APPLYING AGILE METHODOLOGY TO STATISTICAL PROGRAMMING

June 30 2022 Gina Boccuzzi

In any industry, there are various methods for approaching task deadlines to remain within the timeline. It is vital to choose a method that can accommodate challenges faced during the project and reduce time or at least help mitigate any delays. Effective task management is the key to maintaining the project budget and addressing other challenges.

What is Waterfall Project Management?

The waterfall method is the most common project management style used. It follows a linear pattern; i.e., the project moves from one task to the next upon completion, until the end of the project. There is a definitive start and end date for each task, by which the entirety of the work needs to be completed. Timelines are critical and adapting to scope changes may be difficult. When a large change is needed in a project, it can be very time and work-intensive to reach target completion dates.

What is Agile Project Management?

Agile is a cyclical style of project management. A project is broken down into small tasks that are completed and reviewed as the project progresses. Typically, there is a less firm task end date but work progresses on a rolling basis in two-week cycles, with certain tasks due at the end of each cycle. Agile can seem especially daunting when overall project deadlines are approaching but this does not mean that agile can’t be applied; the tools it implements can still be used with great results.

Agile management encompasses a set of twelve principles and four core values. These core values are:

  1.  Individuals and interactions over processes and tools
  2. Working software over comprehensive documentation
  3. Customer collaboration over contract negotiation
  4. Responding to change by following a plan

In the context of our work, these core values can be described best as collaboration, focus on a deliverable, communication, and adaptation.

“PROMETRIKA’s statistical programming team members have begun implementing agile techniques in their approach to building the datasets for analyses of clinical trial data.”

Share This Article

The Tools of Agile Methodology:

Daily Standup – a brief meeting with the entire team working on the specific task. It lasts less than 15 minutes and occurs every day. Discussions in this meeting may include updates on work done the day before and planned work for the current day, and any questions for team members. 

Sprint Planning/Retrospective – a recurring meeting that occurs once every two weeks to review work done in the previous period, estimate the remaining work for the task/project, and anticipate work that can be done during the next sprint. There are many resources available to enable sprint planning.

Post mortem – a meeting conducted at the end of a project after all work has been completed. For example, in the setting of CDISC SDTM programming, questions discussed during this meeting may include: Did we gauge the complexity of the SDTM domains for the specific study correctly? Were there unexpected challenges with the raw data? Were the programmers able to devote enough time to this project? Working together to find the reasons behind the challenges that arose helps the team to account for these in future project planning and handle them more appropriately. This is typically a lengthier meeting that includes team members from the daily standup and management that may not be working on the task/project on a daily basis. Generally, post-mortems result in a set of action items to implement in the next project.

Comparing Waterfall and Agile Methods in Statistical Programming:

Figure 1. Waterfall Project Progression

 

Figure 2. Agile Project Progression

 

In Conclusion:

Agile project management methods may not be a perfect fit for every company and every field. In a CRO environment, in which timelines are set by clinical research sponsors, waterfall project design may remain a more natural fit. However, agile tools may still be introduced, in bite-sized increments, to specific tasks with great results.

PROMETRIKA’s statistical programming team members have begun implementing agile techniques in their approach to building the datasets for analyses of clinical trial data. Beginning the approach to agile management with small steps may require little to no commitment and can greatly optimize the team’s workflow, efficiency, and communication.