Epidemiology studies how disease is distributed across populations (by place, time, and group) because systematic variation in who falls ill points toward the causes. A disease that clusters in one occupation, season, or neighbourhood is signalling a determinant, so the pattern of occurrence becomes evidence about the mechanism of occurrence.
To measure how often disease strikes, epidemiology divides the number of new cases by the total person-time at risk, not by a simple headcount, so that people observed for different lengths of time are weighted by their actual exposure. This lets populations of different sizes and follow-up durations be compared on a common footing, separating how much disease there is from how many people there are.
A statistical link between an exposure and a disease can arise from chance, bias, or confounding rather than from one causing the other, so observing correlation is never sufficient to claim cause. Epidemiology judges causality by weighing considerations such as strength, consistency across studies, dose-response, temporal order, and biological plausibility together, since no single study or single criterion settles it.
An apparent association between exposure and disease can be produced entirely by a third factor that is linked to both, as when coffee drinkers also smoke and smoking causes the disease. Unless such confounders are measured and adjusted for, the study will credit the innocent exposure with an effect that actually belongs elsewhere.
A risk factor that raises any one person's chance of disease only slightly can still cause the bulk of cases when the whole population is exposed to it, because a small risk multiplied across millions outweighs a large risk confined to a few. This reframes prevention: shifting the entire population's exposure distribution can prevent more disease than targeting only the high-risk individuals.
In 1854 London, Snow mapped cholera deaths, traced their concentration to a single public water pump, and had its handle removed, demonstrating a waterborne cause decades before the bacterium was identified. The case shows that careful study of distribution can establish a cause and guide effective action even without knowing the underlying biological agent.