Abstract IA04: Data-driven recruitment: Implementation of digital marketing strategies for improved participant response rates and retention

Abstract

Abstract Background: Decline in epidemiology cohort response rates has been documented over decades, with precipitous declines in recent years. Incentives and repeated contact attempts have been shown to increase retention to varying degrees, but the effects of the different types of incentives and methods of contact—details necessary for reliable replication—remain poorly understood. Numerous factors related to study implementation and participant characteristics likely affect response rates, but few cohort studies have published the detailed empirical data that would be necessary to assess their effect. The small number of cohort studies that have published analyses of their recruitment protocols raises the question of how many studies are systematically and rigorously examining their methods. A 2011 review suggested that cohort studies may be less likely to engage in these activities due to concerns of risk to participant loyalty and the cost of assessment. Epidemiology cohorts often involve many concurrent activities and overlapping projects, which may also contribute to the complexity of the task. Considering the long duration of cohort studies, some may perceive that study staff can accurately intuit best practices. However, these challenges are not unique to epidemiology cohorts. The private sector is also concerned with attracting and retaining customers, and as such, companies have developed tools that enable rapid, low-cost reporting on their methods. In the field of marketing, customer relationship management (CRM) systems facilitate both the desired outcome (a customer buying a product) and the collection of data to identify factors that influenced the outcome. Aspects of these targeted marketing approaches and their related software tools provide both a model and a means by which epidemiology cohorts can better understand the constellation of factors that affect attrition among their own populations. Methods: The California Teachers Study (CTS) is a large prospective cancer epidemiology cohort that began in 1995. Since 2013, the CTS has used a commercially available CRM platform, modified with a variety of native applications (apps) and integrations, to manage participant recruitment and data collection for multiple projects. To manage a large-scale biobanking project and multiple tissue collection projects, we began with the platform’s standard contact data management and activity tracking features. We added an app for geolocation fieldwork management and a custom mobile app for surveys and biospecimen chain-of-custody tracking. To manage two follow-up questionnaires, we used available integrations for survey development, email marketing automation, a cloud-based call center with automated activity logging, and a web form for participants to submit information securely from our study website. Central features of all projects have included native reports and dashboards, workflow automations that reduce data entry, and data security measures. Results: The CRM platform has scaled to meet the needs of multiple concurrent projects with highly varied study designs and staffing requirements. By conducting our recruitment activities within the CRM, each interaction with the system (an incoming call from a participant, a blood collection, an automated survey reminder) passively generates useful metadata. These metadata, in concert with structured and unstructured data from participants and staff, enable timely monitoring and retrospective analysis using native reporting tools within the system. The ability to view and analyze clean data from the outset of each project revealed key—and sometimes unexpected—factors that were influencing response. Analysis of data across projects has revealed that characteristics of participants, staff, protocol decisions, technology, and equipment all influence outcomes in distinct ways. Conclusion: Existing commercially available tools, such as a CRM system, can be readily configured to meet the specific research, process, and security requirements of population sciences research. In the CTS, using a CRM system helped reveal that different factors affected participation differently, even within the same project. This presentation will discuss these factors, the approaches studies can use to collect data about these factors, and lessons learned about ways to further improve response rates. Broader adoption of CRM tools could enable the type of data collection and analysis necessary to more precisely define the factors that affect participation and develop more targeted and successful recruitment methods. Citation Format: Jennifer Benbow. Data-driven recruitment: Implementation of digital marketing strategies for improved participant response rates and retention [abstract]. In: Proceedings of the AACR Special Conference on Modernizing Population Sciences in the Digital Age; 2019 Feb 19-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(9 Suppl):Abstract nr IA04.