Background Strategies that increase participant retention are critical to success in health research to ensure the validity and generalizability of study findings. However, reports of strategies to increase retention within the published literature are typically brief, descriptive and inconsistent if they are mentioned at all. Researchers can and should be applying evidence-based and theoretically-informed approaches to developing comprehensive retention plans early on in the research design phase. The reporting of “retention protocols” presents an opportunity to enhance the field of retention methodology by increasing transparency around effective retention practices through systematic reporting. The BC Healthy Connections Project (BCHCP) is a large-scale, long-term trial examining the effectiveness of the Nurse-Family Partnership (NFP) compared to existing services (usual care) in improving child and maternal health within the province of British Columbia. The BCHCP’s Scientific Team is developing and implementing a detailed retention protocol. This process requires a literature review to support the implementation of a comprehensive, evidence-based, theoretical framework for retention. Objectives This systematic review is intended to determine whether evidence exists of the application of evidence-based, theoretical approaches to retaining “hard-to-reach”, “vulnerable” populations within the published health literature. If they do, they will then be evaluated to determine their applicability to the BCHCP research context and whether they can be incorporated into the BCHCP retention protocol. Methods Seven electronic databases were searched systemically using a detailed search term strategy that included retention, attrition, and population terms in addition to any relevant compact vocabulary terms. Selection criteria for objective one included the following retention-specific conditions: had to be retention within a research context specifically, could not involve post-hoc strategies only, had to be applied to at least one follow-up point beyond enrolment or randomization, and selection of strategies had to involve some rationale. Studies also had to refer to the involvement of “hard-to-reach”, “vulnerable” populations as described for the purposes of this review. Publication limits were set for health-related research involving humans, published in English between January 1, 1980 and May 1, 2016. Objective two restricted eligibility further to studies that more closely matched the BCHCP study context in terms of study design and population as well as their approach to retention. Data items were collected in Microsoft Excel and were grouped under three main categories that included general study characteristics, retention characteristics, and follow-up characteristics. They were then presented in three summary tables (one for each category of characteristics) by full-text article according to assigned study ID. Counts for individual data items from each table were used to assist in interpretation of results. Results Of 1,337 original articles, 49 articles met objective one eligibility criteria. Of those 49, two met the original eligibility two criteria. Restrictions for criteria two were retrospectively lowered so that articles meeting three of the four original criteria were included. Eight articles qualified with adjusted criteria and were assessed for comparison to the BCHCP. The articles presented a diversity of study contexts with disparate lengths of follow-up, number of follow-up points, numbers and types of strategies used, level of description provided, and analyses of retention. The ways in which strategies were reported and assessed also differed substantially. However, few made mention of cost, and less than half made an effort to garner feedback, either from participants or research staff, on the research (and retention) process. While the majority of the articles reported some form of evidence to rationalize which retention strategies were selected for a given study, there was a wide variety in type and quality of evidence provided. Furthermore, only five articles reported studies that were explicitly informed by a theoretical approach to retention. While the majority did iterate the importance of planning for retention from the outset, only seven employed the term protocol in the description of their retention approach. Conclusions While retention strategies may be appearing more frequently within the health-related scientific literature, efforts to adopt the practice of developing comprehensive approaches to retention during the research design phase and to include plans to systematically evaluate and report on their outcomes are still missing. Moreover, examples of the practice of employing evidence-based, theoretically-informed, comprehensive retention approaches remain lacking. Certain organizations have taken the approach of encouraging researchers to conduct methodological research within the trial process–terms trials within a trial–in order to enhance the trial methodology evidence-base. The BCHCP retention protocol, on the other hand, presents a more comprehensive and systematic means for researchers to share an entire approach to trial retention. The adoption of retention protocols should therefore be encouraged as a necessary part of the trial publication process, similar to the implementation of systematic review protocols and trial study protocols.
Copyright is held by the author(s).
Member of collection