As an example, consider again the outbreak of Legionnaires’ disease that occurred in Louisiana described at the end of Step 6. Investigators enrolled 27 case-patients into a case-control study. They also enrolled two controls per case, a total of 54 controls. Using descriptive epidemiology, the investigators did not see any connection with the town’s various cooling towers. Using analytic epidemiology, the investigators determined quantitatively that case-patients and controls were about equally exposed to cooling towers. However, case-patients were far more likely to shop at a particular grocery store, as shown in the following two-by-two table.