Variations on a Theme. Let Me Count the Ways We Do Things Differently to Achieve the Same Goal

I delivered a paper at this year’s PHUSE US Connect in Florida. Here’s a synopsis of it, but you can find a video of the presentation and the slides in the PHUSE Archive.

Common goal, different processes

In the world of life sciences clinical data analysis, there is a common goal among companies: to produce statistical results that meet regulatory requirements. However, despite this shared goal, each company has its own process for achieving it. In the paper, I explored the various ways in which companies approach data analysis and I offered a comparison of these approaches. I also discussed the potential benefits of creating industry-wide standards and achieving harmony and efficiency in the process.

Firstly, it is important to note that there are no explicit industry standards for data analysis processes. As a result, companies are free to adopt their own processes to achieve the goal of producing regulated output. However, this has resulted in a wide range of approaches to data analysis, with each company having its own unique method.

One area in which there is a lot of variability is in the workflow and business process followed in using data analysis to create output. While all companies must acquire and manage data, transform it, and analyse it to create reports, the specifics of how this is done can vary greatly. For example, companies may use different languages in the analysis phase or employ different data standards at each step along the way. Additionally, there are differences in folder hierarchies, security models, versioning, and project management processes.

Reasons include legacy systems, history, and politics

The reasons for these differences are varied, but some common factors include legacy systems, history, and politics. If a company has been using a certain process for a long time, it can be difficult to change it. Additionally, if there is a lack of new ideas or new blood coming into the department, the legacy approach may remain for longer than necessary. Furthermore, there may be politics involved, with some people having a stronger voice than others, and some people being resistant to change.

However, there is also a fear of change, especially in areas where the current process has been proven to work and is acceptable to regulators. Companies may be hesitant to change their processes if they fear that it will upset the apple cart and their methods will no longer be acceptable to regulators.

Potential cost-saving benefits

Despite the potential challenges, there are benefits to achieving harmony and efficiency in data analysis processes. One of the key benefits is the potential for cost savings. If companies are able to adopt more consistent processes, they may be able to streamline their operations and reduce costs. Additionally, having industry-wide standards could make it easier for companies to collaborate and share data, leading to faster and more effective drug development.

Overall, while there is no one-size-fits-all approach to data analysis processes, there is potential for greater harmony and efficiency if companies can agree on some common standards. This could lead to cost savings and more effective drug development, making it a worthwhile goal for the industry to pursue.

Do you see benefits in standardising? How might we best make a start on this journey? Send me your comments!