Looking to the 21st century: have we learned from our mistakes, or are we doomed to compound them?

Abstract

First I would like to dedicate this talk to the memory of my beloved friend, Dennis Slone, who died much too young, and much too unrecognized, in 1980. I missed him then, I miss him now and I will miss him for the rest of my life, as I know will a handful of other people present today. Next, I would like to thank the organizers, particularly Sam Lesko, a colleague and friend with whom I had the pleasure of working for many years, for inviting me to give one of the keynote addresses at this conference. The invitation was particularly gracious, since it is no secret that at its inception I opposed the establishment of a pharmacoepidemiological society, and that I have always questioned the need for pharmacoepidemiology as a discrete discipline, on esthetic, political and conceptual grounds. To be gracious in return, I will not discuss my esthetic or political objections until the coffee break; the conceptual objection I will return to at the end of this talk. I must confess that when I selected my topic, I did not appreciate that I had bitten off more than I could chew. When I began to think about it, I realized that the subject matter is exceedingly varied and complex. In Figure 1, I list some of the domains in which I believe there is a need to come to grips with serious limitations to the way in which some approaches have been applied, or misapplied, to the evaluation of drug effects – and that list is by no means exhaustive. If one is to be serious about covering everything, a monograph, rather than a single lecture, is needed. In 1967, Richard Doll published a sponsored monograph, introduced by a public lecture, on the prevention of cancer. I suggest that an analogous model may be needed, and this, if you like, is my public lecture in which I will concentrate only on Item 1, skepticism as a scientific principle. A bit of history first. This conference is taking place just short of 37 years since I arrived in the United States in September 1967 and switched from clinical medicine to epidemiology, more specifically to the exploration of epidemiological methods that could be applied to drug surveillance, then an ‘almost virgin’ subject. When I arrived, the first colleagues with whom I was associated in that effort, Herschel Jick and Dennis Slone, had only been at it for a year or two and at the time that I joined them all three of us were still learning epidemiology, ‘on the road’, as it were. And when it came to the application of what we were learning to drug surveillance, we were for the most part not only groping in the dark, but also groping in the wrong place. We were studying hospitalized patients while the main public health issues concerned drug safety in the population at large. Like the fabled drunkard, we were not looking for the penny where it dropped but where the street-lamp cast some light. Then, to labor the metaphor, as we sobered up and as dawn approached, we began to look further by adding casecontrol methods to our armamentarium and we succeeded in finding a penny or two, after all. I must quickly add that we also found a slug or two. I have described that experience at length and in rather less florid language elsewhere. But what is relevant here is that as we gained experience, Jick on one hand and Slone and I on the other came to differ more and more about how epidemiology should