In a landmark article published almost 20 years ago McGinnis and Foege showed that the actual leading causes of death in the United States were not cardiovascular disease and cancer, which had long headed the “leading causes” rankings calculated from death certificate analyses.1 Using attributable risk data cobbled together from a number of sources, they estimated that smoking, with 400 000 deaths a year, and disease related to diet and lack of physical activity, with 300 000 deaths, were in fact the leading killers of Americans, between them causing about a third of all deaths in 1990.

The headlines then were all about how smoking was at the top of the list and that almost half of that year’s deaths were a result of this and other preventable, behaviour related causes. Many people were also surprised at the huge toll taken by poor diet and lack of physical activity, but it wasn’t a focus of discussion.

The intervening decades have been a terrific success story for antitobacco efforts. As a …