gms | German Medical Science

Background: PEACHES is a cohort of 1671 PEACHES mother-child pairs observed over ten years and documented in 5067 variables since conception. To make this data FAIR requires interdisciplinary effort between clinical researchers, statisticians, data managers, and IT specialists. The project tasks consisted of data migration from MS ACCESS to REDCap, variable labeling, data cleaning, creating a data dictionary, and metadata storage in a catalogue. Problems like inefficient teamwork, poor communication, uncoordinated priorities of members, derailing of research focus, confusion of task requirements, and intellectual disagreement are prevalent. We assess the implementation of Scrum to the current project management to streamline the team processes, and hereby share our experience with Scrum.

Designed for software development, Scrum methods improve complex processes by well-defined role assignment, purposeful Sprint structure, and transparent Backlog. Short and iterative cycles of Sprint are employed, involving task definition and allocation, regular meeting and feedback sessions, and task reviews and adjustment. By defining roles, processes, and communication channels, Scrum ensures that project team members remain focused and informed about their tasks.

Method: A team member was assigned as the Scrum master who was directly related to the developers and represented the ideas of the PEACHES steering group. After discussing requirements with both parties repetitively, the Scrum master initialized and facilitated task communication in the team. A sprint took place about every two weeks and a shortlist of backlogs was selected in each sprint as the prioritized objectives for the team. Shortlisted tasks were expected to be undertaken within the two-week bound by the assignees clearly stated in the sprint, and the progress of those tasks was reviewed in the next meeting. We retrospectively collected data on the count of emails and meetings, the complexity and lead-time of task, under two management frameworks: Waterfall and Scrum, to compare their process efficiency.

Result: The project started in November 2020 and we implemented Scrum in January 2024 with a hard cut. Throughout the whole period, team size increased from five people to ten. Under Scrum, correspondence was more frequent with an 167% increase in average email counts and 72% in meeting counts. Tasks were divided into smaller pieces with a decrease in proportion of unspecific tasks by 10%. Average lead-time per task decreased by 28% and number of months of idle time without communication or work progress reduced.

Discussion: Scrum increased specificity of tasks by a more concrete definition of requirements tailored to responsible persons. Communication between task assignees increased without prolonging lead-time of tasks, allowing faster completion of tasks. The rapid routine retrospective allowed early identification of defects in the process and continual improvement in close collaboration. Clear communication and enhanced adaptability in the team were fostered in this dynamic transparency. Efficiency has been further reinforced by the utilization of IT tools like Nextcloud and Deck that improved persistent communication settings.

Conclusion: By addressing the complexities inherent in FAIRification projects, Scrum offers a structured and adaptable approach. Scrum did not only foster a better working dynamic but also facilitated project progress.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.