ABSTRACT
Introduction Denny walked into my office one day after the reports for Example
3.2 came out looking like he had been run over by a bus and dragged over hot coals. He had been (figuratively) when he reviewed the findings with senior management. They obviously did not like the implications for getting together a marketable formulation in time for filing with the FDA. Nobody ever comes to see you when you release findings they like. That
annoyed me when I first started on the job, but after a while I realized it gave one more time to enjoy the moment. Do take time out to enjoy the good moments on the job. Given the suc-
cess rate of drugs in clinical development (see Chapter 1), statisticians should expect to be the bearer of bad news on the majority of occasions in their working life. This is ok if you are in an organization that recognizes that failure is far more common in drug development than success, but if you are not, grow a thick skin about such matters, or think about changing jobs. Be careful not to get cynical, though. It is an easy trap to fall into and causes one to not enjoy anything (because you always think about the bad thing that is probably right around the corner and guard against keeping your hopes up). Probabilistically speaking, there will be good moments on the job, and one should maintain one’s equanimity so that one can enjoy them. The question Denny posed to me was simple on the surface - can
we explore these data to see if there was any possibility of a follow-up bioequivalence trial being successful? Note the careful use of the word ‘we’. When a clinician uses ‘we’
with a statistician, it is the royal ‘we’ which can be usually translated as meaning ‘you’. I told him that yes, I could, but given the findings of Example 3.2, my
intuition told me that it was going to be pretty unlikely and that he had better prepare his folks for that message. I would run some programs and get back to him with a quantitative assessment next week. He wanted it sooner, but I told him no.
