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?
Lovisa and I together with some others are presenting at the next PSI conference about this topic and today, we dive already into this topic and identify different dimension to be considered to understand better the different shades pre-specified analyses.
Listen to this episode to avoid oversimplification and confusion in discussions in the future.

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.

Risk-based monitoring plays an increasingly important role for clinical trials. Of course, the assessment of the risk is based on statistics. This presents now only interesting career options for statisticians, but also has an impact on the role of statisticians in study teams.

In this episode, we’ll give you an introduction to risk-based monitoring (RBM) as well as speak about the role of statisticians in this area. Further we provide you as a study statistician insights into what you need to know about RBM. Finally, we also give some recommendations in terms of further resources to learn from.

1. Start with a relevant example
2. Collect questions upfront and track progress of answering them during the training
3. Create regular meetings to engage people
4. Interrupt your presentations with asking questions
5. Use contrasts to show the impact
6. Have a physician first introduce the example study
7. Don’t shy away from speaking to very basic things like p-values
8. Prefer white board over slides
9. Use technology for your advantage in virtual settings
10. Make pre-read easy
11. Collect feedback

Ultimately, we work to improve patients lives but traditionally the voice of the patient was not important. However, over the last decades many stakeholders shift to include the patient more in the decision-making process and patients voice their needs and preferences more.

Still capturing and evaluating patient preferences does not belong to the core capabilities of most biostatisticians. This episode will help you to understand patient preferences.

Are you early in your career and you could still go for a PhD in statistics?

Do you wonder, if this pays out short and long-term?

Is it worth the money?

What are the long-term benefits or are there actually short term benefits as well?

Do you want to change your job?
Are you just looking for you first job?
Do you consider to work for your own company or as a free-lancer?

Then this episode will help you!

Benefit-risk assessments are made by various stakeholders including sponsors, regulators, payers, physicians, and patients throughout the life-cycle of a treatment.

The understanding of data plays a vital role in weighting positive and negative effects of medications.

In this episode, we give you an introduction to what benefit-risk (BR) assessment is in order for you to understand your role.

In this episode, we talk about
- how to network
- how to prepare for the conference
- how to
follow up after the conference
As well as some specific topics around the PSI conference itself.

This episode follows the outline of a webinar, that I presented in 2018 during a PSI webinar. I’m introducing cluster analysis and why you as a statistician need to know about it.