Alfredo Morabia’s Enigmas of Health and Disease, which is aimed at members of the lay public who have an interest in epidemiology, is an exceptionally informative and engaging book that covers landmark events in the history of epidemiology while at the same time discussing the discipline’s concepts and strategies. Its leitmotifs are the importance of group comparisons for causal inference (in contrast with the focus on individuals) and the interface of epidemiology with public health and medicine. Faithful to the paraphrase attributed to Einstein— ‘‘Things have to be made as simple as possible, but not simpler’’—Morabia uses a simple and elegant language without sacrificing depth. He states that ‘‘epidemiology is infamously arcane, even though, as this book shows, there is no intrinsic complexity to it. On the contrary, the principles of comparing groups of people involve no more than simple logic’’ (page 184). Although some may disagree with the statement—for example, to the extent that it ignores the complexity of inferring causality based on epidemiologic findings—epidemiology’s concepts and methods are conveyed with such clarity that it is hard to believe that any lay person would find it difficult to follow most of the author’s explanations. Throughout the book, Morabia emphasizes the importance of differentiating between belief and knowledge while cautioning the reader against making unwarranted inferences from knowledge. Take, for example, his statement in the Prologue: ‘‘I used the term knowledge as opposed to beliefs. Don’t read it as a synonym of truth, but rather of evidence’’ (page XIV). Even though many epidemiologic concepts and methods are not explicitly named, Morabia manages to describe them using many interesting examples. On page 200, for instance, he explains that, ‘‘intuitively, it may seem that treating fewer people but focusing in those with a high risk of heart attack...would decrease the NNT (sic, number needed to treat)...this is not necessarily so: twice as many cases of lung cancer occur among smokers of a pack or less per day than among those who smoke more than a pack per day. Why? ... The risk among light and moderate smokers is smaller than that of heavy smokers, but they are more numerous. A small risk applied to a large number of people can generate more excess cases than a large risk that is applied to a small fraction of the population.’’ This is an exquisite description of the population attributable risk! Other key concepts in epidemiology covered in this book include differences between observational and experimental epidemiology (randomized trials), designs in observational epidemiology (including quasi-experimental studies, as exemplified by Snow’s study of water supplying companies; page 53), efficacy, number of individuals who need to be treated to prevent one case of the disease or one death, bias (including inferential bias; see Chapter 16), external validity, and issues of comparability in observational studies (confounding effects). The book also emphasizes the difficulties in applying population (group) findings to individuals, which probably explain why