Journal:Adapting data management education to support clinical research projects in an academic medical center

Abstract

Background

Librarians and researchers alike have long identified research data management (RDM) training as a need in biomedical research. Despite the wealth of libraries offering RDM education to their communities, clinical research is an area that has not been targeted. Clinical RDM (CRDM) is seen by its community as an essential part of the research process where established guidelines exist, yet educational initiatives in this area are unknown.

Case presentation

Leveraging my academic library’s experience supporting CRDM through informationist grants and REDCap training in our medical center, I developed a 1.5 hour CRDM workshop. This workshop was designed to use established CRDM guidelines in clinical research and address common questions asked by our community through the library’s existing data support program. The workshop was offered to the entire medical center four times between November 2017 and July 2018. This case study describes the development, implementation, and evaluation of this workshop.

Conclusions

The four workshops were well attended and well received by the medical center community, with 99% stating that they would recommend the class to others and 98% stating that they would use what they learned in their work. Attendees also articulated how they would implement the main competencies they learned from the workshop into their work. For the library, the effort to support CRDM has led to the coordination of a larger institutional collaborative training series to educate researchers on best practices with data, as well as the formation of institution-wide policy groups to address researcher challenges with CRDM, data transfer, and data sharing.

Background

For over 10 years, data management training has been identified as a need by the biomedical research community and librarians alike. From the perspective of biomedical researchers, the lack of good quality information management for research data and an absence of training for researchers to improve their data management skills are recurring issues cited in the literature and a cause for concern for research overall. Similarly, librarians practicing data management have identified that researchers generally receive no formal training in data management yet have a desire to learn because they lack confidence in their skills.

To address this need, librarians from academic institutions have been working to provide data management education and support to their communities. By developing specific approaches to creating data management education, libraries have found successful avenues in implementing stand-alone courses and one-shot workshops, integrating research data management into an existing curriculum, and offering domain-specific training. Libraries have offered these training programs by providing general data management training to undergraduate and graduate students, doctoral scholars, and the general research community, whereas domain-specific data management can be seen most prominently in the life sciences, earth and environmental sciences, social sciences, and the digital humanities.

While it is clear that libraries have made inroads into domain-specific areas to provide training in data management, the clinical research community—clinical faculty, project, and research coordinators; postdoctoral scholars; medical residents and fellows; data analysts; and medical or doctoral degree (MD/PhD) students—is one that has not received much attention. Clinical research data management (CRDM), an integral part of the clinical research process, differs from the broader concept of research data management because it involves rigorous procedures for the standardized collection and careful management of patient data to protect patient [[Information privacy|privacy] and ensure quality and accuracy in medical care. The clinical research community understands the importance of data standardization, data quality, and data collection and has established good clinical data management practices (GCDMP) to ensure that CRDM is conducted at the highest level of excellence.

Despite this community-driven goal toward CRDM excellence, there is a dearth of literature about data management training for clinical research, with the only evidence coming from nursing training programs, whose research practices are further afield in that they focus on quality improvement rather than clinical investigations. This lack of evidence is surprising considering that the need for CRDM training has been communicated.

My library, located in an academic medical center, has supported CRDM through National Library of Medicine informationist projects by collaborating with clinical research teams to improve data management practices and, more recently, by serving as the front line of support for REDCap (an electronic data capture system for storing research data) by offering consultations and comprehensive training. Through REDCap training, I identified a need to expand my knowledge of CRDM to better support the needs of our research community. While REDCap is a tool to help researchers collect data for their studies, the majority of issues that our clinical research community encountered were related to data management. These issues included developing data collection plans, assigning and managing roles and responsibilities throughout the research process, ensuring that the quality of data remains intact throughout the course of the study, and creating data collection instruments. As this recurring thread of issues expanded the learning needs of our community beyond those provided via our REDCap training, I decided to expand my knowledge to address the questions that our researchers asked, to develop a curriculum to support CRDM, and to offer and evaluate CRDM training for our community.

Study purpose

This case study will discuss (a) the development and implementation of a 1.5-hour CRDM workshop for the medical center research community, (b) the results and outcomes from teaching the CRDM workshop, and (c) the next steps for the library in this area.

Case presentation

Workshop development

Gaining skills

Beyond the experience I gained from working closely with researchers on their clinical research projects and through REDCap support, I took two particularly valuable training opportunities that improved my skills in CRDM: the “Data Management for Clinical Research” Coursera course and “Developing Data Management Plans” course offered through the online educational program sponsored by the Society for Clinical Data Management. These two courses provided me with the knowledge that I needed to teach a CRDM workshop but more importantly gave me the confidence to teach it because they provided a depth of knowledge I did not have before. These courses also served to reinforce that the issues and challenges encountered at my own institution were common data management concerns across the broader clinical research community.

Identifying core competencies and building workshop content

The primary focus for developing a 1.5-hour CRDM workshop was to use the GCDMP core guidelines as the baseline structure for the workshop. The core guidelines are separated into chapters in the GCDMP, which were used as the foundation for the core competencies of the workshop. Once this baseline structure was established, my goal was to weave in answers to the common questions that our clinical research community has asked through our existing REDCap training. These questions related to how to create codebooks and data dictionaries for research projects, how to structure roles in a research team, how to use best practices for building data collection instruments, how to protect their data according to Health Insurance Portability and Accountability Act (HIPAA) regulations that they should be aware of, how to improve the quality of their data throughout a study, and how to best document procedures throughout a study.

The goal of the workshop was to tie as many examples back to REDCap as possible, because the use of REDCap was written into institutional policy as the recommended tool for research data collection, which made it essential to highlight its data management capabilities. The core competencies combined with the questions mentioned above served as the foundation for developing the learning objectives and interactive learning activities for the workshop (Table 1).

Table 1. Clinical research data management workshop core competencies Core competency Learning objectives Interactive learning Data collection planning ▪ Plan a data collection work flow ▪ Document tools and resources used for data collection ▪ Connect study protocol to data collection plan ▪ Describe study goal ▪ Write down first five steps of the data collection plan ▪ Communicate with partner(s)/team to identify gaps Data collection instrument design ▪ Describe data collection best practices ▪ Identify common data collection risks and pitfalls ▪ Review data collection form and identify errors ▪ Revise data collection form to collect data according to best practices Data standards utilization ▪ Define data standards ▪ Describe the benefits of using data standards for research ▪ Locate data standards for use in research study ▪ Navigate the terms of use for specific data standards ▪ Search for relevant data standards in the REDCap Shared Library, National Library of Medicine, and FAIRsharing.org ▪ Explain the terms of use for the chosen data standard Data quality maintenance ▪ Describe the importance of using data quality measures in a clinical research project ▪ Implement data quality work flows using REDCap ▪ Develop a data quality plan for an existing or prospective research project ▪ Implement the Data Resolution Workflow feature in REDCap Data storage, transfer, and analysis best practices ▪ Identify institutionally supported data storage and transfer software ▪ Identify the components of a statistical analysis plan ▪ Describe the documentation needed to perform a successful data transfer ▪ Select the appropriate tool for data storage and transfer based on different scenarios Role and responsibility management ▪ Describe methods for ensuring that roles and responsibilities are clearly assigned ▪ Develop documentation for past, current, and future roles ▪ Assign roles for different project personnel using REDCap ▪ Describe methods used to assign roles with partner(s)/team

The core competencies and learning objectives were designed to make the workshop as practical as possible. While the theoretical components of CRDM are important and are emphasized in the workshop, the main focus was to consistently incorporate interactive learning throughout so that attendees could both apply and contextualize what they learned to their own research. Another goal of this workshop was to encourage communication between attendees to highlight common CRDM errors and provide avenues for attendees to learn about successful and unsuccessful approaches from their peers. To this end, after each core competency was taught, the workshop was designed to have attendees discuss their own experiences.

In addition to the core competencies listed in Table 1, the overarching theme and intention applied across the workshop was the importance of maintaining good documentation throughout a clinical research project (e.g., data collection plan, roles and responsibilities documents, and statistical analysis plan). By stressing the importance of documentation for each competency, I hoped that attendees would understand the value of and be able to develop their own detailed documentation at each stage of the research process. The time dedicated to developing this workshop—which included reviewing the GCDMP core competencies, outlining commonly asked questions from the research community, establishing learning objectives, building the slide deck, and creating the workshop activities—took between 80 and 100 hours to complete.

Notes

This presentation is faithful to the original, with only a few minor changes to presentation. In some cases important information was missing from the references, and that information was added.