The fattening of the human species and the accompanying emergence of the metabolic syndrome and of type 2 diabetes as remarkably frequent clinical entities are among the major epidemiologic events of our time. Control of the diabetes epidemic requires a greater understanding of the pathophysiologic processes underlying these phenomena. Many epidemiologic studies have now shown associations between inflammation markers and diabetes, with the most consistent being for leukocytes and the strongest being for C-reactive protein. Consistent protective associations have also been reported for adiponectin, an adipocyte secretory protein with antiinflammatory actions. Although great variability is seen between reported associations, as a whole these studies suggest a role for inflammation linked to obesity. The variability reported is in part due to differences in model adjustment, in how diabetes was ascertained, and in the different means used to operationalize the concept of low-grade chronic systemic inflammation. It is also due, in part, to sample characterization, as findings are heterogeneous across some subgroups, such as those defined by smoking. Consistent with their association with type 2 diabetes, inflammation markers have also be shown to predict conditions present in the prediabetes state such as weight gain, hypertension, gestational diabetes, and decline in insulin sensitivity.