COMMENTARY When Can Risk-Factor Epidemiology Provide Reliable Tests?
Can we obtain interesting and valuable knowledge from observed associations of the sort described by Greenland and colleagues 1 in their paper on risk factor epidemiology? Greenland argues “yes, ” and we agree. However, the really important and difficult questions are when and why. Answering these q...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.363.69 http://www.phil.vt.edu/dmayo/personal_website/(2004) When Can Risk Factor Epidemiology Provide Reliable.pdf |
| Summary: | Can we obtain interesting and valuable knowledge from observed associations of the sort described by Greenland and colleagues 1 in their paper on risk factor epidemiology? Greenland argues “yes, ” and we agree. However, the really important and difficult questions are when and why. Answering these questions demands a clear understanding of the problems involved when going from observed associations of risk factors to causal hypotheses that account for them. Two main problems are that 1) the observed associations could fail to be genuine; and 2) even if they are genuine, there are many competing causal inferences that can account for them. Although Greenland’s focus is on the latter, both are equally important, and progress here hinges on disentangling the two to a much greater extent than is typically recognized. A THEORY-DOMINATED VERSUS A NEW-EXPERIMENTALIST PHILOSOPHY In advocating the role of observational studies as the provider of facts from which theory follows, Greenland and his colleagues allude to “one popular philosophy of science”—presumably Popper’s, but, in fact, Popper was the exemplar of a “theory-first” philosopher. Nevertheless, Popper’s demand is wrongheaded (except in the sense that one begins with an interest or problem—one doesn’t just “observe”!); moreover, Popper denied that even the best-tested theory could be regarded as reliable. So the question arises as to which philosophy of science best supports the position of Greenland et al. We think the most congenial would be the “new experimentalism, ” in which experiment and statistical analysis could “have lives of their own ” quite independent of substantive scientific theories. 2,3 Despite their championing of “purely descriptive (atheoretical) approaches, ” it is evident by their emphasis on “precision and replication ” and “precise null results ” that Greenland et al. have in mind, not “mere observations, ” but rather something more in line with what has been called “experimental or statistical knowledge”: knowledge of genuine ... |
|---|