In the case control study you begin with the outcome of interest and then estimate exposure. Patients are selected on disease status and we cannot calculate incidence based on exposure. So when we are estimating the risk in a case control study, we calculate the odds ratio, or the likelihood of having the exposure if you have the disease, relative to if you do not have the disease. I'm now going to show you how to calculate an odds ratio. The first step is to construct a 2 by 2 table using your study data. Remember everyone in the study is either a case, or a control and they're either exposed or unexposed. We will put the outcome of interest, or the cases and controls, along the top of the table and the exposure status on the left hand side. Once you have slotted in the required data you should label the table A B C D. So how do you calculate the odds ratio? First let's look at the total number of cases, that is A plus C. And then the total number of controls is B plus D. The next step is to identify the total number of people exposed this is A plus B, and the total number of people not exposed, C Plus D. The odds ratio can be defined as the odds of exposure among the cases over the odds of exposure among the controls. Referring back to the 2 by 2 table, we estimate the odds of exposure among the cases by taking the exposed cases or A and dividing it by the total number of cases, or A plus C, We then divide this by the non-exposed cases C, divided by the total number of cases, A plus C. So we now have the odds of exposure amongst the cases. Now we do the same for the bottom line of the equation to get the odds of exposure among the controls, that is B, the exposed controls, divided by the total number of controls B Plus D. This is then divided by the non-exposed controls D, divided by the total number of controls, B plus D. This calculates out as A over C, Divided by B over D. Which subsequently gives A times D, Divided by C Times B. This is the formula that you will need to know. However one important thing you must remember is that the equation here is based on the set up of this 2 by 2 table. It is based on the exposed cases being in box A and the exposed controls being in box B et Cetera so be careful of the setup of any 2 by 2 table. We examined the odds ratio as the odds of exposure among the cases divided by the odds of exposure among the controls. but the odds ratio can also be thought of as the odds of disease among the exposed, divided by the odds of disease in the non-exposed. Regardless of whether you take the angle of odds of exposure or odds of disease, the odds ratio is the exposed cases by the non-exposed controls, divided by the exposed controls by the non-exposed cases and you will get the same answer for both approaches. You can make inference about the disease based on the exposure status. For example, what does an odds ratio of 4 mean? An odds ratio of 4 means that for individuals who are exposed to the risk factor of interest, the odds of disease is 4 times greater than the odds for those who are unexposed. What about the odds ratio of 0.7? For individuals who are exposed to the odds of disease is 30 percent less than the odds for those who are unexposed. And what about an odds ratio of 1. This leads us to the correct interpretation of odds ratios. If the odds ratio equals 1, this means there is no association. If the odds ratio is greater than one, this indicates a positive association between the exposure and disease. That is that the exposure is associated with an increased risk of disease. On the other hand if the odds ratio is less than one there is an inverse association between the exposure and the disease, and the exposure reduces the risk of disease. However we know we cannot go on the odds ratio alone, so let's move on to statistical significance. We have two things we can look at here. Firstly the p Value, which is the probability that you would get your results by chance alone. As we know, a p value less than 0.05 is generally considered as the threshold to determine that chance is not likely to explain your results. The second thing to determine is the 95% confidence interval. This is a range of values in which the true value will be found 95 percent of the time. Large samples yield small confidence intervals and small samples yield large confidence intervals. This is because with larger sample sizes you have better precision. So how do we calculate the 95 percent confidence interval? The first step is to look at the variance. To do that, we go back to the 2 by 2 table and we calculate 1 over A plus 1 over B Plus 1 over C Plus 1 over D. So then, for the calculation, you take your log odds ratio plus or minus 1.96 the square root of the variance. So if the 95 percent confidence interval includes 1, the odds ratio is not statistically significant. You need to know this equation for calculating the 95 percent confidence interval. So now we've covered how to calculate the odds ratio in a case control study, and we've looked at how to interpret the odds ratio. Next you will put these skills into practice and work through calculations using data.