Design and methods of the StepByStep randomized trial of a mobile health and social media physical activity intervention among adolescent and young adult survivors of childhood cancer: A Report from the Children’s Oncology Group

Abstract

Background:

Interventions to increase physical activity are needed in adolescent and young adult survivors of childhood cancer who are largely inactive but at lifelong elevated risk of multiple chronic conditions improved by physical activity. The goals of the StepByStep study are to evaluate the effects of a 48-week distance-based, multi-component mobile health and social media behavioral intervention on physical activity, biomarkers of cardiometabolic health, and health-related quality of life.

Methods:

This ongoing study is a two-arm, prospective, multi-site randomized controlled trial. 384 childhood cancer survivors age ≥15 years and <21 years who were 3-36 months off therapy and not meeting physical activity guidelines were enrolled. The trial will test the efficacy of a 24-week intensive multi-component physical activity intervention combining a wearable physical activity tracker, social media peer support group, and individualized goal setting followed by a 24-week maintenance phase of the intervention to improve outcomes. The control group receives the wearable physical activity tracker only.

Conclusion:

There is a growing need for novel, developmentally appropriate interventions to increase physical activity and improve the health trajectory of adolescent and young adult survivors of childhood cancer. If efficacious, this portable and scalable intervention would be a much-needed tool to reduce the morbidity from cancer treatment and improve quality of life among survivors after treatment ends.

Clinical Trial Registration:

ClinicalTrials.gov Identifier: NCT04089358; COG Identifier: ALTE2031.

INTRODUCTION

Background

Children and adolescents with cancer have 5-year overall survival rates exceeding 86% due to advances in treatment [1]. Unfortunately, survivors of childhood cancer have an elevated risk of developing chronic conditions such as cardiometabolic disease and reduced cardiopulmonary fitness, obesity, and osteoporosis [2–4]. Physical activity (PA) reduces the risk and severity of many of these conditions and is associated with improved health-related quality of life (HRQOL) [5–9]. Most survivors, however, are physically inactive and may be more sedentary than the general population [10–12]. Specifically, 50-80% of adolescent and young adult survivors of childhood cancer (AYA survivors) do not meet recommended PA guidelines [11].

Novel, developmentally appropriate interventions are needed for the growing population of AYA survivors. Developmentally, AYA survivors, like their peers without cancer, are building independence and taking greater responsibility for their behaviors, including PA [13]. AYA survivors have voiced a strong interest in PA, with 87% reporting they are “very” or “extremely” interested in “getting in shape” [12]. Previous research has identified the barriers, facilitators, and preferences of PA in this population. They prefer to exercise at home [14]. They value peer group acceptance and peer relationships [13, 15], and they have expressed interest in peer support through electronic applications and internet platforms such as Instagram [16]. Individuals in this age group have also expressed an interest in tracking their PA behaviors [17]. While recent reviews on internet-based behavioral programs and mobile health (mHealth) apps to promote PA among AYAs have been reported, few long-term randomized controlled trials have examined wearable PA trackers coupled with mHealth tools to promote PA in AYA survivors [18–23].

Recent pilot studies have demonstrated uptake of mHealth interventions by AYA survivors and their potential for improving outcomes. Howell et al. tested a web-delivered, interactive PA intervention including a PA tracker and rewards system over 24 weeks in 97 children 11-14 years old [24]. The intervention group increased their moderate-to-vigorous physical activity (MVPA) and maintained the increase over time [24]. Le et al. conducted a pilot study of a 6-month wearable PA tracker intervention in 19 AYA survivors >15 years old that demonstrated feasibility as measured by participant retention, device use, and belief of utility, while also demonstrating increases in MVPA and peak oxygen [25]. Mendoza et al. piloted a multi-component PA intervention that combined a PA tracker with a Facebook group and goal setting via text messages to increase PA among AYA survivors [26]. All feasibility criteria were met including the recruitment goal (n=60), intervention adherence (Fitbits worn/synced for 70% of days), and questionnaire completion (≥90% completed all questionnaires) [26].

Building upon promising pilot work, the ongoing StepByStep randomized controlled trial (ClinicalTrials.gov Identifier: NCT04089358; COG Identifier: ALTE2031) investigates a distance-based, multi-component PA intervention combining a wearable PA tracker, a social media peer support group, and individualized goal setting among AYA survivors.

Study aims and hypotheses

The study aims are to compare PA-associated outcomes for patients who receive a 48-week distance-based, multi-component mHealth and social media behavioral intervention to patients who receive a PA tracker only among AYA survivors who do not meet PA guidelines [27]. The study hypotheses are that, compared to participants randomized to the control condition, participants randomized to the intervention will have greater 1) increases in minutes of MVPA as assessed by accelerometry, 2) improvements in biomarkers of cardiometabolic health as assessed by a two-minute step test (2MST), heart rate monitor, and dried blood spot (DBS) analysis, and 3) improvements in HRQOL as assessed by validated self-report instruments at the end of the intervention.

METHODS

Study schema

This two-arm, prospective randomized controlled trial evaluates the effects of a 48-week distance-based, multi-component mHealth and social media PA intervention. Participants are randomized to receive the intervention, including a wearable PA tracker, a social media peer support group, and individualized goal setting, or a PA tracker only (control group) (Figure 1). There is a 24-week intensive phase of intervention followed by a 24-week maintenance phase. Outcomes are assessed at the baseline evaluation occurring before randomization, an interim evaluation at 24 weeks post-randomization, and a final evaluation at 48 weeks post-randomization, which is the primary follow-up time point of interest for all analyses.

FIGURE 1.

Study Schema

Study setting

This study leverages the infrastructure of the Children’s Oncology Group (COG), a member of the National Cancer Institute-funded Community Oncology Research Program and the National Clinical Trials Network. COG is a consortium of >200 academic and community hospitals that care for >90% of children with cancer in the United States [28]. Eighty-three COG sites enrolled patients onto this study.

Coordinating center and participating COG sites

The coordinating center is a team of research staff at Georgetown Lombardi Comprehensive Cancer Center in Washington, DC and Seattle Children’s Research Institute in Seattle, Washington. The coordinating center is the primary source of contact for participating COG sites, implements all intervention components, and conducts remote evaluations. Participating COG sites are responsible for identification, recruitment, and enrollment of study participants and in-person evaluations. Participating COG site staff also enter clinical data into a baseline medical record abstraction form for each participant. The Central Institutional Review Board for the National Cancer Institute serves as the single institutional review board for all participating COG sites. Coordinating center research activities were approved by the institutional review board at Georgetown.

Adaptation to COVID-19 pandemic

The study was initially planned with in-person assessment of outcomes in the oncology clinics of participants’ respective COG sites. However, with the emergence of the COVID-19 pandemic just prior to the start of study enrollment, all evaluations were modified to be distance-based to accommodate social distancing guidelines and hybrid work environments. Adaptations to allow remote assessment include conducting a 2MST [29] at home with phone guidance by research staff instead of a six-minute walk test [30] in clinic, collecting biomarker specimens through patient-administered finger sticks onto DBS [31] instead of venipuncture in clinic, and electronically sending questionnaires to participants rather than administering them in clinic. We included an instrument derived from the World Health Organization’s questionnaire on behavioral insights research for COVID-19 [32] supplemented by additional questions from several relevant instruments [33–37] to assess potential pandemic-related influences on participation and engagement.

Participant eligibility

Eligible patients met all of the following criteria at enrollment: 1) age ≥15 years and <21 years (ensuring participants in social media group are similar in age so all content is age-appropriate, while still meeting accrual goals), 2) first diagnosis of malignant neoplasm (International Classification of Disease for Oncology behavior code of “3”) [38] in first and continuous remission, 3) history of chemotherapy and/or radiation, 4) all cancer treatment completed 3-36 months prior to enrollment, 4) life expectancy ≥1 year, 5) self-report of <420 minutes of MVPA per week as assessed via the Godin-Shepard Leisure-Time PA Questionnaire [39], 6) ambulatory and no known medical contraindications to increasing PA, 7) no known significant physical or cognitive impairment preventing use of the electronic devices used for the intervention, and 8) ability to read and write English. Exclusion criteria included: 1) previous history of an allogenic hematopoietic stem cell transplant, and 2) post-menarchal female patients who were pregnant or planning to become pregnant within the next year.

Recruitment and randomization

Patients were recruited in person, by email, or by phone. Recruitment was facilitated by a recruitment letter, study flyer, and engaging recruitment video. After informed consent/assent was obtained, participants completed baseline assessments and were randomized to the intervention or control group by the coordinating center using a block randomization design stratified by sex assigned at birth and current age (15.0-17.9, 18.0-20.9 years), with random block sizes of 2,4, or 6. Research staff at participating COG sites and data collection staff at the coordinating center are blinded to randomization.

Intervention framework and theory

A socio-ecological model [40] was used in the design and evaluation of the intervention. On the intrapersonal level, self-determination theory (Figure 2) will be the theoretical framework for describing changes to participants’ PA at the individual-level [41–43]. For neighborhood and community influences, we will use the participant’s home address to obtain: daily weather conditions (temperature, precipitation) during the study and neighborhood walkability (Walk Score^®, Transit Score^®) [44–46]. Intervention participants’ engagement on the Instagram group (“likes,” comments) will be assessed as a measure of the potential peer influence from the social media peer support group on participants’ PA (interpersonal-level and community-level influence).

FIGURE 2.

Self-Determination Theory Framework Model

The intervention is guided by the self-determination theory for which there is proven value in PA promotion [47]. Self-determination theory asserts that three psychological needs drive an individual’s motivation for PA: competence (skills, ability), autonomy (choice, control), and relatedness (connection to self and others, e.g. peers). In the intervention, the wearable PA tracker and associated mHealth app will increase competency and autonomy by allowing participants to set personalized step count goals and to track their own progress toward those goals (self-monitoring). Goal setting and self-monitoring are predictive of short- and long-term PA behavior change [48]. Moreover, the social media group is designed to increase relatedness by providing a virtual support group or community of peers to encourage and offer positive reinforcement. We will assess participants’ psychological needs using the Psychological Need Satisfaction in Exercise Scale [49]. We will measure intrinsic motivation for PA using the Behavioral Regulation in Exercise Questionnaire-2 and scales on “amotivation” and integrated regulation [49–51].

Items delivered to both study arms

Educational materials:

All participants received educational materials in hard copy or by email (Figure 1) adapted from material by the American Society of Clinical Oncology regarding risk of late effects from cancer treatment and the importance of PA [52].

Wearable PA tracker:

All participants received the Fitbit (InspireHR^®, Inspire 2^®, or Inspire 3^®) at the baseline clinic visit (if enrolled in person) or by mail (if enrolled remotely). Research staff assisted participants with setting up a user account and initial usage of the Fitbit (wearing, syncing) and the associated mHealth app. The Fitbit tracks users’ movements using a 3-axis accelerometer, which is then converted to estimates of step counts, proxy estimates of PA, including steps, energy expended (calories burned), and distance traveled. The Fitbit displays activity-related data directly on the wristband to provide immediate feedback on goal attainment throughout the day. The data also syncs with the Fitbit mHealth app on mobile devices to provide feedback and custom graphs of the data.

The control group received a Fitbit to reflect current realities that electronic activity trackers are nearly ubiquitous and built into most smartphones. Only the intervention group integrates Fitbit data with the social media peer support group and research staff-assisted goal setting. During the intensive phase (Table 1), the intervention group also receives daily reminders to wear the Fitbit. The Fitbit is used to give behavioral feedback to participants and not as a data collection tool for the primary outcome of MVPA.

TABLE 1.

Description of Intervention Components by Phase of Intervention

	Intensive Phase (Weeks 1-24)	Maintenance Phase* (Weeks 25-48)
Fitbit	Wear daily plus weekly reminders	Wear daily, no reminders from study staff to wear it
Private Instagram Group	Study staff moderate Instagram group  • Post 2-3 times per week with content/discussion questions on physical activity  • Tailor additional posts to interests and topics raised by participants  • Provide badges to participants	Study staff moderate Instagram group  • Study staff only post once per week with general physical activity-related questions (e.g. how do you keep active when the weather is bad?)  • Participants take the lead posting content and sharing experiences  • No badges provided
Individualized Goal Setting	Study staff conduct weekly brief goal setting sessions by phone call or text messaging with participants, with increasing autonomy for personalized step count goal setting (i.e., weekly increase of 10%, keep same goal, try lower goal if too high)	Participants set their own step count goals weekly (i.e., weekly increase of 10%, keep same goal, try lower goal if too high) Study staff check in with participants monthly to assist with goal setting

^*Major differences between maintenance and intensive phases are bolded.

Additional items delivered to the intervention group only

Social media peer support group:

A private, invitation-only social media peer support group was specifically created on Instagram for intervention participants. Coordinating center interventionists assist participants with using all features of the Instagram group on their mobile devices. Separate Instagram groups are operated for the intensive and maintenance phases of the intervention to adhere to phase-specific activities and to maintain security. Interventionists have log-in access to these groups and monitor each group. As participants complete the intensive phase, interventionists remove them from the intensive phase group and grant them access to the maintenance phase group.

During the intensive phase (Table 1), interventionists provide encouraging advice and build participant engagement by posting messages for PA 2-3 times per week, announcing badges/awards weekly, and moderating a forum to discuss PA and use of the Fitbit. Posted messages are based on self-determination theory (e.g., “What motivates you to achieve your step goal?”, “Who inspires you to achieve your goals?”). Participants are encouraged to share content including photos and videos to introduce themselves to the other members of the Instagram group and to share their PA.

During the maintenance phase, interventionists decrease their interactions and posts on Instagram and encourage participants to take the initiative in directing group content. Participants are invited to volunteer to share their experiences and unique content, such as videos and pictures of their PA, in the Instagram group for 3-5 days.

Individualized goal setting:

During the intensive phase (Table 1), participants receive goal setting messages from interventionists weekly by text or phone call to set personalized step count goals using the Fitbit mHealth app and based on their self-monitoring of step counts via the Fitbit (Supplementary Figure A). The individualized advice is proceduralized in an algorithm based on actual Fitbit daily step count data. Ideally, participants increase their average daily step counts by 10% (or at least 500 steps per day) each week. Participants have the option to keep the same or set a lower step count goal for a few weeks in which they had suboptimal opportunities or motivation for PA, to avoid setting overly high expectations. In general, the goal setting activity is initially led by interventionists, but gradually gives participants greater skills and autonomy for setting their own step count goals. During the maintenance phase, interventionists only contact participants once a month to inquire about overall progress of independent goal setting (Supplementary Figure B). Participants have the option to decrease the frequency of messages or opt out entirely.

Length and varying intensity of intervention

The length of 24 weeks for the intensive phase was based on a review of nine information- and communication technology-based interventions designed to improve PA levels among children and adolescents [18]. Study durations ranged from two weeks to two years. The intervention with the largest effect size (0.8 for males, 1.2 for females) used an activity tracker (pedometer) and was six months in length [53]. An ongoing challenge across studies is the maintenance of PA behaviors after the intensive phase ends [18]. Therefore, a maintenance phase is included in the current study. The maintenance phase was designed anticipating that the participants will have increased their independence for PA. Thus, intervention activities are mostly directed by the participants and require little interventionist input.

Participant retention strategies

COG sites were encouraged to have participants complete contact information forms at enrollment so research staff have multiple contact methods, including alternate contacts. To engage participants, study messaging is succinct, and materials are colorful (e.g. eye-catching logo). Communication strategies tailored to AYAs, include electronic study newsletters and 2-way text messaging (SendHub^®[54]) for goal setting, scheduling assessments, and building relationships with participants (e.g. sharing PA-related photos, birthday messages).

Study outcomes

The primary outcome of MVPA is assessed by the ActiGraph GT3x+^® research grade accelerometer worn at the hip [55]. Accelerometers provide an objective measure of duration, frequency, and intensity of movement over time, thereby providing a valid and objective measure of PA [56, 57]. Participants wear the accelerometers for 3-7 days at each of the three study timepoints. Total minutes above the moderate-intensity threshold will be divided by the number of valid days of wear time to obtain minutes of MVPA per day, which is the primary outcome measure. Participants are encouraged to achieve at least four valid days of wear time [56]. To maximize valid accelerometer data, participants wear accelerometers for 24 hours per day as tolerated, except for water-based activities [58], and are asked to re-wear them for another seven days if they have less than four valid days of wear time. Changes in light PA and sedentary time will also be measured as exploratory PA outcomes.

Cardiopulmonary fitness is assessed by the clinically standard 2MST [29], which is administered remotely by phone/video by the coordinating center or in person by staff in clinic. The 2MST is performed in a safe, level, easily accessible location where participants can march in place uninterrupted. Participants are instructed to alternate bending one knee and raising it to hip height (or as high as physically able to) as fast as they can for two minutes (performance measured as number of right-side steps completed in two minutes). During the 2MST, participants count their steps using a counter app on their smartphones. Resting heart rate is assessed by Fitbit at home or in clinic in beats per minute.

Blood biomarkers of cardiometabolic health are assessed by DBS [31] at baseline and final evaluations only. DBS sampling is completed with finger sticks by participants at home or in clinic, ideally after an 8-hour fast. Patients receive print and video instructions on how to collect the DBS [59]. At least three spots (each approximately 50-60 μL) are collected on a collection card and dried for at least three hours at room temperature. DBS are placed in a plastic bag and stored at room temperature until ready to be shipped to a central laboratory for testing. Lipid panels (total cholesterol, high density lipoprotein, low density lipoprotein, triglycerides), homeostatic model assessment for insulin resistance, glucose, insulin, hemoglobin A1c, and high sensitivity C-reactive protein are assessed for all DBS [60, 61].

HRQOL is assessed with standardized instruments validated and well-used among AYA survivors. These include the Pediatric Quality of Life Inventory^™ 4.0 Generic Core Scales [62] (global functioning, physical functioning, and social functioning scales, specifically) and the 18-item Multidimensional Fatigue Scale [63], which encompasses three subscales: general fatigue, sleep/rest fatigue, and cognitive fatigue. Study questionnaires are administered electronically (or via hard copy) by the coordinating center. Participants <18 years old complete study questionnaires with help from their parent/guardian, as needed.

Additional Data Collection

Height (without shoes; by measuring tape at home or by wall-mounted stadiometer in clinic to nearest 1cm) and weight (without shoes or outer clothing; by bathroom scale at home or by scale in clinic to nearest 0.1kg) are collected. Body mass index will be calculated as weight (kg) divided by height (m) squared [64] and will be examined as a potential covariate in analyses.

Qualitative interviews

Semi-structured qualitative interviews will be conducted with 40 intervention participants following their final evaluations to generate feedback regarding the intervention (timing, duration, usability, etc.), acceptability, and suggestions for improvement. Twenty interviews will specifically target participants without private health insurance to obtain feedback on improving and tailoring the intervention for that underserved group. Semi-structured qualitative interviews will also be conducted with 25 research staff at participating COG sites. To generate a representative sample, research staff will be interviewed from sites varying in size (number of faculty), location (urban/rural), academic vs. community focus, and annual patient volume. Interviews will inquire about ease of recruitment, study challenges, and feedback received from patients.

Sample size and power calculations

Based on previous pilot data [26], the study recruited 384 participants, 192 per arm (intervention and control). Assuming a 25% attrition rate, there will be 290 evaluable participants (145 per arm). With this sample size, we will have at least 90% power to detect a 10-minute difference in the average change in MVPA from baseline between groups at a two-sided significance level of 0.05 with a t-test if the pooled standard deviation is less than 26 minutes per day, and at least 80% power if the standard deviation is less than 30 minutes per day.

Statistical analyses

Balance of patient characteristics between study arms will be evaluated using t-test, Wilcoxon Rank Sum test, Chi-square, or Fisher’s Exact test, as appropriate. If significant differences are observed, further analyses will be adjusted for the relevant factors. All analyses will be performed using SAS 9.4 software (Cary, NC) or R [65]. For the primary analytical strategy, outcomes will be formulated as change in measures from baseline to final evaluation (48 weeks). An intention-to-treat analytic approach will be used to the extent possible, with modifications for missing outcome data. Thus, all eligible and randomized patients who provide post-randomization accelerometer data will be used in analyses. Primary analysis estimates of intervention effects on these change scores will be ascertained from linear models and presented along with associated confidence intervals and two-sided p-values. Analyses will model the change in MVPA from baseline as the dependent variable and will include an indicator variable for treatment assignment and potential confounding factors, if needed. In secondary longitudinal analyses, outcomes and covariates from baseline, 24, and 48 weeks will be included in a model with treatment assignment, a time variable (0, 24, or 48 weeks), the interaction between the two, and any necessary adjustment factors as covariates. In additional secondary analyses, change in PA, biomarkers of cardiometabolic health, and HRQOL outcomes will be examined at interim evaluation (24 weeks), immediately at the end of the intensive phase, to isolate the efficacy of the intensive phase from the maintenance phase that follows.

Exploratory analyses will include health insurance status (yes/no private insurance) as a moderator. Additional analyses will assess associations and mediating relationships between the intervention, psychological needs, MVPA, biomarkers, and HRQOL measures.

Study progress

The study recruitment goal (n=384) was achieved on 8/24/2023. See Table 3 for selected characteristics of participants and Figure 3 for the CONSORT diagram. Most participants were diagnosed with leukemia or lymphoma (38.4% and 29.8%, respectively) at a mean age of 15.2 (± 2.2) years. Overall, 46.2% of participants were male at birth, 70% were White race, 35.8% were of Latinx ethnicity, 61.4% had private insurance, and 81.9% lived in urban/suburban areas. As of March 2024, >95% of study participants have chosen remote evaluation of outcomes. The last participant is estimated to complete all study procedures and the final evaluation in February 2025.

TABLE 3.

Demographic and Clinical Characteristics of Enrolled Participants (N=383)^*

Characteristic	N (%) or Mean ± Standard Deviation (range)
Cancer diagnosis
Acute lymphoblastic leukemia	113 (29.5%)
Other Leukemia	34 (8.9%)
Lymphoma	114 (29.8%)
Central nervous system tumor	14 (3.6%)
Sarcoma	54 (14.1%)
Other Solid Tumor	54 (14.1%)
Age at diagnosis (years)	15.2 ± 2.2 (1.9-20.3)
Age at enrollment (years)	17.9 ±1.7 (15.0-20.9)
15.0-17.9	197 (51.4%)
18.0-20.9	186 (48.6%)
History of chemotherapy
Yes	379 (99.0%)
No	4 (1.0%)
History of radiation
Yes	66 (17.2%)
No	317 (82.8%)
Sex assigned at birth
Male	177 (46.2%)
Female	206 (53.8%)
Gender (self-report)
Male	147 (38.4%)
Female	167 (43.6%)
Other	12 (3.1%)
Chose not to answer	57 (14.9%)
Race
White	268 (70.0%)
Black	31 (8.1%)
Asian	19 (5.0%)
Other	22 (5.7%)
More than one race	36 (9.4%)
Unknown	7 (1.8%)
Latinx ethnicity
Yes	137 (35.8%)
No	243 (63.4%)
Unknown	3 (0.8%)
Insurance
Private†	235 (61.4%)
Public	142 (37.1%)
Other‡	4 (1.0%)
Unknown	2 (0.5%)
Residential area§
Urban core	276 (72.0%)
Suburban	38 (9.9%)
Large rural	37 (9.7%)
Small town/rural	32 (8.4%)

^*One participant withdrew consent prior to completion of medical record abstraction form

^†Includes 6 participants with military insurance

^‡Includes self-pay and individuals who have both Medicare and private insurance

^§Residential area was determined based off participant’s zip codes and United States Department of Agriculture 2010 Rural-Urban Commuting Area Codes: https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/

FIGURE 3.

CONSORT Diagram

DISCUSSION

Generalizable interventions tailored to the preferences of AYA survivors are needed to increase PA in this vulnerable, understudied group. This distance-based, multi-component mHealth and social media behavioral intervention seeks to improve PA and health outcomes for this population. While this study builds upon the current discourse to improve PA among AYA survivors [66–68], there are notable differences from previous studies. First, combining a wearable PA tracker with a social media peer support group and integration of activity data leading to individualized goal setting creates a robust intervention to be evaluated. Second, the current intervention is tailored with AYA feedback from previous pilot work [24, 26, 69] and a recent prospective study [70] that showed peer support to be a positive predictor in increasing exercise among AYA survivors. Also, the intervention is community-based and incorporates familiar technology and social connections with peers.

A strength of the study is the inclusion of two study time points post-randomization. The interim evaluation after the intensive phase of the intervention will determine the immediate impact, while the final evaluation after the maintenance phase of the intervention will provide insight on long-term behavior change and sustainability of outcomes. Recruitment within COG is another strength because it leverages the collaboration and diversity of the participating COG sites.

In addition, a notable strength of this study is the adaptation to the COVID-19 pandemic, where there has been increased patient interest in telehealth. The adapted design necessitated by the COVID-19 pandemic shows that a large-scale randomized controlled trial can effectively be conducted remotely. Remote evaluations allow for greater participation among COG sites, thereby increasing the study’s representation. A large proportion of participants are non-White, of Latinx ethnicity, with public insurance, and living in rural areas. Previous studies of home-based interventions for survivors of childhood cancer included research staff accommodating home or clinic visits to collect biospecimens for study data [66]. The current design allows for at-home activity tracking via accelerometers, cardiopulmonary fitness test via the 2MST, biospecimen collection via DBS, and electronic survey collection. Further, the mHealth intervention with text messaging and social media is an emerging tool in AYA survivors and leverages their use of smartphones and desire for distance-based interventions and social connectivity [15]. If efficacious, the proposed intervention is highly generalizable and can be disseminated readily in diverse clinical settings.

CONCLUSION

The ongoing StepByStep study investigates a novel, multi-component, distance-based mHealth and social media behavioral intervention in AYA survivors to improve PA and health outcomes. The intervention is tailored to the needs and preferences of AYA survivors. If shown to be efficacious, this portable and generalizable intervention would be a much-needed tool to reduce the morbidity from cancer treatment and improve quality of life among survivors.

TABLE 2.

Schedule of Outcome Measures

Outcome	Instrument	Baseline Evaluation*	Interim Evaluation	Final Evaluation
Moderate-to-vigorous physical activity (primary outcome)	Research grade accelerometer	✓	✓	✓
Biomarkers of cardiometabolic health		✓	✓	✓
Cardiopulmonary fitness	2-minute step test
Resting heart rate	Fitbit heart rate monitor
Total cholesterol	Dried blood spot
High density lipoprotein	Dried blood spot
Low density lipoprotein	Dried blood spot
Triglycerides	Dried blood spot
Homeostatic model assessment for insulin resistance	Dried blood spot
Glucose	Dried blood spot
Insulin	Dried blood spot
Hemoglobin A1c	Dried blood spot
High sensitivity C-reactive protein	Dried blood spot
Health-related quality of Life		✓	✓	✓
Global, physical, social functioning	Pediatric Quality of Life Inventory™ 4.0 Generic Core Scales (global, physical, social functioning scales)
Fatigue	Pediatric Quality of Life Inventory™ 4.0 Multidimensional Fatigue Scale

^*Baseline Evaluation = prior to randomization; Interim Evaluation = 24 weeks post-randomization; Final Evaluation = 48 weeks post-randomization (primary endpoint)

Abbreviations:

AYA

Adolescent and young adult

AYA survivors

Adolescent and young adult survivors of childhood cancer

COG

Children’s Oncology Group

DBS

Dried blood spot

HRQOL

Health-Related Quality of Life

MVPA

Moderate-to-vigorous physical activity

mHealth

Mobile health

PA

Physical activity

2MST

Two-minute step test