An epidemiologists’ toolkit

 The measure of association is among the most useful items in an epidemiologists’ toolkit. It can be the critical piece of information to judge whether an exposure is likely to be a causal factor in disease occurrence. However, our ability to judge what is useful for epidemiologic inference is driven by what measures of disease frequency can be estimated in a study. You’ve already learned the study designs, so use this knowledge in participating in this discussion. In a cross-sectional study, a survey of health concerns was conducted in a community where half the residents live within two miles of a power plant. Residents self-report their health status on the day they receive the survey in the mail, over a period spanning about a week. The investigators analyze the information and conclude that individuals who live nearest to the power plant have greater risks of asthma, neurologic disease, toothache, and headaches than those who live farther away. They conclude that exposure from the power plant is the likely cause of these health outcomes. Given what you’ve already learned in previous modules about the appropriate measure of disease frequency in a cross- sectional study, do you agree with the investigator’s conclusion that the development of these outcomes are linked to the power plant? Why or why not? Do you think the approach of comparing individuals who live close to the power plant and those who live farther away makes sense? Why or why not? Now applying what you’ve learned from the readings and videos, let’s go through the rest of the discussion together, one question at a time. The utility of relative versus absolute measures of effect in determining public health interventions has long been a question of epidemiologists. You’ve learned in this module that both are useful in a general sense, but the type of information they provide may be importantly different. Therefore, the interpretation of one independent of the other may directly influence intervention relating to social inequality. King et al. provide a nice summary of the use of relative versus absolute measures of effect in an important body of epidemiologic literature — health inequalities. Consider whether inequality is an endpoint that requires one or both of these measures to fully understand. King et al. suggest that the majority of studies report ratio measures of effect in relation to this outcome. Discuss the advantages or disadvantages of a study reporting and interpreting only the relative or only the absolute measure of effect. . How might a researcher reconcile when the absolute and relative measures of effect lead to opposite conclusions? Consider how this issue applies to the topic (DIABETES) you have selected for your final paper, and discuss which of them is most meaningful. Use the first paragraph of your response to summarize your final topic so the class has enough information to understand your choice.