Statisticians in the Pharmaceutical Industry

The Association for Clinical Data Management (ACDM) and Statisticians in the Pharmaceutical Industry (PSI) publish an excellent document called Computer Systems Validation in Clinical Research A Practical Guide, which can be found at http //www.cr-csv.org/. 

Statisticians in the Pharmaceutical Industry (PSI), at http //www.psiweb.org. They have published the excellent Guidelines for Standard Operating Procedures for Good Statistical Practice in Clinical Research, at http //psiweb.org/pdf/gsop.pdf. 

J. McKellar. 1995. Assessment of dose proportionality Report from the Statisticians in the Pharmaceutical Industry/Pharma-cokinetics UK Joint Working Party. Drug Information J. 29 1039-1048. 

It should be clear that the purpose of this chapter is not to review the normal role of the statistician in the pharmaceutical industry today. Instead, this chapter is intended to challenge the status quo and to highlight some critical problems and largely unmet needs that logically intersect both the expertise and the sphere of influence of professional statistician in the pharmaceutical industry. In some ways the intention is to revisit the basic tenets of experimental design and analysis to see where we have drifted away from sound scientific principles, and where we may have unexplored opportunities for the future—a future certain to be different from the past. 

Once the study is completed and the plasma or serum samples are analyzed for drug concentrations and AUC and Cmax are determined for each subject. AUC and Cmax are treated as the dependent variables (Y) in the analysis. At this point, there are a number of ways to assess for dose proportionality. The Statisticians in the Pharmaceutical Industry/Pharmacokinetics UK Joint Working Party (SPI/PUK JWP) have reviewed the statistical methods used to assess dose proportionality and have published a summary of their findings (Gough et ah, 1995). These methods will now be summarized. In the untransformed regression approach, the simple linear model is fit to the data... 

Gough, K., Hutchison, M., Keene, O., Byrom, B., Ellis, S., Lacey, L., and McKellar, J. Assessment of dose proportionality Report from the statisticians in the pharmaceutical industry/pharmacokinetics UK joint working party. Drug Information Journal 1995 29 1039-1048. 

The medical statistician also has the opportunity to join a number of specialist societies. The International Society for Clinical Biostatistics organizes an annual conference, as does the Society for Clinical Trials, and proceedings of these meetings are published in Statistics in Medicine and Clinical Trials, respectively. There are a number of organizations specifically for statisticians working in the pharmaceutical industry, such as the UK based PSI (Statisticians in the Pharmaceutical Industry) and there is a European Federation of Statisticians in the Pharmaceutical Industry (EFSPI), of similar national societies. Other national organizations such as the American Statistical Association and the French Association pour la Statistique et ses Utilisations have specialist sections. 

EFSPI Working Group (1999) Qualified statisticians in the European pharmaceutical industry report of a European Federation of Statisticians in the Pharmaceutical Industry (EFSPI) working group. Drug Information Journal 33 407 15. 

When the first edition of this book appeared in 1997, it was still the case that many statisticians in the pharmaceutical industry adhered to the type III philosophy. Since then the espousal of a type II philosophy seems to have become more common. I would like to think that this book played its part in this evolution. There have, however, been other influential statements in favour of the type II philosophy. A particularly clear account is given by Gallo (2001). 

In spite of its current popularity in the pharmaceutical industry, the use of two control groups is opposed by some statisticians on the grounds that a significant difference between the two groups may indicate that the study was compromised by excessive, uncontrolled variation. Haseman et al. (1986), however, analyzed tumor incidence data from 18 color additives tested in rats and mice and found that the frequency of significant pairwise differences between the two concurrent control groups did not exceed that which would be expected by chance alone. 

Many of the quality improvement goals for implementation of PAT in the pharmaceutical industry have been achieved by companies in other industries, such as automobile production and consumer electronics, as a direct result of adopting principles of quality management. The lineage of modern quality management can be traced to the work of Walter Shewhart, a statistician for Bell Laboratories in the mid-1920s [17]. His observation that statistical analysis of the dimensions of industrial products over time could be used to control the quality of production laid the foundation for modern control charts. Shewhart is considered to be the father of statistical process control (SPC) his work provides the first evidence of the transition from product quality (by inspection) to the concept of quality processes [18,19]. 

The chapters can be used in a number of ways by junior statisticians in order to get a quick overview of issues affecting a particular type of trial they are working on for the first time, by physicians and other life-scientists working in the pharmaceutical industry to help them discuss statistical issues with their statistical colleagues, and by university departments as the basis for student seminars and journal clubs. 

At rather infrequent intervals, statistical issues in the pharmaceutical industry have been debated by the Royal Statistical Society. Three read papers contain much of relevance to the work of the pharmaceutical statistician. Even though the first (Lewis, 1983), appeared nearly twenty-five years ago (at the time of writing), it is interesting to note that a number of the issues are still live. The second paper (Racine et al., 1986) is an early exposition of the use of Bayesian methods within the pharmaceutical industry, which, despite the fact that it pre-dates the Markov chain Monte-Carlo revolution introduced by Gelfand and Smith (1990) is still ahead of current practice in many respects. The third (Senn, 2000a) was written by me and, as might be expected, touches on a number of issues covered in this book. 

This Is an issue where consensus now seems to have been achieved. An ingenious theorem due to Fieller (Fieller, 1940, 1944) enables one to calculate a confidence interval for the ratio of two means. The approach does not require transformation of the original data. (Edgar C. Fieller, 1907—1960, is an early example of a statistician employed in the pharmaceutical industry. He worked for the Boots company in the late 1930s and 1940s.) For many years this was a common approach to making inferences about the ratio of the two mean AUCs in the standard bioequivalence experiment (Locke, 1984). 

There have been many developments since the first edition of this book and it was high time for a second. My own period working in the pharmaceutical Industry is now a distant memory but the ten years working as an academic since the first edition has had its compensations. I have been fortunate enough to be able to consult for many pharmaceutical companies during this time and this has certainly widened my appreciation of the work that statisticians do within the industry and the problems they face. 

I continue to hope, of course, that this book will aid dialogue between statisticians and life-scientists within the pharmaceutical industry but also hope that it will contribute to a wider appreciation of the interesting challenges that statisticians within the pharmaceutical industry face and the seriousness with which they are met. I hope that the reader finds both stimulation and enjoyment in encountering these challenges. 

Finally, this book is written for the graduate students or postdoctoral fellows who want to specialize in pharmacometrics and for pharmaceutical scientists, clinical pharmacologists/pharmacists, and statisticians in academia, regulatory bodies, and the pharmaceutical industry who are in pharmacometrics or are interested in developing their skill set in the subject. 

---