To recap, a cohort study recruits a group of people without disease, who are observed over a period of time to see what happens to them. They are generally prospective because you start with an exposure of interest, and follow-up participants to observe if they developed the condition or disease. As a result, this allows you to calculate incidence, which is the new cases of disease that occur during your follow-up period. Then you look at the rate at which the disease develops in a group in which a certain exposure is present compared with the rate in a group where the exposure is absent. This is referred to as a relative risk or risk ratio, which I am now going to show you how to calculate. Relative risks measured the likelihood of getting the disease if you are exposed, relative to if you are not exposed. It is the incidence in the exposed divided by the incidence in the unexposed. So, how do we calculate the relative risk? As with case-control studies, the first step is to construct a two-by-two table using your study data. Here you have your cases and non-cases along the top of the table, and the exposure status along the left-hand side. Once you have sorted in the required data, you should label the table A, B, C, D. One thing to note is that in cohort studies, we refer to controls as non-cases due to the study design. Any non-case in a cohort study has the potential to become a case if you wait long enough. However, you will see some studies refer to non-cases as controls. Before you calculate the relative risk, you must calculate the incidence or risk of disease in both the exposed and the unexposed groups. The incidence in the exposed group is the total number of exposed cases or A divided by the total number of exposed people A plus B. Similarly, the incidence of disease in the unexposed group is the number of unexposed cases C divided by the total unexposed people C plus D. To calculate the relative risk or the risk ratio, you simply divide the incidence in the exposed by the incidence in the unexposed. From this, we can then easily calculate the attributable risk. This is also referred to as the risk difference. To calculate this, you simply subtract the incidence in the unexposed from the incidence in the exposed. This is expressed per 1,000 or 10,000 exposed individuals. For example, this could be something like 150 per thousand smokers develop lung cancer because they smoked. So, how do you interpret relative risks? For example, what is the relative risk of five mean? This can be interpreted as the exposed individuals are five times more likely to develop the outcome of interest compared to the unexposed individuals. In general, if the relative risk is equal to one, it suggests there's no difference in risk between the exposed and the unexposed groups. A relative risk greater than one, suggests an increased risk of the outcome in the exposed group, and a relative risk less than one, suggests a reduced risk in the exposed group. How do we calculate the 95 percent confidence intervals? As for the odds ratio for case-control studies, the 95 percent confidence interval is calculated for the relative risk. You will need to know this equation. If the 95 percent confidence interval includes one, the relative risk is not statistically significant. Now, let's briefly look at risk versus rates in the context of cohort studies. So, we have already described this when we calculated the incidence or risk of disease in the exposed and unexposed groups, and subsequently, the relative risk or risk ratio and attributable risk. However, if you want to look at person years, you can calculate rates. So, your two-by-two table becomes the number of cases in the person years. You calculate the incidence rate in the exposed by taking the number of cases that were exposed divided by the person years. The incidence rate in the unexposed is the number of cases in the unexposed divided by the person years. The incidence rate ratio can then be calculated as the incidence rate in the exposed divided by the incidence rate in the unexposed. Now, you have learned how to calculate and interpret measures of association using cohort study data. Next, we will put these skills into practice and work through calculations using real data.