We are all living in the era of globalization and, like it or not, it is going to change the way we practise epidemiology, the kinds of questions we ask and the methods we use to answer them. However, the methods, and ways of thinking about the health of populations, that will be required for epidemiology in the 21st century are in some instances quite different from the standard epidemiological techniques that are taught in most textbooks and courses today. As we develop epidemiological methods for addressing the scientific and public health problems of the 21st century, it is important that we consider, once again, the distinction between the analysis of variance and the analysis of causes. This has primarily been considered with respect to genetic research, and also with regard to the problems of making comparisons between different populations and environments at the same point in time. It has not been considered in depth with regard to the issues of conducting epidemiological research in a world that is changing over time. In this article, I first consider the statistical and scientific issues involved in the distinction between the analysis of variance and the analysis of causes. I then discuss some examples of the implications of this distinction for the theory and practice of epidemiology in a changing world, particularly with regard to risk factors that become ubiquitous over time. Sometimes the most important causes of disease are invisible because they are everywhere.