To recap, in a case control study, you begin with disease status and go on to estimate exposure. So you already know who has the outcome, and hence, you can now calculate incidence based on exposure. When you estimate the risk in a case-control study, you estimate the likelihood of having the exposure in those who have the disease, relative to those who do not. Now, I will work through an example of how to calculate and interpret the odds ratio in a case control study, examining the association between smoking and lung cancer. Let's imagine 200 patients with lung cancer, and 200 patients without lung cancer, were enrolled into a study. Using a survey, patients were asked about their tobacco use over the past 20 years, 180 of the lung cancer patients reported smoking, and 20 patients without lung cancer reported smoking. The aim of this study is to test the association between smoking and lung cancer, and to do so by using a quantitative measure of the association. Firstly, how do we know this is a case control study? Well, this particular study takes individuals with and without the disease and goes back in time, 20 years in this case, to check for an exposure. As you know, this is a key characteristic of case-control studies. To take the disease of interest, and explore potential exposure or risk factors. Here, you need to calculate an odds ratio, which is exactly what it implies, a ratio of the odds. In this case, it is the odds of being exposed to cigarette smoke in those with lung cancer, over the odds of being exposed to cigarette smoke in those without lung cancer. Okay, let's start by constructing a two-by-two table. By convention, disease status goes along the top of the table. Here, that is your cases. Lung cancer patients under controls, patients without lung cancer. The exposure status in this study, smoking status, goes along the side. You know 180 of the 200 patients with lung cancer smoked, and 20 of the 200 who did not have lung cancer smoked. If you slot this information into the table and complete the blanks, it should look like this. After completing the two-by-two table, you can now calculate the odds ratio. To do so, you simply divide the odds of exposure among cases, by the odds of exposure among controls. The odds ratio in this study, is simply the odds of smoking among those with lung cancer, divided by the odds of smoking among those without lung cancer. The odds of exposure among the cases, is the ratio of exposed cases to unexposed cases. Here, that is 180 divided by 20 which is nine. Similarly, the odds of exposure and the controls is the ratio of exposed controls to unexposed controls. In this case, 20 divided by 180 which is 0.111. Next, you divide these, so nine divided by 0.111, and you get an odds ratio of 81. That means if you had lung cancer, you were 81 times more likely to smoke compared to if you didn't have lung cancer. There appears to be a strong positive association between smoking and lung cancer. A simpler way to remember how to calculate the odds ratio however, is to use the cross products of the boxes in your two-by-two table. If you label your boxes A, B, C and D, that is, A multiplied by D over B multiplied by C. This would give you 180 multiplied by 180, over 20 multiplied by 20. You end up with 32,400 divided by 400, which equals 81. As you have seen, calculation odds ratios is simple. Hopefully, you now understand where the numbers come from, when the case-control studies report an odds ratio of for example 81.