Graham, D.J. et al. 2005. Risk of acute myocardial infarction and sudden cardiac death in patients treated with cyclo-oxygenase 2 selective and non-selective non-steroidal anti-inflammatory drugs: nested case-control study. The Lancet 365: 475-481. Don't attempt to read this from start to finish. It's dense! We will spend the next several weeks on this one paper. Let the following questions be your guide. Jump around. Read little bits. And ponder each sentence you do read. A scientist would never read a paper from from start to finish, and you should not either. Focus on the Abstract, Introduction and Discussion. We will look at the Methods and Results in future classes.
Pease and Bull. 2000. Science for Business, Law and Journalism. Chapter 4. Models are the building blocks of science. Chapter 5. All models are false. But some are still useful.
2. What is the goal, study design and conclusion of Graham's paper? That is, what did he want to learn, how did he go about studying this problem, and what did he conclude? Use language that the Governor of Vermont can understand.
3. How did Graham's paper build on previous studies undertaken by Merck? Hint: See the Discussion.
4. What are the key models that Graham uses in his paper. Give an examples of physical, abstract and sampling models.
5. What exactly is a model? Why did Graham et al. need to employ models? Does not this leave them open to unnecessary criticism? More generally, why is it necessary for scientists to employ models?
6. Suppose Merck contracted with you to critique the Graham et al. paper. What is the most substantive criticism you could raise?