Innovation

Interview with Marco Boeri

As statisticians in the medical field, we’re used to study subgroups of patients with respect to all kinds of biological variables: from demographics to genomics. This provides us with a good understanding of how the benefit-risk profile for a given patient looks like.

However, the patient might have a completely different view on the importance of the different benefits and risks. And importantly, these preferences might be less driven by biologic factors and more by personal experiences and situations as well as psychological traits. How can we assess patient preferences in this regard?

Marco Boeri and I worked on such questions in the past and some work has been presented at last years PSI conference. In todays episode, we give you some insights into what’s possible and how you can approach this problem.

We’ll dive into questions like:

Interview with Michael O’Kelly

Estimands continue to be a hot topic, but many statisticians struggle to put it into practice. As statistician, we wonder about the correct interpretation and how to analyse different estimands.

In todays episode, we speak with Michael O’Kelly, an expert on this topic with lots of presentations around estimand (see e.g. the PSI events). He also won the award for Statistical Excellence in the Pharmaceutical Industry, jointly run by the RSS and Statisticians in the Pharmaceutical Industry (PSI).

In the interview we will dive deep into a specific approach – the composite strategy – and cover questions like:

Interview with Andy Nicholls

One of the most common questions I got asked during my nearly 2 decades of being a statistician sounds similar to this: “Which patients have the best response to treatment?”

I’m sure, we all face this situation sooner or later and not surprisingly lots of research has happened in the last years on this area. In todays episode, we will help you to understand one of the best approaches I have come across to solve this problem in a rigorous yet sophisticated way: the SIDES approach.

Both Andy Nicholls and I have applied this approach in the past and we’ll use an example, which he presented during a PSI webinar.

Listen to this episode to learn step by step how to apply the SIDES method.

Interview with Lovisa Berggren

Prespecified=good and post-hoc=bad. This is how we as statistician see it usually and I did too. However, over the past years I realized more and more, that it’s not that easy.

How many details do you need to have to call an analysis pre-specified? Should we label a request to analyse a certain subgroup by regulators as well as a fishing expedition to find a significant subgroup both in the same way: post-hoc?

Overcoming the simplification strategy

Analysing binary or continuous data usually doesn’t cause any headaches for statisticians. But when we step into ordinal data, most of us ignore their specific nature and either dichotomize them or analyse them as if they are continuous.

Recently, these problems have becoming much more prevalent due to the nature of composite endpoints (watch out for an interesting episode on this in a few weeks).

Now Benjamin and I have worked on better tools to analyse such data already at university. We’ll dig back into what we learned then and what is still relevant today.

Specifically, we’ll cover the following questions: