Scalability of Cancer SurvivorLinktm: A Cluster Randomized Trial among Pediatric Cancer Clinics

Abstract

Background:

Children diagnosed with cancer are living longer and the survivor population is growing. However, most survivors develop late effects of radiation and chemotherapy shortly to years after completion of therapy, and the receipt of follow-up visits that are recommended by the Children’s Oncology Group (COG) is suboptimal nationally.

Aims:

The aims of this study are to: 1) evaluate the impact of a patient-controlled electronic personal health record (ePHR) and system (SurvivorLink) on care visit attendance, risk-based surveillance, and other secondary outcomes (i.e., patient activation, quality of life (QOL)); 2) measure the use, acceptability, and perceived usefulness of, and satisfaction with SurvivorLink; and 3) assess facilitators and barriers to implementation.

Methods:

This hybrid effectiveness–implementation, clustered randomized control trial (RCT) evaluates the effect of SurvivorLink among pediatric cancer survivors and their parents on receipt of follow-up cancer care. We will recruit 20 pediatric survivor clinics with half receiving the intervention and half acting as a waitlist control. Parents of survivors and survivors will complete baseline, 3 and 12 month surveys that assess SurvivorLink use, patient self-efficacy, and intentions to return for follow-up. We will use mixed methods and multi-informant assessment to assess implementation outcomes (i.e., acceptability, feasibility, appropriateness).

Discussion:

New approaches are needed to facilitate the receipt of long-term follow-up care among pediatric cancer survivors. This study will assess whether SurvivorLink is effective in increasing receipt of follow-up cancer care. Moreover, it will explore the influences of context and other moderators of clinical practice change in pediatric cancer survivorship.

1. Introduction

Due to improvements in cancer therapy over the past three decades, children diagnosed with cancer are living longer and the survivor population is growing. Data from the National Cancer Institute show that the overall 5-year survival rate for childhood cancer has increased from 45% in 1970 to over 84% in 2014 [1]. However, childhood cancer and its subsequent treatment predispose survivors to a higher risk of certain life-threatening and debilitating diseases called late effects [2]. Numerous reports and reviews of late effects of chemotherapy and radiation describe the sequelae that may present after cancer shortly following the end of active treatment until years after the completion of treatment [3–6]. These publications show that type and intensity of therapy, as well as age at therapy, are important factors in determining both overall survival and risk for developing late effects [7–11]. Long-term follow-up of pediatric cancer survivors has found that the cumulative incidence of chronic health conditions by age 45 years is 95.5%. Moreover, 80.5% of survivors have conditions graded as serious, life threatening, or disabling [12]. Pediatric cancer survivors often have more than one chronic condition; 38% reported having at least two conditions and 24% reported having three or more conditions [13]. In addition, survivors experience increased cause-specific mortality from subsequent malignancy (Standardized Mortality Ratio [SMR]= 15.2), cardiac (SMR=7.0), and pulmonary (SMR=8.8) causes relative to the general US population [14].

These data summarized above clearly indicate that cancer survivor care should be lifelong and include monitoring for early detection and treatment of late effects. A panel of experts and medical specialists working with the Children’s Oncology Group (COG) developed the Long-Term Follow-Up Guidelines for Survivors of Childhood, Adolescent and Young Adult Cancers (LTFU Guidelines) to improve individual monitoring for specific late effects [15, 16]. These evidence-based guidelines, organized by therapeutic exposure, include a list of all late effects associated with that treatment and screening tests that are recommended for each late effect. Aligned with the Institute of Medicine (IOM)’s call for survivorship care plans for all cancer survivors [17], the COG LTFU Guidelines can be used to develop survivorship care plans for pediatric cancer survivors that include survivors’ cancer treatment summary and individualized late effects risk profile and surveillance plan. The personalized survivorship care plan serves as a guide to long-term follow-up care for the patient and their healthcare providers.

1.1. Challenges of follow-up cancer survivor care

Currently, the COG recommends that pediatric cancer patients be referred to a cancer survivor program for evaluation at two years after the completion of cancer therapy, with regular follow-up visits thereafter. In addition to follow up in a survivor program, survivors often require medical care from a number of other healthcare providers in both primary and subspecialty fields. Many of these healthcare providers are unaware of the complex medical needs of pediatric cancer survivors [18]; consequently, cancer survivors or their parents are often left to ensure these providers are aware of their pediatric cancer history and recommended surveillance tests. Communication with and between a survivor’s various healthcare providers can be challenging but crucial, particularly during times of a health crisis, move to a new city, or acquisition of a new provider(s) during the transition from the pediatric care system to the adult care system, for example.

1.2. Evidence-based pediatric cancer survivorship interventions

Promotion of evidence-based, cancer survivorship interventions to impact improve lifelong survivor care and quality of life is in its infancy. A review of the National Cancer Institute’s Research-Tested Intervention Programs (RTIPs) website yielded only 19 interventions for survivorship/supportive care, a majority of which were supportive care interventions for adults diagnosed with cancer [20]. Only one intervention (Surviving Cancer Competently Intervention Program (SCCIP) centered on pediatric cancer survivors. Further, a recent review examined the application of implementation science across the cancer control continuum and found only 6% of NCI-funded implementation science grants (n=4) were focused on survivorship [21].

Cancer SurvivorLinktm, www.cancersurvivorlink.org, was created to improve patient-provider communication and increase adherence to survivor care recommendations. SurvivorLink is a patient-controlled electronic personal health record (ePHR) where users can upload and store their important health documents and electronically share their health record with all of their healthcare providers who are registered on SurvivorLink. If the clinic where they receive survivor care has an affiliation with SurvivorLink, registrants have the opportunity to affiliate with their survivor clinic, allowing the clinic to upload health documents on their behalf [19]. The SurvivorLink functionality allows survivor clinic users to manage patients who select them as their clinic, the medical record release requirements of their institution, and the relationship with their own survivors. Educational materials about survivor care and late effects of cancer therapy are also available for patients/parents and providers.

1.3. Aims and Objectives

To address this gap of effective pediatric survivorship interventions in the literature, this is the first study to conduct a hybrid I effectiveness-implementation trial of a survivorship intervention in the pediatric survivor clinic setting. The overarching goal is to implement SurvivorLink in pediatric survivor clinics to improve a survivor’s adherence to late effects surveillance.

The overall aims of this study are to examine how context influences the implementation and sustainability of SurvivorLink and to assess its impact on survivor’s adherence to cancer survivor care and late effects screening. Key objectives are:

1. Develop a standardized process to implement SurvivorLink within pediatric cancer clinics.

1.2. Determine the impact of SurvivorLink on patient outcomes, including survivor clinics’ attendance and engagement in risk-based surveillance, as well as patient reported outcomes including activation, quality of life (QOL) and health self-efficacy to manage a chronic condition.

2.3.Evaluate the uptake, acceptability, and utilization of SurvivorLink among adolescent childhood survivors/ and their parents and at the clinic level.

3.4.Assess the integration of SurvivorLink into pediatric survivor clinics’workflows and identify contextual facilitators and barriers to implementation.

2. Methods

2.1. Trial design

Using the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework and the Consolidated Framework for Intervention Research (CFIR) [22, 23] the study design is a hybrid 1 effectiveness-implementation, clustered randomized waitlist control trial, considering pediatric survivor clinics as units of clustering. The intervention will be packaged for delivery and training materials will be developed in Year 1 of the study. We will recruit pediatric clinics in Year 2 and implement the program in Years 2 (intervention) and 3 (waitlist). The Emory IRB has approved the study protocol.

2.2. Study population

2.2.1. Pediatric survivor clinics

The COG, a National Cancer Institute supported clinical trials group, is the world’s largest organization devoted exclusively to childhood cancer research [15]. More than 90% of 14,000 children and adolescents diagnosed with cancer each year in the United States are cared for at COG member institutions.

For this trial, 20 pediatric survivor clinics will be recruited from COG centers. According to the most recent data, 130 of the 220 COG centers (59%) reported having a designated pediatric cancer survivor clinic (i.e., presence of late effect clinics/services) with a specialized late effects provider(s) that followed recommended clinical practice as described in the COG LTFU Guidelines [24]. Additionally, clinics must meet the following eligibility requirements: 1) provide a survivorship care plan to survivors seen in clinic, 2) see > 100 pediatric cancer survivors annually, 3) be willing to become a SurvivorLink partner clinic and complete a Business Associates Agreement with Emory University, and 4) be able to enroll a minimum of 75 patients in one year.

2.2.2. Patients

Eligibility criteria for this study include that survivors be at least two years off therapy and attendance at a pediatric cancer survivor clinic. Each survivor clinic will document number of eligible patients approached and if they choose to register on SurvivorLink. Clinic staff will approach caregivers/parents of survivors < 18 years old and survivors 18-22 years of age who are either English or Spanish speaking during their clinic visit to assess their interest in participation in this study. The only exclusion criterion is having a terminal diagnosis.

2.2.3. Randomization

Pediatric cancer survivor clinics will be randomized to either the SurvivorLink arm or the waitlist control arm in Year 2 of the study. Clinics will be matched in pairs based on patient population size, level of current survivorship care, and location (urban/rural) using the urban/rural continuum codes [25]. The clinics will be block randomized in groups of 2 or 4 and one member of each pair will be randomly assigned to the SurvivorLink arm and the other to the waitlist arm. See Figure 1 for the CONSORT diagram for our clustered randomized trial.

Fig 1.

Clustered RCT CONSORT flow diagram

2.3. SurvivorLink Implementation

The core elements of SurvivorLink implementation are: 1) training of clinic staff; 2) education and registration of childhood cancer survivors and/or their parents; 3) distribution of the survivorship care plans to survivors via SurvivorLink; and 4) management and tracking of survivor and parent use of SurvivorLink. To increase the fit of the program, some adaptations of core elements (Elements 2 and 3) in the format of content and delivery will be allowed given the different infrastructure of the clinics.

Applying principles of the Replicating Effective Programs (REP) model [26], all training materials for SurvivorLink will be standardized and packaged in a toolkit with supporting materials. Stakeholders serving on the SurvivorLink Advisory Board will review the toolkit for user-friendliness of the protocol, training, and tools, and any missing information. This Board currently includes a pediatric cancer nurse practitioner and oncologist, family physician, representatives from patient advocacy groups, and parent and patient members. The Board will be expanded to include representatives from the COG’s Survivor and Outcomes Committee beginning in Year 1 and patient/parent and provider stakeholders from participating institutions in subsequent years.

Training of clinic directors and staff at randomized clinics will be conducted through webinars, PowerPoint presentations and use of the toolkit. During the training, the SurvivorLink team will provide an overview of the intervention, a logic model of program elements leading to key healthcare outcomes, key intervention components, and the system itself. They will demonstrate the functionality of SurvivorLink from all user perspectives (patient, parent, and clinic). In addition, discussions will apply principles of workflow design to 1) ensure fit of the core elements into their clinic workflow, and 2) map clinic flow to determine possible places to integrate the introduction, registration and follow-up reinforcement of SurvivorLink. Finally, a readiness for implementation tool and a fidelity checklist will be discussed and shared. Clinic staff will be allowed to ask questions during the training or submit questions at any time for technical assistance via email or phone calls. Clinic staff will also be trained on education and registration of users, distribution of the survivorship care plan (SCP) and uploading of each patient’s SCP into their SurvivorLink account, and user management and tracking of registration by their patients.

2.3.1. Intervention - SurvivorLink arm

Once randomized to the SurvivorLink arm, clinic directors will be contacted to schedule the initial training and will be sent the SurvivorLink toolkit. Each site will be individually trained to allow for tailoring for their current workflow.

To enroll survivors in SurvivorLink, clinic staff will verbally assess participants’ health literacy, and if they have access to the Internet (either computer or smartphone) [27]. After obtaining consent and assent as appropriate, clinic staff will use a script explaining the purpose of SurvivorLink and how to register for the website. Adolescent survivors (ages 15-17) who provide assent, will create a SurvivorLink account and their parent will then share their health record with them.

2.3.2. Control - Waitlist intervention arm

Clinics randomized to the waitlist arm will receive the intervention in Year 3 of the study. After the year of data collection from waitlist clinics, the waitlist control cancer clinics will then be individually trained and will implement the SurvivorLink program at their sites. In the waitlist clinics, survivors will be enrolled via a method that is similar to the one described above for the intervention clinics.

2.4. Data Collection Procedures

The sources of data for this study include: usage data automatically captured by SurvivorLink, patient- and parent-reported outcomes collected via surveys, and medical records abstraction. All patient and parent level data will be collected and managed via REDCap [28], and medical records abstraction. All consented parents and survivors in both the intervention and waitlist clinics will complete 3 surveys – at baseline (T1), 3-months (T2), and 12-months (T3). In Year 2, data will be collected from those who receive survivor care from a waitlist clinic. Data collection will include baseline, 3- months and 12-months follow-up surveys similar to the intervention group. The full set of surveys (with SurvivorLink questions) will be distributed in Year 3, during their intervention year. A subset of adolescents will also complete a 3- and 12-month survey regarding their use, feedback about SurvivorLink, and other self-management scales (e.g., readiness to transition to adult healthcare). Table 1 shows the primary and secondary measures for the effectiveness study.

Table 1.

Primary and secondary intervention effectiveness outcomes.

Primary Outcomes:

Validated 1-year follow-up survivor care visit from electronic medical

record (EMR)

Other follow-up tests and surveillance screenings from EMR

Secondary Outcomes and Measures:

Parent/Survivor’s patient activation

Patient Activation Measure (PAM) [30]

Changes in Parent/Survivor’s quality of life

Patient-Reported Outcome Measurement Information System (PROMIS®) Global-10 [31–34]

PROMIS® Scale v1.2 – Global Health [35]

Changes in Parent/Survivor’s self-efficacy to manage chronic disease

Adapted from Chronic Disease Self-efficacy Scale [36]

Changes in young adult survivors’ readiness for transition to adult-oriented care

29-item Readiness for Transition Questionnaire-provider version (RTQ Provider) [37]

Overall satisfaction with oncology care

25-item Pediatric Quality of Life Inventory® Healthcare Satisfaction

Hematology/Oncology Specific Module (PedsQL® Healthcare Satisfaction

Hematology/Oncology Specific Module) [38]

Barriers to follow-up care

What are reasons for your/your child missing the annual follow-up appointment?

Creation of question with responses from literature [39]

Clinic Adoption scales:

SurvivorLink readiness to implement scale [40]

Evidence-Based Practice Attitude Scale-36 (EBPAS-36) [41]

Context Assessment Index (CAI) (context, culture, leadership, evaluation [42])

Organizational Readiness to Change Assessment (ORCA) [43]

2.4.1. Objective 1. Measures of impact on survivor care and quality of life

The effectiveness outcomes of interest for this study consist of two broad categories: 1) one year follow-up cancer care visit and other utilizations (risk based tests/visits, general medical care) or recommended services to other sub-specialists (cardiology, nephrology, neurology, psychology) abstracted from medical records, constituting our primary endpoints, and 2) diverse domains of parents’ and survivors’ patient activation, quality of life, intentions to seek follow-up care, as well as perceived attitudes about the intervention. Institutional medical records will be reviewed to assess whether survivors attended recommended cancer survivor clinic one-year follow-up visit and completed other screening. Parent’s patient activation, QOL, and self-efficacy will be collected through the online patient self-report surveys. Covariates are the patients’ cancer and treatment history, demographics and technology use assessments.

2.4.2. Objective 2. Process measures of parents and adolescents

A process evaluation of the uptake of SurvivorLink will be conducted to measure use, participation, acceptability, perceived usefulness and satisfaction among childhood survivors and their parents. SurvivorLink automatically captures when a user registers, verifies their account, creates their health record, stores and/or access a health document and the total number of times a user logs in, these data analytics will be used to describe the reach of SurvivorLink (% of approached parents/patients who register). Participants will provide permission for study staff to monitor SurvivorLink site utilization during the study period through the dashboard and Google Analytics. Parents and adolescents will also be surveyed about SurvivorLink to assess uptake, intervention use, frequency participation, acceptability, perceived usefulness and satisfaction of SurvivorLink.

2.4.3. Objective 3. Measures of survivor clinic reach, adoption and integration (implementation outcomes)

Our primary implementation outcomes for this aim are reach, adoption and level of implementation of SurvivorLink by clinics from the RE-AIM framework. For reach, clinic staff will collect information from both the intervention and control clinics on who was approached, who participated and reasons for non-participation, and steps taken for registration on SurvivorLink. Socio-demographic information on all survivors will be obtained from clinic medical record review.

Adoption is defined as the number of clinics that establish a clinic portal in SurvivorLink and measure level of adoption (e.g., phases - formative to full implementation). We will create a composite score of completion of its 4 core elements as an index of implementation. At baseline, clinics will complete an organizational profile (e.g., size, late effects clinic services, age of survivors, etc.) and other clinic adoption scales. These same measures will be assessed at follow-up with the addition of the Sustainability Model at 12 and 24 months post-implementation [29].

The clinic implementation phase of the hybrid study will involve mixed methods data collection through surveys and interviews. Online surveys will be completed by the designated clinic Project Manager at baseline and 12-month follow-up, who will be instructed to consult with clinic records or other staff as needed to provide full and accurate responses. At baseline and 12-month follow-up, there will also be 1-hour in-depth interviews to assess the level of implementation with the Project Manager, a survivorship clinic provider, and another staff member closely involved in using SurvivorLink (may be a provider, social worker, nurse, or other staff position) for each site.

In our in-depth interviews, the CFIR will be used to explore the implementation process as well as constructs influencing adoption, implementation, and maintenance [44]. We will also draw from the Technology Acceptance Model (TAM) to identify specific intervention characteristics that may be particularly relevant for adoption and implementation of technology interventions such as SurvivorLink [45]. Interview questions will be adapted from prior interview guides; see Table 4 below for questions that map to selected CFIR constructs.

Secondary implementation outcomes will also be explored and focused on maintenance and sustainability of SurvivorLink. The timeframe for this study does not allow us to assess early maintenance beyond 2 years; we will focus on capacity for sustainability and plans for maintenance in Year 1 and its continuation in Year 2. Capacity for sustainability (RE-AIM) will be measured at 12 and 24 months through clinic surveys using the Program Sustainability Assessment Tool (PSAT), and the Sustainability Model [29, 46]. Additionally, our in-depth interviews will include questions about the clinic’s plans to maintain Survivor Link.

2.5. Data analysis

2.5.1. Health outcomes

We will use PROC GLIMMIX to compare baseline demographics in the intervention and the control clusters for both baseline and follow-up samples to assess how well the randomization procedure worked and conditional attrition. Next, we will use multilevel modeling to assess the intervention impact. Multilevel modeling allows to model both individual-level and clinic-level variances, predictors, and cross-level interactions simultaneously [47–49]. This allows for unbiased estimates of effects. For the main outcome of one-year follow-up cancer visit, we will assess the impact of the intervention using hierarchical binary logistic regression adjusting for individual and cluster-level covariates while accounting for unmeasured differences between clusters. Models will be built sequentially, adding first individual level covariates, then clinic level covariates, and finally the intervention assignment [47]. For secondary outcomes, multilevel models with the appropriate link function (i.e., linear for continuous outcomes, Poisson for count outcomes, binary for dichotomous outcomes) will be built similarly to those for the main outcome. Given the moderate number of clusters, Mplus will be used to calculate unbiased 95% confidence intervals [50].

2.5.2. Implementation outcomes

We will document the number of clinics that adopt SurvivorLink by establishing a clinic portal. Reach will be calculated as the proportion of eligible patients that register for SurvivorLink and effectiveness will be measured by the proportion of eligible patients that meaningfully use SurvivorLink (upload/store a document to their health record, download a stored document, and/or share their record with a registered provider). We will also assess which types of patients are more likely to register for or use it considering demographic characteristics (i.e., race), therapeutic exposure, patients vs. parents, and proximity to transition out of pediatric cancer center care.

We will conduct weighted clinic-level analyses (using t-tests), adjusting adoption, reach, summary implementation and sustainability scores due to estimated variances [51]. We will also use multilevel structural equation modeling (MSEM) to assess mediators to implementation (i.e., attitudes toward evidence, organizational readiness). Additionally, we will assess moderators to reach, level of implementation, capacity for sustainability and sustainability index based on organizational characteristics and inner/outer context variables. These data will inform important facilitators and challenges to implementation (in addition to the qualitative data on process) to discuss in training materials, webinars and future technical assistance with cancer clinics. HLM7 and Mplus 8 will be used to estimate the general multilevel and MSEM models. We will run descriptive statistics on process evaluation questions and compile open-ended comments related to users’ reactions, thoughts about the intervention and suggestions for improvements. We will assess RE-AIM and organizational-level outcomes focused on implementation and sustainability through the interview data that will be transcribed verbatim. NVivo 10.0 will be used for text coding of the transcripts and to facilitate the organization, retrieval, and systematic comparison of data across a range of dimensions (e.g., clinic size, geographic region, patient demographics). A codebook will be developed from the first set of transcripts, using CFIR. We will use standard content analysis methods to identify themes, as well as site-ordered matrices [52]. Qualitatively, we will use site-ordered matrices (ordered by overall reach) to identify patterns in level of implementation by selected CFIR constructs and how context affects level of implementation and sustainability (salient inner and outer setting variables). We will also “converge” data by creating tables with qualitative data side by side with quantitative findings [53, 54]. Salient themes will inform future changes to intervention content, technical assistance and training, and delivery in discussions with Advisory Board.

2.5.3. Power calculation

We conducted power calculations based on the following assumptions: A meaningful effect size of 10 percentage point differences between the two arms and 20% attrition. We also predetermined a cluster size of 75 patients per clinic as feasible. Furthermore, we assumed a conservatively high (especially for a matched-pair design) intra-class correlation of 3% [55, 56]. Close-form solutions are available for our design by Moerbeek and Teerenstra [55] and indicate the need for 20 clinics for adequate power for a small effect for follow-up visit based on our preliminary study (to detect mean difference of 10%) [57]. Thus, we determined the need for 75 patients each from 20 clinics yielding a total baseline sample size of 1500. Parallel, we determined that this sample size will also allow us to detect a small to medium effect size of d=.26 or larger for continuous outcomes, such as quality of life. One review has found small effects on psychosocial variables for SCP interventions to date [50].

3. Discussion

The Institute of Medicine (IOM) recommends all cancer patients should receive a survivorship care plan; however, scant research has explored how interventions which provide summary care documents have impacted a survivor’s utilization of health care, and adherence to surveillance guidelines [58]. Although web-based tools, including electronic personal health records (ePHRs) to facilitate long-term follow-up (LTFU) have been called for in the childhood cancer survivor literature [59, 60], very little is currently understood about ePHR use in this survivor population [61]. SurvivorLink functions as a web-based, patient-controlled communication tool for shared care. This ePHR system transcends the inability of different electronic medical record systems to exchange information, and allows cancer survivors to store and share their SCP and other health documents to promote and support delivery of evidence-based survivor care.

The number of visitors to SurvivorLink has been steadily increasing, and in 2017 alone Survivor Link had 1,526 unique visitors. While our recruitment efforts have been focused on the state of Georgia, we have had visitors from all 50 states and 116 countries. We currently have over 1,600 registered users including 719 caregivers/parents of survivors < 18 years, 622 young adult patients, 303 healthcare providers. Our team’s previous work has demonstrated that SurvivorLink™ can serve as an important bridge for young adult patients who transfer their care [57]. Young adult SurvivorLink registrants who transferred to adult care during an observational study of SurvivorLink usage were more likely to be meaningful users (uploaded or downloaded a document from their survivor health record and/or shared their record) (aOR: 2.6 (95% CI: 1.1, 6.1)) and used the ePHR twice as frequently as those who continued to receive care in our institution’s pediatric survivor clinic.

Our proposed research has numerous strengths. Importantly, it will contribute to implementation science by exploring context and other mediators and moderators of clinical practice change and by developing a clearinghouse of outer setting measures. Second, it will advance the field of pediatric cancer and offer the first hybrid effectiveness-implementation study to evaluate the impact of a technology-based, survivor health record and education system for pediatric cancer survivors and their parents. The achievement of these aims will help to promote evidence-based cancer survivor care to improve health and quality of life (QOL) of pediatric cancer survivors nationally.

Fig. 2.

Hybrid I effectiveness-implementation trial flowchart.

Table 2.

Consolidated framework for implementation research (CFIR) selected constructs for interviews.

Domain and Construct Sample Areas for Questions
Invervention characteristics	Flexibility/Adaptability of program, Complexity of use, Cost of staff time to deliver, Relative Advantage to clinic’s performance
Process	Engaging opinion leaders, Champions for process change, Executing for optimal workflow
Individuals involved	Knowledge, Beliefs, Efficacy and skills sufficient to implement SurvivorLink
Inner setting	Structures currently within clinic to deliver program, Climate-compatibility with current clinic practice, Leadership engagement in delivering program
Outer setting	Patient needs and resources, Barriers and facilitators to meet needs, Partnerships for survivorship care, External policies or incentives parent

Abbreviations:

CAI

Context Assessment Index

CFIR

Consolidated Framework for Intervention Research

COG

Cancer Oncology Group

COG LTFU Guidelines

Cancer Oncology Group Long-term Follow-up Guidelines

EBIs

Evidence-based interventions

EBPAS-36

Evidence-Based Practice Attitude Scale-36

EMR

electronic medical record

ePHR

electronic personal health record

IOM

Institute of Medicine

LTFU

long-term follow-up

MSEM

multilevel structural equation modeling

NCI

National Cancer Institute

PAM

Patient Activation Measure

PedsQL

Pediatric Quality of Life Inventory

PROMIS

Patient-Reported Outcome Measurement Information System

PSAT

Program Sustainability Assessment Tool

QOL

quality of life

RE-AIM

Reach, Effectiveness, Implementation, and Maintenance framework

RTQ-Provider

Readiness for Transition Questionnaire-provider version

REP

Replicating Effective Programs model

RTIPs

Research-Tested Intervention Programs

SCP

survivorship care plan

TAM

Technology Acceptance Model