Aims: A career in pharmacometrics, biostatistics and computational biology attracts individuals who enjoy working with mathematical equations, data and writing computer programmes, but at the same time are interested in collaborating with clinicians and laboratory scientists on problems in medicine and biology. There is a lot of overlap in the knowledge and skill set required for each of these careers, highlighting the importance of integrating postgraduate training. The aim of this presentation is to review current course offerings worldwide to see what cross-disciplinary courses are being offered to students.
Methods: As part of the redevelopment of the University of Melbourne Master of Biostatistics, I recently reviewed postgraduate biostatistics courses at leading Universities in the US and UK (e.g. Harvard University, and London School of Hygiene and Tropical Medicine). This was carried out to ensure the core and elective subjects offered at the University of Melbourne, were benchmarked against market leaders. Furthermore, to offer multiple career pathways for graduates, I reviewed other postgraduate degrees in disciplines that also attract students with an aptitude for mathematical sciences and an interest in biomedical research (in particular, the emerging discipline, Computational Biology) to expand the offering of elective subjects.
Results: A pharmacometrics subject was not offered as an elective in any of the Master of Biostatistics courses reviewed. All courses offered a core or elective subject on the analysis of longitudinal (repeated measures) data, however, these subjects only covered linear mixed-effects modelling and generalised linear mixed-effects modelling (e.g. where there is a nonlinear link function for binary outcomes that is linearly associated with the regression model parameters). Non-linear mixed-effects modelling, the main statistical method applied in population pharmacokinetic-pharmaocdynamic modelling, requires knowledge of specialist programmes which are tailored for solving ordinary differential equations (e.g. NONMEM) and have flexible algorithms (e.g. MONOLIX) to circumvent convergence issues. With regards to other discipline Master degrees, a google search only revealed a Master of Science in Pharmacometrics at the University of Maryland, US. The Master of Computational Biology emerged as a growing market with degrees now offered at many Universities worldwide. The inter-disciplinary nature of computational biology was highlighted with students required to complete compulsory biostatistics, biology and computer science subjects.
Conclusions: Currently there is very little cross-over in postgraduate training of biostatisticians and pharmacometricians, with pharmacometrics training primarily offered as a short course taught by Pharmacometrics groups (e.g. PAGANZ offer annual short courses). Postgraduate coursework degrees in biostatistics and computational biology share core and elective subjects to ensure students have the necessary depth and breadth for their future careers. Why is pharmacometrics absent from many of the programmes, when clearly pharmacometricians need adequate statistical knowledge and computational skills as well as an understanding of clinical pharmacology? Is the current training model for pharmacometricians sufficient, or should there be more bridging with postgraduate courses in biostatistics and computational biology? I welcome a lively debate.