IMPROVING RECRUITMENT AND RETENTION OF ADOLESCENTS AND YOUNG ADULTS WITH CANCER IN RANDOMIZED CONTROLLED CLINICAL TRIALS

Abstract

Participation of adolescents and young adults (AYAs) with cancer in randomized clinical trials (RCT) is necessary to advance treatments and psychosocial programs. Exploring AYA experiences in a RCT will inform strategies to support recruitment and retention. A qualitative design was used to study the experiences of 13 AYAs in the SMART I trial. Key themes included: Weighing the Pros and Cons; Randomization Preferences; Completing Measures; and Worthwhile Experience. The experiences of AYAs during RCTs can bring insights that inform the design and management of AYA trials. Strategies include improving assent/consent processes, design of electronic interfaces, and encouraging researcher flexibility.

Background

Survival rates for adolescents and young adults (AYAs) with cancer have not kept pace with improvements in outcomes for children and older adults.^1–4 In addition, AYAs experience significant long-term consequences resulting from their cancer treatment including decreases in emotional and mental health, fertility, and health-related quality of life.^5–9 Participation of AYAs in research is necessary to advance treatments and psychosocial support programs for this unique age group;^10–15 however, fewer AYAs enroll in randomized clinical trials (RCT) than children or older patients.^{4,10,13,15–17} Although reasons for lower AYA enrollment include logistical factors like RCT availability, there is growing evidence that modifiable AYA and investigator factors including negative AYA perceptions of research, and recruitment strategies lacking in developmental considerations are also responsible.^15–25

Empirical study of AYAs’ experiences during RCT has focused almost exclusively on recruitment and consenting processes.^{18,23,26–29} A more in-depth exploration of the experience of AYAs across the entire RCT process is lacking. For example, while it is clear that trust contributes significantly to adolescents’ decisions about research participation,^{18,27,29–34} establishing and maintaining trust across the RCT can be challenging related to factors such as random group assignment, measurement completion and additional clinical visits .^35–38 For AYA undergoing cancer treatment, these additional activities may exacerbate feelings of lost control, fatigue, and burden, that can result in study withdrawal.³⁹ Understanding AYAs’ experiences across an RCT is essential to developing trials that encourage enrollment and retention. This article reports findings on AYAs’ experiences across a behavioral health RCT and discusses strategies to support AYA participation.

Methods

Stories and Music for Adolescent and Young Adult Resilience during Transplant (SMART I) was a RCT of a therapeutic music video intervention funded by the National Institutes of Health (R01 NR008583) and the Children’s Oncology Group (ANUR0631, U10 CA098543). The primary aim of SMART I was to test the efficacy of a therapeutic music video (TMV) intervention versus a low-dose control group on resilience and quality of life in AYA undergoing stem cell transplantation. Participants were recruited from eight Children’s Oncology Group institutions. Trial design and outcomes were guided by Haase’s Resilience in Illness Model and the intervention structure and content were guided by Robb’s Contextual Support Model of Music Therapy.^40–42 Primary findings indicated that AYA who received the TMV intervention reported significantly better courageous coping immediately post-intervention, and significantly better social integration and family environment at 100 days post-transplant.⁴³

The secondary aim of the SMART I was to describe AYA perceptions about the meaningfulness and helpfulness of the TMV intervention in coping with the distress associated with stem cell transplantation. A stratified purposive sample of 13 AYAs (4 female; 10 white, 1 black, 2 more than one race) from the six RCT sites, ages 13 to 22 years (mean age = 17 years) participated in open-ended interviews.

Trained study personnel conducted the interviews either privately in-person or over the telephone. The primary, data-generating statement, “Please tell me about your experiences of being involved in the music video study”, encouraged AYAs to discuss thoughts, feelings, and experiences related to the study, and was followed by probes to deepen reflections. Interviews ranged from 6 to 24 minutes (mean length = 15 minutes). Haase’s adapted version of Colaizzi’s method was used to analyze data by: (1) gaining understanding of the experience as a whole by thoroughly reading interview transcripts; (2) identifying significant statements; (3) restating significant statements in general language; (4) formulating meaning of significant statements.^44,45 The multi-site study team worked together via web-based conferencing to: (5) reach consensus on the formulated meanings, and (6) organize the meanings into theme clusters and categories to develop the essential structure of the AYAs’ experience. Trustworthiness procedures used throughout data collection and analytic phases of the study enhanced credibility, dependability, conformability, transferability, and authenticity⁴⁶ of the findings. Procedures included: a) data collection -interviewer training, transcript accuracy checks against audio recordings. b) data analysis - team members analyzed transcripts first in dyads, then presented preliminary meanings to full team at monthly analysis meetings, c) checking for representativeness of the meaning units to the data set as a whole⁴⁷, d) face validity checks by analysis team that included five advanced practice nurse clinicians.

Results

Four theme categories describe the AYAs’ experiences of being in a controlled trial: (1) Weighing the Pros and Cons of Participation; (2) Randomization Preferences—Dealing with Disappointments; (3) Completing Measures—Long, Repetitive, Worth It; and (4) An Overall Worthwhile Experience.

Weighing the Pros and Cons of Participation.

Prior to consenting, AYAs experienced uncertainty about study expectations and they thoughtfully considered the overall benefits and options before making their decision to participate. They considered their group assignment options and the general benefits of having something to do. “I thought it would help me cope with the situation and it…. and not have to worry about all the other things going [on]” (21-year-old male). Helping others was also considered, “being … part of a [study] is pretty cool, helping out with other kids in the future” (13-year-old male). Knowing withdrawal was an option was helpful when uncertainty persisted. “I thought, okay, I’ll try it. If I really don’t like it then I could drop out” (16-year-old male).

Randomization Preferences—Dealing with Disappointments.

AYA often had preferences for group assignment to either the low-dose control audiobook group, “At first I was hoping to be in the [books] group” (21-year-old male), or to the TMV group, “…if I didn’t get the video I really wouldn’t have been very enthusiastic…” (15-year-old male). When assigned to their second choice, AYA either found their assignment was better than expected, “At first I wanted to do the audiobook but now I don’t think that would have been as fun…” (16-year-old male); made the best of their assignment, “It was just books on tape, so I could have done whatever I wanted while I was listening” (15-year-old male); or they continued based on a sense of obligation, “I’d say it [lack of enthusiasm] was kind of the whole time. … I was hoping to be in the other group…I felt a little bit obligated” (21-year-old male).

Completing Measures—Long, Repetitive, Worth It.

Measures for the SMART I trial were completed online using laptop computers.⁴⁶ Overall, the electronic format, which included age-appropriate and ethnically diverse graphics, encouraging messages, and scheduled audio-alerts to mark progress and cue breaks, and provided an appealing and flexible interface: “…the laptop was a lot better than using pen or paper…very interesting questions and it made me think…It wasn’t boring at all” (21-year-old male). However, some AYA found the measures to be long and sometimes repetitive, which dampened their enthusiasm. “They were pretty long….It seemed like a lot of the questions were repetitive.” (20-year-old female). Some expressed frustration with having forced-choice answers that did not adequately capture their experience, “Some …answers they had didn’t…reflect my answer. … some responses were too broad for the answer I had …. so I had to just pick…” (13-year-old female). Some AYA saw measure completion as a task that needed to be done, “…[it] was just kind of [like] taking a test.” (21-year-old male) or a “…means to an end.” (13-year-old male).

Despite the length, many AYA found the questionnaires were meaningful and as such their value outweighed the time required to complete them, “…was kind of long but it …talked about a lot of ways you could actually…cope with being sick and, what you use to pull through…” (21-year-old male). Many AYA described how completing the measures caused them to reflect on their cancer experience, their strengths, and sources of support they could use during transplant. Comments included, “…there were some things I didn’t really think about until filling out the questionnaire.” (21-year-old male); “It opened up my eyes to different stuff, like the questions asked, ‘would you share something with your mother and father, or your friends; someone you could trust’. It just made me think.” (16-year-old male); and, “it reminded me again that I do have people who care about me, and my family really is there for me and they’re trying their best through this experience…” (20-year-old female).

An Overall Worthwhile Experience.

Overall, AYA experienced participation in the SMART I RCT as a worthwhile experience they would recommend to other AYA, “I advise everybody my age to do it because it does, it does pass the time and bringing my mind off all the bad things going on.” (21-year-old male). AYA saw study benefits of developing new relationships and having something to help pass time when in the hospital, “I liked the attention because sure, you’ll have visitors and stuff like that, but just having the SMART I study people come and take a survey of you was kind of nice.” (21-year-old male). They expressed appreciation for study personnel’s’ efforts to be flexible and understanding when symptoms interfered with their desire to engage in study activities. AYA appreciated the genuine desire of study personnel to support AYA beyond study activities. “You all, like the [music therapist] and all them [study personnel] … they helped me through it [the transplant]. Thank you all for helping me through it.” (21-year-old male).

Discussion

The findings support the importance of researchers building trust and honoring AYAs’ autonomy throughout the research process.^{18,31–33,48} These goals are supported by well-conducted, thorough assent and consent processes that go beyond general details and provide a deeper understanding of what to expect during the clinical trial experience. Johnson’s self-regulation theory of coping with health care events indicates that individuals cope in ways that are consistent with their understanding of an experience, and that providing concrete preparatory information, that addresses potential emotional responses, facilitates coping with distressful events.^49–53 Extending self-regulation theory into RCTs for AYA is instructive to the informed consent process. Including disclosure of time required for measure completion and providing a description of potential negative emotional responses to study activities (i.e. frustration filling out long measures) is likely to improve AYAs’ preparedness to deal with study activities and enhance trust.

Strategies to support AYA autonomy include the flexibility of researchers in their interactions with AYAs and last-minute scheduling changes. Flexibility is critical when conducting research with ill adolescents who experience unpredictable symptoms or treatment scheduling changes. In addition, AYAs’ decisions to participate in a RCT hinge on their knowledge of their right to withdraw at any time, for any reason. While such knowledge supports AYAs’ autonomy and builds trust, it also requires researchers to identify ways they can be flexible within the RCT protocol, encourage AYA to express any hesitation or reluctance to complete study activities, and consistently assess AYA for signals indication their desire to withdraw.

There is evidence that additional time and activity requirements undermine AYAs’ participation in research.^23,29 This analysis highlights how researchers can minimize the burdens of completing study activities. AYA frustration regarding measures can be addressed by: (a) advising about the potential for emotional frustration, (b) emphasizing the importance of the measures and providing reasons for having seemingly repetitive questions, (c) providing AYA with real-time, visible tracking of their progress while completing measures, and (d) including free-text comment fields so AYAs can indicate when they feel constrained by response options.

AYAs in the SMART I expressed both positive and negative experiences of trial participation. Some AYAs found the frustration of completing measures was offset by the meaningfulness of the research experience, which included the opportunities for self-reflection and forming supportive relationships with researchers. AYAs’ initial preconceptions about participation in research often changed during the trial. Recognizing the dynamic nature of the AYA RCT experience and seeking opportunities to enhance positive and meaningful aspects will likely improve AYA recruitment and retention rates in RCTs.

Findings reported here resulted from a qualitative analysis that was nested within a behavioral health intervention RCT, and not a separate study examining AYA RCT experiences. Participant experiences with a behavioral intervention are likely different than experiences with a drug-based RCT due to the intervention type (drug vs. psychosocial) and outcome measures (e.g., patient-reported outcomes, psychosocial self-report). However, the themes discussed here have relevance to most RCTs involving AYAs, as they tackle common challenging research processes such as assent, consent, and attrition.

Conclusions

The RCT experiences of AYAs with cancer provide insights that inform the design and management of future AYA trials. Improving assent and consent processes, as well as the design of electronic measure interfaces, and encouraging researcher flexibility may facilitate AYAs’ participation and help overcome disparities in outcomes for AYAs with cancer.