The Effective Statistician - in association with PSI

By Alexander Schacht and Benjamin Piske, biometricians, statisticians and leaders in the pharma industryScience Natural Sciences

The Effective Statistician - in association with PSI episodes

October 05Integrated Evidence Planning That Connects Drug Development to Patient Needs
A Conversation with Gorana Capkun
Why You Should Listen
Understand integrated evidence planning and how it connects evidence generation across functions, stakeholders, and geographies.
Look beyond regulatory approval to consider what payers, physicians, and patients need from the evidence.
Strengthen your contribution as a statistician by asking questions that clarify endpoints, study designs, and intended conclusions.
Learn why patient perspectives matter when assessing treatment benefits and the practical challenges of clinical care.
Recognise common planning mistakes that can delay access and reduce the impact of otherwise strong research.
Episode Highlights With Timestamp
02:23 — From pure mathematics to patient-focused evidence
Gorana shares how her career across statistics, epidemiology, and pharmaceutical functions shaped her approach to evidence generation.
04:29 — What makes an evidence plan integrated?
We discuss a shared strategic roadmap that connects different methods and stakeholder needs throughout the product life cycle.
07:17 — Why evidence planning should start in research
Gorana explains how early collaboration helps teams understand unmet needs, disease burden, and how to measure success.
15:56 — Balancing expert advice with real-world data
We explore how advisory boards can guide research—and why their observations may not tell the whole story.
18:49 — Aligning evidence needs before launch
We discuss why early input from market access, medical affairs, and other functions matters when choosing endpoints and study designs.
25:59 — What changes after a treatment reaches patients?
We examine how clinical practice reveals new questions about tolerability, treatment burden, and outcomes beyond trial populations.
29:44 — Common mistakes and how to avoid them
Gorana shares why starting too late, working in silos, and unclear communication can undermine integrated evidence planning.
Links and Resources
Merck Healthcare
The Effective Statistician
The Effective Statistician Academy — Explore the Academy and its learning resources. The Effective Statistician
Join PSI — Learn about membership in the statistical community supporting healthcare research. psiweb.org
Further reading: Integrated Evidence Planning for Enhancing Patient Care — Harnessing the Power of Real-World Evidence — A related publication co-authored by Gorana Capkun. PMC
34 min
September 28How to Make Statistics Click with Schools Outreach and Interactive Data Visualization
A Conversation with Steve Mallett
Why You Should Listen
Discover how PSI helps students see where maths can take them, including careers in pharmaceutical statistics.
Learn how interactive visuals can connect a histogram to the experiences of individual patients.
Find practical ways to design statistical training around the questions your audience actually has.
Hear how you can support schools outreach, even if you have never led a classroom workshop.
Major Episode Highlights
02:00 — Steve shares why he joined the PSI Schools team and what the team hopes to achieve.
[- 06:13 — We discuss translating outreach activities and bringing them to schools beyond the UK.
07:39 — Steve explains how new volunteers can work alongside experienced presenters.
09:35 — We walk through a school visit featuring R Shiny workshops, a clinical trial simulation, and career conversations.
12:56 — Steve shows how an interactive histogram can help students connect variability with patient outcomes.
17:15 — We discuss how to build training around your audience’s challenges and help concepts stick.
23:01 — Steve explains why medical colleagues want the confidence to discuss clinical trial results accurately.
25:22 — I bring together our key lessons on knowing your audience, using clear visuals, and encouraging exploration.
Links and Resources
list text herePSI Schools website: https://www.psiweb.org/careers/schools-zone
list text here Email address for people interested in volunteering: [email protected]
list text here is an example of one of our Shiny workshop activities: mallettstats.shinyapps.io/Clinical-Trial-Workshop/
Steve Mallett's LinkedIn
28 min
September 07Will AI Replace Statisticians? The Future of Statistics in Pharma
A Conversation with Chris Harbron
Why you should listen
Listen to this episode to:
Understand how AI may change statistical work in the pharmaceutical industry
Recognize the limitations of using AI in clinical trial design and decision-making
Explore why human judgment, collaboration, and accountability still matter
Learn which capabilities statisticians should strengthen for an AI-enabled future
Discover practical ways to experiment with AI in your daily work
Approach the future of statistics with greater confidence and curiosity
Episode highlights with timestamps
01:30 – Will AI make statisticians redundant?
I introduce the central question behind my conversation with Chris.
03:27 – Can AI create an entire clinical study package?
We examine what AI can generate—and why plausible outputs still require careful scrutiny.
05:16 – A protocol is more than a document
Chris explains how document development supports alignment, precision, and better scientific discussions.
07:08 – Navigating real-world trade-offsWe discuss the challenge of balancing timelines, budgets, sample sizes, biomarkers, and broader development goals.
10:18 – Combining human and artificial intelligence
We consider how AI can complement experience rather than simply compete with it.
12:41 – Why keeping a human in the loop matters
Chris highlights an important challenge surrounding review quality and confidence in AI-generated work.
13:37 – Building trust in an AI-enabled workplace
We explore what statisticians provide beyond technical competence.
15:53 – Who remains accountable when AI gets it wrong?
Our conversation turns to responsibility, risk, and the limits of delegating decisions to technology.
17:17 – How statisticians can prepare for the future
Chris offers practical guidance for professionals working with clinical trials, real-world evidence, and other data projects.
22:07 – A promising future for statisticians
We close by considering how AI could create new opportunities for the profession.
Links and Resources:
🔗 Read Chris Harbron’s paper: Will the Pharmaceutical Industry Need Statisticians in an AI World?
🔗 Learn more about Chris Harbron and PoshStatsConnect with Chris Harbron on LinkedIn
🔗 Learn more about PSI and become a member
🔗 The Effective Statistician Academy – I offer free and premium resources to help you become a more effective statistician.
🔗 My New Book: How to Be an Effective Statistician - Volume 1 – It’s packed with insights to help statisticians, data scientists, and quantitative professionals excel as leaders, collaborators, and change-makers in healthcare and medicine.
Join the Conversation:
Did you find this episode helpful? Share it with your colleagues and let me know your thoughts! Connect with me on LinkedIn and be part of the discussion.
Subscribe & Stay Updated:
Never miss an episode! Subscribe to The Effective Statistician on your favorite podcast platform and continue growing your influence as a statistician.
25 min
September 01How to Build an Effective Patient-Reported Outcome Strategy for Clinical Trials
A conversation with Julia Poritz
Why you should listen
You’ll learn:
Learn how to build a cohesive patient-reported outcome strategy from measure selection through publication.
Choose PRO measures that reflect what truly matters to patients.
Capture treatment effects that traditional clinical endpoints may overlook.
Understand how regulatory, payer, and clinical trial requirements influence PRO planning.
Avoid common mistakes involving patient burden, translation, analysis, and reporting.
Generate PRO evidence that is scientifically rigorous, comparable, and meaningful to patients.
Episode highlights with timestamps
01:29 – The Role of Patient-Reported Outcomes
Julia shares how PROs help bring the patient voice into clinical research.
04:19 – Connecting Treatment to Quality of Life
We discuss a useful framework for understanding how treatment can affect symptoms, functioning, and patients’ overall well-being.
08:59 – Building a Cohesive PRO Strategy
Julia introduces the essential areas researchers should consider when planning their approach.
15:00 – Common Challenges in PRO Planning
We explore practical issues involving implementation, translation, and international studies.
16:23 – When Should PRO Planning Begin?
Julia explains why timing matters and what development teams should consider when introducing PRO measures.
19:14 – Keeping Patients at the Center
We close with the most important principle behind every effective PRO strategy.
Links and Resources:
🔗 PROTEUS Toolbox: Patient-Reported Outcomes Tools, Engaging Users and Stakeholders
🔗 National Cancer Institute: Introducing the PROTEUS Toolbox
🔗 CONSORT Reporting Guidelines
🔗 STROBE Reporting Guidelines
🔗 PRISMA Reporting Guidelines
🔗 The Effective Statistician Academy – I offer free and premium resources to help you become a more effective statistician.
🔗 My New Book: How to Be an Effective Statistician - Volume 1 – It’s packed with insights to help statisticians, data scientists, and quantitative professionals excel as leaders, collaborators, and change-makers in healthcare and medicine.
Join the Conversation:
Did you find this episode helpful? Share it with your colleagues and let me know your thoughts! Connect with me on LinkedIn and be part of the discussion.
Subscribe & Stay Updated:
Never miss an episode! Subscribe to The Effective Statistician on your favorite podcast platform and continue growing your influence as a statistician.
21 min
August 11How the R Consortium Is Transforming Regulatory Submissions and AI in Clinical Trials
A Conversation with Ning Leng
Why you should listen
If you work in statistics, data science, statistical programming, or clinical development, this episode will give you a practical look at how the industry is moving toward more modern and collaborative approaches.
You’ll learn:
How the R Consortium is helping pharmaceutical companies work together and collaborate with the FDA.
What the R Consortium’s regulatory submission pilots can teach you about using R in real-world submissions.
How open-source collaboration can influence regulatory guidance and industry practices.
How tools like R, Shiny, containers, and WebAssembly could make regulatory submissions more interactive and reproducible.
How the community is beginning to explore generative AI for clinical trials and statistical programming.
Why benchmark datasets, challenging test cases, and quality control will be critical for using AI responsibly.
How you can explore the R Consortium’s public resources and get involved in its work.
Whether you’re already using R or simply want to understand where statistical programming and regulatory submissions are heading, this conversation with Ning Leng will give you valuable insights into the future of our field.
Episode highlights with timestamps
01:30 — Meet Ning Leng and her journey into R
Ning introduces her background in statistics and computational genomics and explains how she became involved in R adoption, cloud migration, Git, and Shiny at Roche Genentech.
03:57 — Why R wasn't being used for regulatory submissions
Ning explains the misconception that the FDA did not accept R and identifies the real challenge: the industry lacked practical examples showing how to submit R-based materials.
05:46 — What is the R Consortium?
We discuss the R Consortium's role as a nonprofit organization that promotes good practices and the use of R across industries.
06:16 — How the R Consortium collaborates with the FDA
Ning explains the submission working group, validation hub, and how the Consortium provides a platform for collaboration between pharmaceutical companies and the FDA.
08:13 — R Consortium FDA submission pilots
Ning walks through the different pilots, from submitting TLGs to incorporating Shiny, ADaM code, containers, WebAssembly, and alternative data formats.
10:29 — How companies are using the pilot submissions
We explore how pharmaceutical companies use the public R Consortium pilots as practical templates when preparing their own R-based regulatory submissions.
11:24 — How collaboration influenced FDA guidance
Ning explains how lessons from the pilots helped clarify FDA guidance around file formats, including .r and .zip files.
12:17 — Moving beyond PDF-based electronic submissions
We discuss the opportunity to use interactive graphics, HTML, Shiny, and other modern technologies to make electronic regulatory submissions more useful and traceable.
14:14 — Generative AI enters the picture
Ning explains how the R Consortium is beginning to explore generative AI for clinical trial reporting, trial design, and statistical programming.
15:12 — Building AI skills and benchmark test cases
We discuss open-source AI skills, benchmark datasets, and the importance of testing AI against difficult and unusual corner cases.
16:28 — How you can get involved with the R Consortium
Ning shares practical ways to explore the community, including its public working-group materials, meeting minutes, recordings, and Slack channel.
Links and Resources:
🔗 R Consortium: Learn more about the R Consortium and its work to promote the use of R and good practices across industries: https://r-consortium.org/
🔗 R Consortium R Submissions Working Group: Explore the working group's projects, meeting materials, submission pilots, and opportunities to get involved: https://rconsortium.github.io/submissions-wg/
🔗 R Submissions Working Group — Pilot Projects: Learn about the different FDA submission pilots and how the community is exploring R for regulatory submissions: https://rconsortium.github.io/submissions-wg/pilot_background.html
🔗 R Consortium 2026 Plans and 2025 Success: Read about the latest work from the R Submissions Working Group, including the development of Pilots 6 and 7: https://r-consortium.org/posts/submissions-wg-2026/
🔗 Pilot 4 — WebAssembly and Containers: Learn how the R Consortium explored submitting a Shiny application using WebAssembly and containers for FDA review: https://r-consortium.org/posts/using-r-to-submit-research-to-the-fda-pilot-4-successfully-submitted/
🔗 R Consortium Working Groups: Browse the R Consortium's different working groups and projects: https://r-consortium.org/all-projects/isc-working-groups.html
🔗 The Effective Statistician Academy – I offer free and premium resources to help you become a more effective statistician.
🔗 My New Book: How to Be an Effective Statistician - Volume 1 – It’s packed with insights to help statisticians, data scientists, and quantitative professionals excel as leaders, collaborators, and change-makers in healthcare and medicine.
Join the Conversation:
Did you find this episode helpful? Share it with your colleagues and let me know your thoughts! Connect with me on LinkedIn and be part of the discussion.
Subscribe & Stay Updated:
Never miss an episode! Subscribe to The Effective Statistician on your favorite podcast platform and continue growing your influence as a statistician.
19 min
July 21Project Optimus and what you need to know about it
A Conversation with Ayon Mukherjee
Cancer treatments have changed dramatically over the past decade, but have our dose-finding strategies kept pace?
In this episode, I speak with Dr. Ayon Mukherjee, who leads statistical innovation in early oncology development at Eli Lilly. Together, we explore Project Optimus, the FDA initiative that is changing how we think about dose optimization in oncology.
Instead of simply finding the highest dose patients can tolerate, Project Optimus encourages us to identify the dose that provides the best balance between efficacy, safety, pharmacokinetics, pharmacodynamics, and long-term tolerability.
Ayon explains why the traditional maximum tolerated dose approach worked well for chemotherapy but often falls short for targeted therapies and immunotherapies. We also discuss how statisticians can help lead this transformation by designing better dose optimization studies and collaborating more effectively with clinicians, pharmacologists, and regulators.
Why Listen to this Episode:
Understand what Project Optimus is and why it is transforming oncology drug development.
Learn why the traditional maximum tolerated dose (MTD) approach is no longer sufficient for many targeted therapies and immunotherapies.
Discover how statisticians can use PK/PD, exposure-response, efficacy, safety, and tolerability data to support better dose optimization.
Explore the progress the industry has made since Project Optimus was launched and the challenges that remain.
Gain practical insights on collaborating effectively with clinicians, pharmacologists, regulators, and academic partners.
Find out what statisticians can do today to help advance innovative dose optimization strategies and improve patient outcomes.
Episode Highlights
00:00 – Introduction to the episode
01:31 – Ayon Mukherjee introduces himself and his work in early-phase oncology and dose optimization.
02:58 – What is Project Optimus, and why did the FDA introduce it?
03:28 – Why traditional chemotherapy dose-finding approaches no longer fit modern targeted therapies and immunotherapies.
05:21 – The key principles of Project Optimus: balancing efficacy, safety, PK/PD, and long-term tolerability.
08:14 – The types of data needed to support dose optimization beyond dose-limiting toxicities.
09:40 – How far has the industry come since Project Optimus launched in 2021?
11:06 – Why communication and cross-functional collaboration are essential for successful implementation.
13:29 – Regulatory acceptance and the gap between published methodologies and industry adoption.
15:27 – The value of industry-academia collaboration and cross-company working groups.
16:53 – Why education and training are critical for increasing awareness and adoption.
20:06 – Open-source tools, R Shiny applications, and practical resources for implementing innovative trial designs.
23:23 – Final thoughts on how statisticians can improve dose optimization and ultimately serve patients better.
Links and Resources
Connect with Dr. Ayon Mukherjee on LinkedIn
FDA Project Optimus – An initiative from the FDA Oncology Center of Excellence to reform dose optimization in oncology drug development.
TrialDesign.org – Open-source tools and R Shiny applications for innovative clinical trial designs.
Innovative Design Scientific Working Group (IDSWG) – A collaborative group advancing innovative clinical trial designs in early-phase oncology.
26 min
June 30Unlocking Growth: The Power of Coaching and Mentoring for Statisticians
A Discussion with Alun Bedding and Emma May
In this insightful interview, Emma May and Alun Bedding explore the nuances of coaching and mentoring, sharing personal stories, frameworks, and practical tips to enhance professional growth. Discover how these powerful tools can transform statisticians' careers and foster leadership development.
Key topics:
Differences and overlaps between coaching and mentoring
Frameworks for mentoring conversations (Challenges, Choices, Consequences)
The importance of independence in coaching and mentoring
Building trust and confidence in professional relationships
Role of reflective practice and visualization in growth
Episode Highlights
02:00 – Emma shares how coaching and mentoring helped her overcome limiting beliefs.
04:00 – Coaching vs. mentoring: the key differences and why both matter.
12:20 – Why technical expertise alone isn't enough for career growth.
17:00 – The value of having an independent coach or mentor.
22:00 – Building leadership through small, consistent actions and accountability.
26:40 – How improv and role-play strengthen communication and leadership.
32:25 – Choosing between coaching and mentoring—and why one session can make a lasting impact.
Links
Understanding Coaching & Mentoring PDF
Emma May: LinkedIn https://linkedin.com/in/emma-may
36 min
June 09Understanding and Mitigating Endpoint Bias in External Control Arms
A Conversation with Benjamin Ackerman
External control arms are becoming increasingly important in drug development, but creating valid comparisons requires more than matching patient populations.
In this episode, I speak with Ben Ackerman, Director of Real-World Biostatistics at GSK, about one of the most overlooked challenges in external control arm studies: endpoint bias. We discuss why differences in how outcomes are measured can influence study results, what researchers should consider when designing studies, and how the field is evolving to address these challenges.
If you work with real-world evidence, causal inference, or innovative clinical trial designs, this episode offers valuable insights into improving the credibility and transparency of external control arm analyses.
**Why You Should Listen
**
Learn why endpoint alignment matters as much as population matching.
Understand how measurement differences can create bias in external control arm studies.
Discover practical methods to quantify and mitigate endpoint bias.
Hear how regulators are increasingly evaluating endpoint comparability.
Gain insights into better study design and pre-specification strategies for real-world evidence research.
**Episode Highlights
**
00:01:31 – Introducing Ben Ackerman and external control arms
00:04:41 – Why endpoint bias deserves more attention
00:08:38 – Understanding the challenges of comparing different data sources
00:12:30 – Practical considerations for study design
00:16:32 – The role of transparency and pre-specification
00:20:30 – Regulatory perspectives and future expectations
00:26:07 – Where the field is heading next
**About Ben Ackerman
**
Ben Ackerman is Director of Real-World Biostatistics at GSK and a PhD biostatistician specializing in causal inference, real-world evidence methods, and the integration of randomized trial data with observational data sources. His research focuses on improving evidence generation through innovative statistical methods that bridge clinical trials and real-world healthcare data.
29 min
June 02The Future of Statistical Methodology in Drug Development
A Discussion with Alun Bedding and David Wright, Jürgen Hummel, and Jenny Devenport
This episode features three leading statistical methodology experts discussing the role, impact, and future of methodology groups in the pharmaceutical industry. They explore organizational structures, skill sets, AI integration, and strategies to accelerate adoption of innovative methods.
**Key topics:
**
Role and impact of methodology groups
Organizational considerations for methodology teams
Skills and traits of great statisticians
Integration of AI and machine learning in pharma
Strategies to accelerate adoption of new methods
**Episode highlights:
**
00:00 Introduction to Statistical Methodology Groups
02:18 Exploring the Paper's Insights
06:49 The Role of Methodology Statisticians
10:13 Consultation and Collaboration in Drug Development
12:54 Addressing the Innovation Problem in Drug Development
16:06 Qualities of a Great Methodology Statistician
20:31 The Future of Methodology Groups and AI
25:46 The Importance of Human Insight in Clinical Trials
28:27 The Prevalence of Methodology Groups in the Industry
30:29 Goals of Methodology Departments
**Links and resources:
**
Statistical Methodology Groups in the Pharmaceutical Industry Paper
Https://statistics.biopharmaceutics.com/article/10.1177/15501477221112345
EFSPI Statistical Leaders Group
https://www.efspi.org/statistical-methodology-leaders/
EFSPI Ecosystem
https://www.efspi.org/ecosystem/
**Guest links
**
Jenny Devenport: https://www.linkedin.com/in/jenny-devenport/
David Wright: https://www.linkedin.com/in/david-wright/
Jurgen Hummel: https://www.linkedin.com/in/jurgen-hummel/
32 min
May 26Rethinking Programming Validation and Traceability in Clinical Trials
A Discussion with Andrew (Andy) York
**Episode Summary
**
What does “quality” really mean in statistical programming?
In this episode of The Effective Statistician, I speak with Andrew (Andy) York about the evolving world of programming validation, traceability, and quality assurance in clinical trials. Andy has decades of experience in statistical programming, leadership roles across pharma and CROs, and now works with AI-driven solutions focused on improving validation and traceability.
We discuss why traditional approaches to validation are becoming increasingly difficult to sustain, how expectations from regulators continue to grow, and why traceability is far more than just linking programs and datasets.
Andy also shares how modern AI-powered tools can automatically map programming workflows, connect datasets and outputs, and create end-to-end traceability from raw data to final tables, figures, and listings.
If you work with statistical programming, clinical data workflows, submissions, or validation processes, this episode will challenge some long-held assumptions and introduce you to where the future may be heading.
**Why You Should Listen
**
Learn what “quality” in programming really means beyond simply writing working code
Understand the challenges of maintaining traceability across complex clinical trial workflows
Discover why manual validation processes are becoming harder to scale
Hear how AI is starting to transform validation and traceability in programming
Explore the balance between regulatory expectations, efficiency, and confidence in outputs
Gain insights from someone who has seen the evolution of statistical programming from the very beginning
**Episode Highlights
**
00:01:30 — Andy York’s journey into statistical programming
Andy shares how he started programming during the early days of SAS in pharma and how the role of programmers evolved over the decades.
00:04:41 — What does programming quality actually mean?
We discuss confidence in outputs, customer expectations, regulatory requirements, and creating programs that your future self can still understand years later.
00:06:46 — The regulator’s perspective on validation and traceability
Andy explains why full traceability from raw data to final outputs is essential for regulatory confidence.
00:08:15 — The limitations of traditional traceability approaches
We reflect on the common experience of manually navigating folders, programs, and datasets to reconstruct programming logic.
00:09:45 — How automated traceability changes the game
Andy explains how modern tools can automatically create end-to-end traceability matrices across programs, datasets, and outputs.
00:10:45 — Forward traceability vs. backward traceability
A fascinating discussion about not only tracing outputs back to source data, but also understanding where every data point flows forward through the analysis process.
**Links and References:
**
Verisian - https://verisian.com/
Bayer case study https://verisian.com/customer-stories/how-bayer-uses-verisian-ai-to-automate-submission-document-generation
24 min

About The Effective Statistician - in association with PSI

From the publisher's feed

The podcast from statisticians for statisticians to have a bigger impact at work. This podcast is set up in association with PSI - Promoting Statistical Insight. This podcast helps you to grow your…