventions does not always translate to increased physical activity levels children with or who have survived cancer [63], suggesting a need for portable interventions with unique motivational strategies. This study utilizes social engagement and rewards to increase initial engagement and motivation, targeting systems salient for children and young adolescent survivors of childhood cancer at an age where they are responsive to extrinsic stimuli when learning or adopting a new activity. This study also allows participants to continue to use the platform beyond the intervention period, providing sustained access to social interaction and rewards to maintain physical activity behavior. The results of this study will demonstrate impact of social interaction and rewards-based motivation strategies on fitness and examine if the association is mediated by physical activity.

Using a readily available website for an intervention allows our team to examine impact of a readily available resources, saving time and money, optimizing early translation. A study of portable interventions is additionally timely, as the COVID-19 pandemic impacted children’s levels of physical activity and sedentary time [64]. Children with or who have recently completed treatment for cancer may be at an increased risk, as their immunocompromised state could cause further social isolation. Finding ways to optimize remote physical activity interventions for children with or who have survived childhood cancer is especially pertinent.

Other design factors contribute to strengths of this study. This study leverages existing COG framework to identify children who are finishing curative therapy and engages them in physical activity early in survivorship. Multi-site data accrual also provides a robust sample and a geographically, racially, ethnically, and socioeconomically diverse sample, maximizing generalizability.

The most significant potential limitation to this study is usability of the intervention platform. It is important to note that, while our pilot data shows occurrence to be rare, there is a potential that this intervention will not be well-received by individuals without internet interest or accessibility. A mobile phone-compatible option of the intervention was created to help with accessibility in those without reliable internet access. In case of technical issues with using the original device, an iPad with the app already downloaded will be provided for use or the enrolling site will contact the participant to complete a manual log to capture physical activity data. If a child/adolescent wants to participate and does not have access to internet at home or to a mobile device, with parent/guardian permission, study staff from the coordinating center will provide written and verbal (Skype, phone) instructions regarding the device and the website interface to school personnel so the child/adolescent can participate. Technical support is available for use of the website via a HELP button or a toll-free support number. Also, some participants might not remember study team queues to sync data or could be less convinced of importance of syncing. Activity monitors can malfunction or become lost, which can halt any incentive earning on the part of the participant and demotivate the participant; however, our budget accounts for 30% of devices to be replaced. Finally, as is the case with all physical activity interventions, there is a possibility of injury during participation. Participants are unlikely to incur physical injury past what could be experienced in daily life because participants are encouraged to increase time spent in activity that they actively engage in rather than increase intensity.

CONCLUSION

COG ALL Stars (ALTE1631) is an ongoing, multi-site intervention designed to encourage and motivate child and adolescent survivors to engage in physical activity soon after therapy completion. The effects of social interaction and rewards-based motivation strategies will be examined, as they are untapped resources that could be useful in developing long-term habits in young survivors. Findings from this study will inform future interventions geared towards physical activity adoption and maintenance for long-term health in children with or who have survived cancer.

Abbreviations

BMI

body mass index

COG

Children’s Oncology Group

MVPA

moderate and vigorous physical activity

PCI

physiologic cost index

QoL

quality of life

SD

school days

SJCRH

St. Jude Children’s Research Hospital

DATA STATEMENT

Protocol documents are available upon request to the corresponding author. Study data are embargoed until the study is complete but will be available upon request once the planned analyses are complete.