Video-on-Demand Library


Collection
Keyword or phrase

Topic

13 September 2023

Medication planning is an important issue for multimorbid patient. It can be measured by the Medication Appropriate Index (MAI). But how can it be visualised? Martin Brown presents the proposed visualisations in contrast to the regular tabulation of the results. Visualisations are available on the Wonderful Wednesday blog.

Read more...

Medication planning is an important issue for multimorbid patient. It can be measured by the Medication Appropriate Index (MAI). But how can it be visualised? Martin Brown presents the proposed visualisations in contrast to the regular tabulation of the results. Visualisations are available on the Wonderful Wednesday blog

An advantage of the visual display is the possibility to include individual patient data. This can be done in combination with a box-and-whisker plot. The dumbbell plot is showing individual changes and is giving a good impression on the distribution, especially with the right sorting applied. Depending on  the message to be transported also scatter plots or a simple histogram can be useful. The next challenge is on exploring dirty data. See the Wonderful Wednesday homepage for more detail.

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, David Carr, Steve Mallett, Rhys Warham, Lovemore Gakava, Zara Sari, Paolo Eusebi, Martin Brown, Greg Ball, Martin Karpefors, Benjamin Lang

12 September 2023

In this session we discuss the advances of AI and its applications within the pharmaceutical industry. We focus on opportunities, challenges and changes that lie ahead for the roles of programmers and statisticians. Join in to hear more on what ChatGPT is and whether the robots are really going to take over our roles!

Read more...

Hosted by Claire Britain and Chetan Mistry, for this session we're joined by: Kleio Kipourou, Andy Nicholls, and Tom Fisher.

In this session we discuss the advances of AI and its applications within the pharmaceutical industry. We focus on opportunities, challenges and changes that lie ahead for the roles of programmers and statisticians. Join in to hear more on what ChatGPT is and whether the robots are really going to take over our roles!

There are no expectations or requirements on the previous experience on using ChatGPT or AI for this session. This session is aimed at anyone who is interested in learning more about these topics. 

11 September 2023

In this webinar we will be showing you how you can inspire NextGen and help shape the future workforce of pharmaceutical statistics.

Read more...

Claire Brittain, Steve Mallet, Nicola Voyle, Doug Thompson, Emma Crawford.

Do you remember your career talks at school? What opportunities were promoted for maths and science? When did you   first discover the pharmaceutical industry as a career choice?  Raise awareness of our rewarding careers in schools For me, I only discovered the pharmaceutical industry when searching for industrial placements as part of my undergraduate degree. My decision to continue studying maths was fairly generic in that I was good at it at school and thought it was general enough to give me a few different career opportunities - although I was adamant I didn’t want to end up in teaching or finance. Any awareness of drug development I had would have come from science lessons, and never once had it crossed my mind that it was a field where maths skills were widely used and sought after - alongside several other data centric industries.  School outreach plays a vital role However not all students are encouraged to share the same passion and interest for maths that I did. This is where school outreach activities can play an important role in helping to raise awareness of the many exciting opportunities for working with data and inspiring the next generation to continue their studies in relevant subjects to allow them to do so.  Learn about the benefits of schools outreach volunteering Join us in this webinar to kick off the 2023/24 academic year, and find out how PSI can support you in engaging in school outreach activities and how you can take part in our new award scheme, “Outreach Champions Challenge Cup” which will recognise members who make exceptional contributions to outreach throughout the year. Listen to reflections from our panelists of their real-life experiences of participating in schools outreach activities and the benefits they have found from doing so.  If as a PSI community, every member engaged in at least one school's outreach activity per academic year, we could reach an impressive number of students and help to shape the future workforce of pharmaceutical statisticians.

09 August 2023

This month's webinar, chaired by Rhys Warham, focussed on the submitted line plot summarising benefits and risks over time. There followed an interesting discussion of the pros and cons of some alternative graphical approaches to summarising benefit-risk.

Read more...

This month's webinar, chaired by Rhys Warham, focused on the submitted line plot summarising benefits and risks over time. There followed an interesting discussion of the pros and cons of some alternative graphical approaches to summarising benefit-risk.

The main discussion focused on a line plot summarising both efficacy responders (benefit) and incidence of an AE of special interest (risk), on the same graph. This allowed the balance between benefits and risks to be summarised over time. Some alternative graphical approaches to summarising benefit vs risk were presented, and the pros and cons were discussed.  The new challenge is on the topic of medication planning for multimorbid patients. See the Wonderful Wednesday homepage for more detail.

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, David Carr, Steve Mallett, Rhys Warham, Lovemore Gakava, Zara Sari, Paolo Eusebi, Greg Ball, Benjamin Lang.

12 July 2023

This month Steve Mallet presents epidemiology data on measles, polio, and rubella from the 20th century. The challenge is to display the incidence change over time, especially after the availability of the vaccine. Visualisations are available on the Wonderful Wednesday blog.

Read more...

This month Steve Mallet presents epidemiology data on measles, polio, and rubella from the 20th century.  The challenge is to display the incidence change over time, especially after the availability of the vaccine. Visualisations are available on the Wonderful Wednesday blog.

To visualise the big amount of data in one plot heatmaps were used. Having the time on the horizontal axis is intuitive and helps to quickly understanding the data. On the vertical axis the way of sorting or arranging the data can support understanding of relationships. One question is how to display missing data. Another question is when to use diverging and when to use sequential colour scales. Have a look. The new challenge is on benefit-risk assessment. See the Wonderful Wednesday homepage for more detail.

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, David Carr, Steve Mallett, Rhys Warham, Lovemore Gakava, Zara Sari, Paolo Eusebi, Greg Ball, Benjamin Lang.

27 June 2023

This webinar discusses an investigation to improve nonlinear mixed-effects approach for EC50 estimation based on multi-donor dose-response data.

Read more...

Dr. Weiliang Qiu and Dr. Cheng Wenren

Dose–response relationships are important in assessing the efficacy and potency of drugs, which can usually be characterized by a 4-parameter logistic model: EC50, slope, lower asymptote, and upper asymptote. EC50, the concentration of a drug that induces a response halfway between the baseline and maximum, is a key quantity to evaluate drug potency. For multi-donor dose-response data, it is often the interest to estimate the overall EC50 and its 95% confidence interval. A few multi-donor EC50 estimation methods have been proposed in literature. Jiang and Kopp-Schneider (2014) systematically compared meta-analysis and nonlinear mixed-effects approaches and concluded that meta-analysis approach is simple and robust to summarize EC50 estimates from multiple experiments, especially suited in the case of small number of experiments, while nonlinear mixed-effects approach has issue of convergence failure probably due to overparameterization. In this talk, we investigated ways to improve nonlinear mixed-effects approach to alleviates its issue of convergence failure.

22 June 2023

Dr. Heng Li & Dr. Lilly Yue (FDA), and Dr. Jason (Jixian) Wang (Bristol Myers Squibb) present their recent work on propensity score analysis with discussions led by Amy Mulick.

Read more...

Dr. Heng Li & Dr. Lilly Yue (FDA), and Dr. Jason (Jixian) Wang (Bristol Myers Squibb) present their recent work on propensity score analysis with discussions led by Amy Mulick.

Presenters and Papers:
1. Heng Li, Lilly Q. Yue: Propensity score-based methods for causal inference and external data leveraging in regulatory settings: From basic ideas to implementation
https://onlinelibrary.wiley.com/doi/10.1002/pst.2294
2. Jixian Wang: To use or not to use propensity score matching?
https://onlinelibrary.wiley.com/doi/10.1002/pst.2051

14 June 2023

Standard & reference-based imputation methods based on conditional mean imputation Many RCTs compare a continuous outcome variable that is assessed longitudinally at scheduled visits between subjects assigned to an intervention vs. a control group. Missing outcome measurements may occur because subjects miss an assessment or drop out from the trial. Missing data methods based on MI are increasingly used to align the analysis strategy with the defined estimand. Imputations may be based on a MAR assumption (with or without the inclusion of time-varying covariates such as ICE indicators) or on a reference-based imputation assumption. We propose & justify deterministic conditional mean imputation (based on maximum likelihood estimation of imputation parameters) combined with the jackknife for inference as an alternative approach. In an application and a simulation study, we demonstrate that our proposal provides unbiased treatment effect estimates & correct frequentist inference with accurate SE estimation & type I error control.​

Read more...

Marcel Wolbers, Alessandro Noci 

Standard & reference-based imputation methods based on conditional mean imputation Many RCTs compare a continuous outcome variable that is assessed longitudinally at scheduled visits between subjects assigned to an intervention vs. a control group. Missing outcome measurements may occur because subjects miss an assessment or drop out from the trial. Missing data methods based on MI are increasingly used to align the analysis strategy with the defined estimand.  Imputations may be based on a MAR assumption (with or without the inclusion of time-varying covariates such as ICE indicators) or on a reference-based imputation assumption. We propose & justify deterministic conditional mean imputation (based on maximum likelihood estimation of imputation parameters) combined with the jackknife for inference as an alternative approach. In an application and a simulation study, we demonstrate that our proposal provides unbiased treatment effect estimates & correct frequentist inference with accurate SE estimation & type I error control.

Many randomized controlled clinical trials compare a continuous outcome variable that is assessed longitudinally at scheduled follow-up visits between subjects assigned to a intervention treatment group and those assigned to a control group. Missing outcome measurements may occur because subjects miss an assessment or drop out from the trial altogether. Moreover, intercurrent events (ICEs) such as discontinuations of the assigned treatment or initiations of rescue medications may affect the interpretation or the existence of the outcome measurements associated with the clinical question of interest. The ICH E9(R1) addendum on estimands presents a structured framework to link trial objectives to a precise description of the targeted treatment effect in the presence of ICEs and missing data. Missing data methods based on multiple imputation (MI) are increasingly used to align the analysis strategy with the defined estimand.  Imputations may be based on a missing-at-random assumption (with or without the inclusion of time-varying covariates such as ICE indicators) or on a reference-based imputation assumption. Reference-based methods impute missing data in the intervention treatment group based on observed data from a reference group which is typically defined as the control group of the trial. Typically, imputation is implemented using Bayesian random multiple imputation and Rubin’s rules for pooling results across imputed datasets. However, this approach requires the specification of prior distributions and MCMC sampling. Moreover, it overestimates the frequentist standard error for reference-based imputation. We propose and justify deterministic conditional mean imputation (based on maximum likelihood estimation of imputation parameters) combined with the jackknife for inference as an alternative approach. In an application and a simulation study, we demonstrate that our proposal provides unbiased treatment effect estimates and correct frequentist inference with accurate standard error estimation and type I error control. Additionally, it can result in substantially more efficient treatment effect estimators under reference-based imputation assumptions than the Bayesian approaches. A further advantage of the method is that it does not rely on random sampling and is therefore easily replicable and unaffected by Monte Carlo error. The implementation of the method in the publicly available R package “rbmi” will also be described.

14 June 2023

​In this session the speakers talk about platform trial designs covering different therapeutic areas. Jason Cooper (AstraZeneca) discusses the feasibility of a platform phase II trial for moderate-to-severe asthma, motivated by the need to offset declining recruitment rates with more adaptive and efficient patient-cantered trials. Michaela Maria Freitag (Charité – Universitätsmedizin Berlin) presents design considerations for a phase II platform trial in Major Depressive Disorder and discusses the selection of important design elements, such as allocation ratios and analysis methods as well as the operating characteristics of such trials under various realistic scenarios. Peter Jacko (Berry Consultants) concludes the session with a presentation on designing and simulating a platform trial for rare diseases, discussing rules for adding new arms, randomisation procedures, statistical analysis, and operating characteristics of the various designs.

Read more...

Jason Cooper; Michaela Maria Freitag; Peter Jacko

In this session the speakers talk about platform trial designs covering different therapeutic areas.  Jason Cooper (AstraZeneca) discusses the feasibility of a platform phase II trial for moderate-to-severe asthma, motivated by the need to offset declining recruitment rates with more adaptive and efficient patient-cantered trials.  Michaela Maria Freitag (Charité – Universitätsmedizin Berlin) presents design considerations for a phase II platform trial in Major Depressive Disorder and discusses the selection of important design elements, such as allocation ratios and analysis methods as well as the operating characteristics of such trials under various realistic scenarios.  Peter Jacko (Berry Consultants) concludes the session with a presentation on designing and simulating a platform trial for rare diseases, discussing rules for adding new arms, randomisation procedures, statistical analysis, and operating characteristics of the various designs.

14 June 2023

Major statistical challenges exist in design, analysis and decision-making in vaccine development. This session is intended to stimulate and exchange scientific knowledge exchange among individuals working on similar topics within development of prophylactic and therapeutic interventions. To that end, a range of case studies and methods will be presented, illustrating the statistical challenges posed in vaccine trials and development, and novel solutions to address them.

Read more...

Hege Michiels (Ghent University), Andrea Callegaro (GSK), Daniel Backenroth (Janssen) 

Major statistical challenges exist in design, analysis and decision-making in vaccine development. This session is intended to stimulate and exchange scientific knowledge exchange among individuals working on similar topics within development of prophylactic and therapeutic interventions.  
To that end, a range of case studies and methods will be presented, illustrating the statistical challenges posed in vaccine trials and development, and novel solutions to address them.

A per-protocol analyses is typically performed in vaccine trials, to evaluate the effect of the vaccine as soon as full immunization is expected to be reached. In the first presentation Hege Michiels (Ghent University) will present on the causal interpretation that can be given to those per-protocol analyses and proposes an alternative estimand to measure the intrinsic vaccine efficacy in settings with delayed immune response.  
In the second presentation, Andrea Callegaro (GSK) will present a case study (The Boostrix Maternal Immunization US case) in which Bayesian dynamic borrowing was used to assess vaccine effectiveness. 
In the third presentation, Daniel Backenroth (Janssen) will present on using negative control methods for assessing the success of confounding adjustments and robustness of open-label analysis with non-randomized interventions. 

14 June 2023

Very popular session from the 2023 conference where 6 speakers give short (8-9 minute) talks on non-technical topics. The speakers and talk titles were as follows: Bob Murray “Communication of Statistics: The Formula for Success”, Kimberley Hacquoil “Deal or No Deal: Make a Decision”, Emma May “Building Community in a virtual environment”, Eileen Holmes “Supporting People in FSP Roles”, Ian Ratcliffe “I Want to Fail!” and Amy Newlands “Increasing Statistics Capability in sub-Sahara Africa – The Next Wave”. ​

Read more...

Bob Murray, Kimberley Hacquoil, Emma May, Eileen Holmes, Ian Ratcliffe, Amy Newlands

Very popular session from the 2023 conference where 6 speakers give short (8-9 minute) talks on non-technical topics.  The speakers and talk titles were as follows: Bob Murray “Communication of Statistics: The Formula for Success”, Kimberley Hacquoil “Deal or No Deal: Make a Decision”, Emma May “Building Community in a virtual environment”, Eileen Holmes “Supporting People in FSP Roles”, Ian Ratcliffe “I Want to Fail!” and Amy Newlands “Increasing Statistics Capability in sub-Sahara Africa – The Next Wave”.  

14 June 2023

Health Technology Assessment (HTA) has historically been a topic for the select few statisticians specializing in HTA, and something that took place relatively siloed from regulatory statistical work. Once regulatory statisticians had argued efficacy and safety of a new treatment, HTA statisticians could take over to help quantify the added clinical and economic value. With healthcare systems increasingly moving towards a value-driven approach, HTA will demand the attention of statisticians across the product life cycle. Cross-border initiatives like EU HTA will very literally put HTA side by side with regulatory assessment, requiring statisticians to carefully consider how to integrate HTA needs in pivotal trials, and how to manage complex HTA and regulatory statistical work simultaneously. The road ahead has challenges, but also unique opportunities for statisticians to break down silos and take leadership - as experts in strategically anticipating and mitigating uncertainty across many different types of evidence.

Read more...

Anders Gorst-Rasmussen (Novo-Nordisk); Ralf Bender (IQWiG); Zoe Garrett (NICE); Justine Rochon (Boehringer-Ingelheim)

Health Technology Assessment (HTA) has historically been a topic for the select few statisticians specializing in HTA, and something that took place relatively siloed from regulatory statistical work. Once regulatory statisticians had argued efficacy and safety of a new treatment, HTA statisticians could take over to help quantify the added clinical and economic value.  With healthcare systems increasingly moving towards a value-driven approach, HTA will demand the attention of statisticians across the product life cycle. Cross-border initiatives like EU HTA will very literally put HTA side by side with regulatory assessment, requiring statisticians to carefully consider how to integrate HTA needs in pivotal trials, and how to manage complex HTA and regulatory statistical work simultaneously.  The road ahead has challenges, but also unique opportunities for statisticians to break down silos and take leadership - as experts in strategically anticipating and mitigating uncertainty across many different types of evidence.

'In the crystal ball: how the world might look for a pharmaceutical statistician when EU HTA is here' - Anders Gorst-Rasmussen, Novo Nordisk
Pan European Health Technology Assessment (EU HTA) is coming. Starting in 2025, manufacturers will be required to submit an HTA evidence dossier while EMA regulatory review is ongoing. The evidence  dossier must simultaneously address all EU member states' questions on the relative effects of a new treatment, and will form the basis of the so-called Joint Clinical Assessment (JCA). This may sound a lot like the EMA regulatory dossier and review but the scope of JCA will be very different: to support member state judgement of the added clinical and economic value of the new treatment within the context of their healthcare system. Accordingly, to ensure relevance within their health care system, a member state may ask questions involving different comparators, different populations, or different outcomes than those used in pivotal trials. And to address those questions with the available evidence, the HTA dossier may involve anything from subgroup analyses of pivotal trials, to observational studies, to network meta analyses.  The concurrency with EMA regulatory review, the very short timelines, and the sheer scope of work needed for JCA is likely to have a major impact on how we operate as pharmaceutical statisticians. In this talk, I will reflect on how the world might look for a pharmaceutical statistician when EU HTA is here.

'German requirements for direct and indirect comparisons in HTA dossiers' Ralf Bender, IQWiG, Germany
According to the law on the reorganization of the pharmaceutical market, new approved drugs have to undergo an early benefit assessment in Germany since January 2011 with the aim of a structured price regulation. The basis of the early benefit assessment is given by dossiers prepared by the manufacturer submitted to the Federal Joint Committee (G-BA), which can commission the Institute for Quality and Efficiency in Health Care (IQWiG) with an assessment of the dossier. According to the Regulation on Health Technology Assessment (EU) 2021/2282, a European joint clinical assessment of HTA dossiers comes into application in January 2025 after a 3-year implementation period. The EUnetHTA 21 joint consortium was established to develop guidance documents to support the implementation of the EU HTA regulation. The first EUnetHTA 21 deliverables are already published, e.g., the methods and the practical guideline on direct and indirect comparisons. Although the joint clinical assessment has to be taken into account in the future, each EU member state remains responsible for drawing conclusions on the added value for their health system and taking decisions on pricing and reimbursement. In this talk, an overview of the German requirements for the HTA dossiers is given with a special focus on direct and indirect comparisons.

'Where can collaboration add value to HTA – A look at the UK, Australia and Canada' - Zoe Garrett, NICE 
In May 2022 HTA agencies in the UK, Australia and Canada signed a collaboration arrangement to formalise a relationship that started during the COVID-19 pandemic. The collaboration arrangement identified 5 priority areas for collaboration including: (1) COVID19 related intelligence sharing, (2) futureproofing HTA systems, (3) collaborating with regulators, (4) work-sharing and efficiency gains, (5) digital and AI. This presentation will describe learnings from the first year of activity arising from the arrangement and where the group has chosen to focus their activities to provide most value to the agencies. It will cover activities described in the collaboration arrangement including development of science and methods and work sharing.  /  Statistical leadership: Do we hold the key to the future of HTA? Justine Rochon, Boehringer Ingelheim -  No one holds the key to the future of HTA that is fast approaching. Statisticians can, however, shape the future by treating data as a strategic asset they are responsible for throughout the entire lifecycle engagement. Learn why HTA is really a customer problem, and how statisticians can use their knowledge and creativity to solve it. Come for an informative session and leave it with some food for thought on why and how we should all learn to speak HTA statistics. The ultimate goal is to make your HTA initiatives successful and your HTA work more enjoyable in the future. 
Page:

Upcoming Events