PHASTAR at PSI conference 2017

The conference of the PSI (Promoting Statistical Insight) organisation is in London this week. Our statisticians at PHASTAR are pleased to be presenting on a number of statistical topics:

1. Our head of statistical consulting (and Honorary Professor of Clinical Trials at Manchester University) is presenting on "A unified framework for synthesis of safety data in the presence of varying exposure and risk". The talk will cover the using meta-analysis of individual patient safety data to assess risks when there are different windows of clinical relevance. A model will be presented that has been developed by the EFSPI Integrated Data Analysis SIG.
2. A presentation on the statistical considerations for safety assessment in cancer immunotherapy trials, focusing on the design, conduct and analysis challenges, with specific recommendations using case studies from approved immunotherapy products.
3. Looking at overdispersion in event rate models for asthma and COPD exacerbations in respiratory studies. We present a case study where non-constant event rates and incorrectly specified negative binomial models can dramatically influence the dispersion parameter and mislead a sample size re-estimation process. A number of strategies are described to investigate the behaviour of the dispersion process over time and under different model specifications.
4. We have a poster presentation on "Big Data" in the pharmaceutical industry, on how SAS can be leveraged for large datasets in drug discovery, genomic data analysis and real-time streaming of data from wearable devices.

In addition, we will have a stand where you can meet PHASTAR staff, and talk about how PHASTAR can help deliver your statistics, programming and data management projects, or alternatively you can find out about current vacancies for statisticians at PHASTAR.

We hope you have a great conference and enjoy one of our favourite cities.

For more details on how we can help deliver your projects, email [email protected]. For information about statistical positions at PHASTAR, email [email protected].