More sponsors, including the NIH, are requiring Data Management Plans (DMPs) as part of the application process. Researchers with little or no experience developing DMPs may be curious about how to construct a good plan and could benefit from learning from their colleagues. To highlight researchers who have derived impact or benefit from sound data management practices, the Research Data Initiative (RDI) within the Duke Office for Research & Innovation is launching the Data Management Exemplar Series. If you or someone you know is a data management exemplar, please connect with us so we can continue to foster a research community fluent in strong data management practices.
Our interviewee today is Ryan Shaw, Ph.D., R.N., an associate professor in the School of Nursing and former faculty director of the Duke Mobile App Gateway in the Duke Clinical and Translational Science Institute. Ryan frequently uses data management plans (DMPs) and describes how DMPs have benefited him as well as his take on how DMPs could benefit others.
"Can you tell us a little bit about your research area? Describe what researchers/research teams look like in your discipline?"
"I am a researcher in Digital Health and Director of Duke School of Nursing’s Health Innovation Lab. Adjacent to Duke Hospital, the lab provides a simulated clinical space for product development, modeling, and testing new care delivery processes. I discover how patient generated health data and digital health technologies can be integrated into care delivery models to manage and improve health outcomes and promote health equity for patients with complex chronic illnesses. This includes data from wearables, apps, sensors and devices that monitor and augment patient care.
I have led a number of clinical trials including a current NIH-funded telehealth trial in which we integrate multiple streams of real-time remote monitoring data into Duke’s electronic health record (EHR). In this study, a patient is prescribed a suite of devices in Epic (Duke’s EHR) to self-monitor their uncontrolled diabetes and hypertension (i.e., blood sugar, blood pressure, weight, and activity) and the data is captured in their health record. This is coupled with team-based telehealth in which nurses deliver remote care to patients in-between in clinic visits and pharmacists provide medication management.
Research teams in my area consist of clinicians, behavioral scientists, data scientists, computer programmers, database developers, and clinical research staff."
"How did you first hear about research data management plans (DMPs) as a concept? What do you see as the benefits of DMPs to your research community or others?"
"I first heard about research data management plans (DMPs) through an NIH grant requirement. The grant mechanism required a DMP in the proposal so the decision was made when we decided to apply.
A DMP creates standardization and expectations from the beginning that are documented for how we will collect, manage, analyze, store and share data over the life of the research study. There are many benefits including making it easier to plan, budget, and communicate between stakeholders. Stakeholders include the study team (especially as new people are on-boarded), regulators, security office, funders, participants, and the community, among others. It has also helped us with publications and presentations when we describe studies and the complexity of digital health trials."
"What are common practices in your community related to Data Management?"
"A couple common processes related to good data management include creating a data flow diagram and approaching data as a lifecycle. A data flow diagram is helpful to illustrate how data flows in a project to different kinds of stakeholders (i.e., clinicians vs. software developers). It can be difficult to describe complex data in writing especially if cloud computing is involved and data moves across a variety of settings. By approaching data as a lifecycle, it helps to create structure and consistent communication in the way it is described and is understandable. The data lifecycle is a series of stages for how data is acquired, used, analyzed, stored, shared and destroyed."
Figure 1: Example of a Data Flow Diagram from an ongoing NIH study. Credit: Ryan, Shaw, Ph.D., R.N.
"Has having a data management plan ever assisted you or others in understanding or remembering how project data is organized and/or processed? What about when sharing data?"
"Yes, especially for longer clinical trials where 5 or more years may have passed. New study personnel and trainees come on board over time and it's useful to have documentation for how project data is organized and processed. It’s also useful for yearly documentation needs such as progress reports to the IRB and funding agencies.
The DMP helps with having language available that describe the project data. When responding to requests to share data, we can share the DMP for efficient communication and consistency."
"What problems or barriers have you faced when trying to implement usage of a DMP? Why do you think people are reluctant to use DMPs?"
"Documentation burden for clinical research trials is high and this adds to that. I wouldn’t say that data management plans themselves have a lot of barriers but receiving approval for the plan across all stakeholders can be a lengthy process and requirements change frequently.
I feel that people will be less reluctant to use DMPs as they learn the benefits of them. While they require upfront work, the content can be recycled and is useful for yearly reporting and communicating consistently with stakeholders over time."
"What advice would you give others who are just getting started in terms of using DMPs?"
"Reach out to your department to ask who would be the best person to assist with your DMP. The needs will vary by the technical complexity of your study. If you’re on the health system side, the OASIS team (Office of Academic Solutions and Information Systems) may be able to provide additional assistance. Consultations can be requested here."
"Any closing thoughts around Data management planning you would like to comment on?"
"While data management planning may feel like more work upfront, it will save you time in the long term and provides many benefits. Not only may they be required by funders and regulators, they are useful for communication with team members, stakeholders, and useful in scholarship such as presentations and publications."
Read more about Dr. Ryan Shaw
Duke’s Shaw Highlights Transformative Opportunities in Digital Health
State-of-the-Art Technology in Surprising Places
How You Can Help Scientists Better Understand COVID Variants With Wearable Devices
New School of Nursing Maker Space to Encourage Collaborative Health Innovation
