Recruitment and Retention Strategies Used in Dietary Randomized Controlled Interventions with Cancer Survivors: A Systematic Review

Simple Summary The purpose of this systematic review is to evaluate the quality of recruitment and retention methodologies in the context of diet-related intervention trials among cancer survivors. Findings suggest investigators are meeting reporting guidance for recruitment; however, reporting of retention methods and rates is less consistent, raising concern as to the interpretation of study findings. There is a need for researchers to consistently report retention methods and rates to inform best practices and enhance the rigor of future diet intervention trials in cancer survivors. Abstract Background: The purpose of this review was to systematically evaluate the quality of reporting of recruitment and retention methods in diet-related intervention trials among cancer survivors. Methods: A systematic search of five databases in Spring 2023 identified dietary intervention randomized controlled trials with a minimum of 50 cancer survivors, an intervention of at least eight weeks, and at least six months of study duration. Outcomes investigated include methodologic description and reporting of recruitment and retention rates. Results: Seventeen trials met inclusion criteria. Recruitment methods included cancer registry and clinician referral, hospital records, flyers, and media campaigns, and were reported in 88.2% of studies. Eleven of 17 studies (64.7%) met a priori recruitment goals. Eleven studies identified an a priori retention goal and seven met the goal. Retention goals were met more often for studies of less than one year (71.4%) versus greater than one year (50%), and for studies with remote or hybrid delivery (66.7%) versus only in-person delivery (50%). Conclusions: Recruitment goals and methods are frequently reported; reporting of retention methods and goals is limited. Efforts are needed to improve reporting of retention methods and rates to inform best practices and enhance the rigor of future dietary intervention trials.


Introduction
It has been estimated that as of 2022, over 18 million people in the United States were cancer survivors, and this number is expected to surpass 22 million by 2030 [1].Cancer survivors are at an increased risk of second primary cancers along with other diet-related chronic illnesses such as diabetes, osteoporosis, and cardiovascular disease [2].Incorporating healthy eating behaviors plays a key role in reducing comorbidities, improving overall health, promoting a longer lifespan, and supporting higher quality of life for cancer survivors [2].As the population of those who have lived with cancer expands, leading organizations have proposed guidelines for dietary behaviors to promote survival, reduce co-morbidity, and increase quality of life.The World Cancer Research Fund (WCRF) and the American Institute for Cancer Research (AICR) recommend maintaining a healthy weight during adulthood as well as consuming a diet with a high daily intake of whole grains, vegetables, fruit, and legumes, and restricting consumption of sugar-sweetened beverages, processed foods, red meat, and alcohol [3,4].The American Cancer Society (ACS) and American Society of Clinical Oncology (ASCO) provide similar recommendations for cancer survivors with additional advice to limit saturated fat intake [5,6].
Since diet is a factor in morbidity and mortality outcomes among cancer survivors, numerous interventions have been conducted to enhance adherence to dietary guidance for survivorship [7][8][9][10][11].Several published reviews acknowledge the beneficial effects of these dietary interventions on health outcomes for cancer survivors [7][8][9].To ensure the scientific rigor of these trials, adequate and appropriate methodologies and transparency of reporting must be followed [12].Two of the important aspects that impact the validity of study results are the recruitment and retention of participants in clinical research.In fact, concerns related to the appropriate interpretation of findings from randomized controlled trials arise when recruitment (accrual) falls below statistically powered estimates, or when drop-out rates (attrition) exceed a priori estimates [13].Differential recruitment or dropout rates by study randomization assignment is an additional concern that reduces trial integrity and rigor [14,15].Reporting of a priori sample size is a necessary step for other researchers to determine if the study reached recruitment and retention goals and thus amply tested the proposed hypothesis.
Recruiting adequate participants to randomized controlled trials and retaining them for the duration of the intervention requires coordinated and strategic efforts.Some methods of recruitment include searching cancer registries for potential participants and sending invitation letters to those eligible, promoting the study through in-person and online networks, and advertising through media outlets, including traditional television, radio, and print [16].Retaining participants in clinical trials can be difficult, often related to burden on the participant.Burden may include frequency of measurements, travel time and cost to attend study visits, time required to adhere to the intervention, time to prepare food and measure for self-reported data, time to perform self-assessment (tracking food intake or steps), family time commitments, and the social burden with family and friends not participating in the intervention [8,[17][18][19].It is crucial that researchers report the reasons why participants exit the study, thus providing key insight to inform the balancing of participant burden with research rigor in future research trials.The application of CONSORT (CONsolidated Standards of Reporting Trials) diagrams to randomized controlled trial outcome publications is one step journals have taken to increase transparency in reporting participant engagement throughout a trial [20].
The purpose of this systematic review is to evaluate the quality of reporting in the context of diet-related intervention trials among cancer survivors and to describe recruitment and retention methods previously reported.Our focus is on published randomized controlled dietary intervention trials among four cancer survivor populations: breast, prostate, colorectal, and lung and bronchus.This review focuses on accrual and attrition outcomes of these trials, and descriptions of recruitment and retention methods, offering insight into approaches that may enhance reporting and promote optimal recruitment and retention in dietary trials among cancer survivors.

Aims
The aims of this systematic review are to (1) describe recruitment and accrual rates and methods used to recruit cancer survivors into randomized, controlled dietary intervention trials, and to (2) describe retention and attrition rates and methods used to promote retention of cancer survivors enrolled in randomized controlled dietary intervention trials.

Methods
The conduct and reporting of this systematic review adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [21].The protocol was registered with PROSPERO, the International Prospective Register of Systematic Reviews (CRD42018070396).

Search Strategy
The following five electronic databases were searched for relevant articles in January 2023: PubMed, CINAHL, Cochrane Central Register of Controlled Trials, Embase, and PsychINFO.Search strategies were developed with assistance from a librarian and adapted for each database using standardized terminology.The search string was developed for PubMed and then translated to the other databases.Keywords or Medical Subject Heading (MeSH) terms included in the electronic database search comprised (1) diet, (2) cancer survivors-specifically breast, prostate, colorectal, lung and bronchus, and (3) randomized controlled trials and controlled clinical trials (see Supplementary Figure S1 for example of the search strategy).Filters included human adults and peer-reviewed articles published between 2013 and 2023 in English.

Eligibility Criteria
This systematic review evaluates dietary behavior change and diet-inclusive weight loss intervention randomized controlled trials, including pilot and feasibility studies with a minimum of 50 participants at the time of randomization, at least 8 weeks of active intervention, and at least 6 months of total expected participation.Eligible studies included adult (18+ years old) breast, prostate, colorectal, and lung and bronchus cancer survivors.Studies with other cancer types were allowed if one of these primary cancer types was included in the study.Outcomes investigated in this review include recruitment and retention methods and rates, which required the presence of a CONSORT flow diagram of participant engagement numbers, attrition rates, and reasons participants discontinued participation in a trial [20].Trials still in active recruitment were ineligible.

Study Selection and Data Extraction
A single individual (SW) performed the database search, and all citations were exported to EndNote X9 for data management.Duplicate citations were removed following the process described by Bramer et al., 2016 [22].Two reviewers (SL, SW) then independently dual-screened the titles and abstracts of all publications to assess eligibility criteria.Articles that did not meet the inclusion criteria were excluded, and any articles for which there was a discrepancy between reviewers moved on to full-text screening for further review.Full-text articles were screened independently by two reviewers (SL, SW) to assess eligibility.Non-relevant articles were excluded, and conflicting votes were resolved by a third reviewer (CT).
Data were extracted independently by SL and SW using a data extraction form developed by SL, SW, TC, and CT.SL and SW screened the full-text articles and reference lists for protocol or design articles for the selected studies.Data collected from each study or protocol article included first author, study title, publication year, country where the study was conducted, study name, study design, cancer type and stage, a priori sample size, intervention type, mode of delivery, length of intervention and follow-up, methods of recruitment and retention, time necessary for recruitment, number recruited, number screened, number eligible, number randomized, and number retained.Retention rate was calculated as the percent of the number of participants who completed study follow-up divided by the number randomized.Attrition rate was calculated as the percent of the number of participant withdrawals divided by the total number randomized.

Recruitment Results
Of the 15 studies that reported their target accrual, eleven (73.3%) met or exceeded their recruitment goals [10,26,28,29,32,33,35,37,39,40,42].The four (26.7%) studies that did not meet their recruitment goals [23,30,31,36] identified lack of adequate accrual time, challenges with ineligibility criteria, and the loss of funding or the end of the funding cycle as reasons for not meeting their accrual goals.Two studies did not report their target accrual goals [25,34].Overall, studies reported a lack of interest as the primary reason potentially eligible participants did not enroll in the study, followed by lack of access to transportation for in-person sessions or assessments, and lack of time.
The average target sample size for the eleven studies which met their a priori accrual goals was 212.4 ± 192.9 participants and 674.7 ± 985.2 participants for the four studies which did not meet accrual goals, though this difference was not significant (p = 0.42).Accrual time across studies ranged from three months to four years, with the average accrual time being 19.9 ± 11.8 months.There was no difference in average accrual time between studies which met their target accrual goals and those which did not.
Commonly reported reasons for participant attrition included loss to follow up or noncompliance, lack of willingness to be wait-listed for the intervention or join a control arm, primary outcome event (i.e., death, recurrence, new disease), and inability to attend inperson sessions or assessments.Retention methods included regular study contact through text, phone call, or email (47.1%) [10,28,30,[33][34][35]37,38,40]; participant compensation or travel allowance (23.5%) [10,24,31,36]; personalized coaching or lifestyle recommendations (23.5%) [27,32,37,42]; and flexible scheduling (11.8%) [10,31].Consistent contact was the most commonly employed method by studies achieving retention goals.Compared to studies which did not meet their identified retention goals, studies that did meet their retention goals more commonly included delayed delivery of intervention materials to the control group, rapport building and personalized messaging, or tailored materials.

Discussion
This review highlights recruitment and retention of cancer survivors in dietary randomized controlled trials published over the past decade.In general, reporting of recruitment methods and study target accrual was high.Although all studies reported overall participant attrition rates, a key finding of this review is the overall limited reporting of protocol-specified participant retention goals (30% of trials provided) and related retention methodology used to achieve retention goals.Of note, trials of shorter duration and with remote or hybrid intervention delivery more commonly reported achieving recruitment goals and maintaining high retention.This has implications in terms of sustaining engagement in lifestyle behavior programs, a factor that will drive long-term health promotion, and related oncology care outcomes in cancer survivors [43].
In terms of recruitment strategies, oncology clinic-based recruitment was the most common approach.This has implications in terms of the timing of interventions as cancer survivors are encouraged to move back to primary care within five years of treatment [44]; thus, recruitment in oncology care may bias early survivors and promote recruitment for individuals with higher access to care.This may indirectly contribute to inequities in access to these trials and related lifestyle health coaching programs and is a problem that has been noted in previous work.In fact, the lack of diversity in trials has been a major impetus for the community outreach and engagement role in comprehensive cancer center care [45], wherein the National Cancer Institute called to action the need for broader catchment area enrollment in trials that represent the diversity of people with cancer to reduce cancer health disparities [46].Current data suggest diversity is not achieved in most oncology research [47].Similar to previous work, here we show that every trial recruiting across race and ethnic groups resulted in over 82%, on average, being non-Hispanic White (NHW) race and ethnicity, despite the higher cancer mortality rates and likelihood for diagnosis with advanced disease in Blacks and Hispanics compared to NHW [1].Further, Black and Hispanic cancer survivors, as compared to NHWs, report lower adherence to dietary guidance for cancer survivors [48,49], supporting a greater need for programming focused on attaining a healthful diet.
Our results highlight the need for more transparency in reporting retention goals and methodologies in lifestyle behavior oncology research.We identified that remote-or hybrid-delivered interventions are most likely to meet retention goals.Similar to what has been concluded in previous work, this suggests continued efforts to develop effective programming using web-based and telephone resources [50].Additionally, studies which reported protocol-specified retention goals and strategies were less likely to see differences in attrition by study arm compared to those that did not report.Importantly, 80% of the studies which did not report a priori retention goals saw greater attrition in the behavior change condition compared to usual care or control.This is compared to the studies which did report retention goals, of which less than 10% saw greater attrition in the intervention arm.Differential drop-out rates between study conditions have implications for trial rigor and validity, potentially biasing study results and therefore dietary recommendations for survivors [15].Systematic and protocol-driven retention methodology should be developed early in the planning stages of a randomized, controlled trial and reported clearly to support interpretation of trial results.
Several resources and recommendations for improving recruitment and retention in clinical trials are available.The NIH-funded Recruitment Innovation Center was established in 2016 to help increase enrollment of diverse participants in clinical trials and offers support to investigating teams to ensure timely completion of trials and with statistical power to make valid recommendations for care [51].To overcome barriers for trial participation, the community-engaged recruitment model [52] and the patient-centered recruitment and retention model [53] have been developed.Suggested approaches include cultural tailoring of intervention materials, consideration and adaptation for participants with geographic barriers, offering digital or remote alternatives to program implementation, and providing travel compensation and flexibility [54,55], which our results supported as effective in achieving enrollment and retention goals.Building community partnerships is a welldescribed and relevant strategy to improve accessibility to trials [54,56], including diet trials in survivors, which was not well utilized in the included studies, as only one study reported recruiting at community-based events [39].
This review has several strengths, including being the first to systematically address recruitment and retention in completed dietary-controlled trials for cancer survivors.The systematic approach to trial selection and data collection limits the risk of bias.Study eligibility criteria for this review captured trials with moderate or high levels of research rigor (i.e., randomized trials, greater than 50 participants at randomization, at least six months of follow-up, etc.).Our results inform effective strategies to improve recruitment and retention that are applicable for advancing trial participation in future diet behavior change programming and in turn supporting completion of more rigorous and evidencebased clinical trials.
With a limited number of trials included and from a wide range of geographic locations, comparisons between recruitment and retention methodology must consider heterogeneity in trial contexts and settings.A potential limitation to these data is publication bias.Studies with statistically significant findings are more likely to be published than those with non-significant or negative findings or those which were ended early due to futility, a potential result of low enrollment or high attrition in a study [57].Another potential threat to the validity of study results and a limitation for our study is healthy subject bias.Participants experiencing greater symptom or comorbidity burden related to their cancer treatment may be more likely to drop-out or be lost to follow-up, biasing trial results towards the healthiest participants or those with greater access to resources for cancer-related care [58].Future research should also explore other factors that may impact survivor engagement and adherence to an intervention such as individual affect, cultural and social influences, dietary preference, and age.Clear reporting of patient-reported outcomes is necessary to understand if attrition and resulting missing data are related to poorer outcomes, methodological issues, or lack of interest in the intervention [59].

Conclusions
Achieving and retaining study participants in dietary intervention trials among cancer survivors are important to the interpretation of study findings.The most common recruitment methods utilized for studies which achieved their a priori accrual goals included clinical recruitment and provider referral or recruitment from a cancer registry.Studies with a duration of six months or less, and studies with remote or hybrid delivery models were most likely to have met their accrual goals.Our findings suggest investigators are meeting reporting guidance for transparency in study recruitment; however, reporting of retention methods and rates is less common and inconsistent, raising concerns as to the interpretation of study findings.Studies which met their identified retention goals most frequently reported regular contact and interaction between study staff and participants.This indicates that rapport building is an important factor for survivor engagement and continuation with a trial.Efforts to improve retention reporting should be enhanced, especially for trials of longer duration.

Table 1 .
Study characteristics, recruitment, and retention data for randomized, controlled dietary trials conducted with cancer survivors over the past ten years (n = 17).