Podcast

Group sequential trials, interim analyses, final analyses, updated analyses… what do these terms actually mean, and why is there so much confusion? In this technical yet highly practical episode, I speak with Kaspar Rufibach, Principal Biostatistician at Roche, to unpack some of the most commonly misunderstood terminology and concepts in clinical trial design and analysis. If you’ve ever questioned what really qualifies as an “interim analysis” or struggled to explain why a “final analysis” isn’t always the last word, this conversation is for you.

In this episode, I had the pleasure of speaking with Prof. Holger Fröhlich, who leads the AI and Data Science Group at the Fraunhofer Institute and is an honorary professor at the University of Bonn. We explored one of the hottest topics in healthcare data science right now: synthetic data. Holger and I discussed how synthetic data is generated using AI, what role digital twins could play in the future of clinical trials, and how these innovations could fundamentally reshape how we design and conduct research. We dove into the Cynthia Project, which is part of the Innovative Health Initiative (IHI) – the largest public-private partnership for health research in Europe.

In this episode, I’m diving into a topic that’s very personal to me—working in an English-speaking environment as a non-native speaker. If you’ve ever felt unsure about your English skills in meetings, emails, or presentations, you’re not alone. I’ve been there myself, and I want to share what’s helped me grow more confident and effective over the years. I’ll walk you through practical strategies that go beyond grammar—things that have really made a difference for me in both speaking and writing, and most importantly, in getting my message across clearly.

In this episode, I’m once again joined by Daniel Sabanés Bové for a deep dive into one of the most impactful tools for statisticians working with data visualization—R-Shiny. We explore how interactive data visualizations can help you iterate faster, collaborate better across functions, and focus more on the actual scientific questions rather than just coding. Daniel shares some excellent examples from clinical trials and gives practical tips on how to avoid common pitfalls when building Shiny apps. Whether you’re designing your first app or maintaining a more complex one, you’ll find plenty of value in this conversation—from best practices around UI/UX design to strategies for modular development and testing.

In this episode, I’m sharing three personal stories where soft skills—or better yet, human skills—made a huge difference in my work as a statistician. Whether it was building trust to access critical data, presenting results in a way that truly resonated, or negotiating a fair contract, these experiences reminded me how essential these skills are alongside our technical expertise.

In this episode, I’m joined once again by Daniel Sabanés Bové to talk all about R packages—why they’re so useful, when to create one, and how to do it effectively. Whether you’re just starting out with writing reusable functions or thinking about building a more robust and reusable R package, you’ll find plenty of hands-on advice in our discussion. Daniel shares his experiences from working at Roche, Google, and now through his consultancy, Rconis. We dive into everything from writing clean and consistent code, to testing, documenting, and even promoting your package in the open-source world.

In this Friday episode, I’m sharing some hard-earned lessons on delegating programming tasks—something that completely changed the way I work and lead. I didn’t start out knowing how to delegate effectively. Like many of you, I just figured it out as I went. Over time, I reached a point where I didn’t even need a SAS license anymore because I had fully delegated all my programming tasks. But getting to that level of trust and clarity wasn’t always straightforward—especially with SOPs in the mix. SOPs are meant to guide us, but I’ve found they can both support and limit effective delegation. In this episode, I break down what SOPs do well, where they fall short, and what really matters when assigning work to others.

In this episode of The Effective Statistician podcast, I dive into the art of persuasion. As statisticians and data scientists, we often rely on logic and data, but true influence requires more than just being right. Drawing from ancient Greek philosophy, I explore the three pillars of persuasion—ethos (credibility), logos (logic), and pathos (emotion)—and share practical strategies to help you effectively convince others.

In this episode of The Effective Statistician podcast, I dive into the art of persuasion. As statisticians and data scientists, we often rely on logic and data, but true influence requires more than just being right. Drawing from ancient Greek philosophy, I explore the three pillars of persuasion—ethos (credibility), logos (logic), and pathos (emotion)—and share practical strategies to help you effectively convince others.

In this episode of The Effective Statistician, I sit down with Kaspar Rufibach to tackle a topic that affects statisticians every day—how to interpret p-values, confidence intervals, and statistical hypotheses. We explore the differences between Fisher’s and Neyman-Pearson’s approaches, clear up common misconceptions, and discuss how misinterpreting statistical significance can lead to flawed conclusions. Using real-world examples from clinical trials and drug development, we highlight best practices for communicating statistical results effectively.