Editorial

Smoking and coronary disease

During the late 1950s Sir Ronald Fisher questioned the already popular, but in his view precipitate, causal interpretation of the association between smoking and lung cancer. His pungently expressed views began a controversy that has smouldered and sometimes flared ever since. The most recent attack on Fisher's constitutional hypothesis was launched by Reif and in this paper I consider the validity of his criticisms. A range of evidence shows that it is not yet possible to distinguish between constitutional and causal-plus-constitutional interpretations although recent studies indicate that a pure causal hypothesis is incapable of explaining the full association as observed in Western populations. Unfortunately, errors of diagnosis and death certification still impede the rigorous testing of adequately formulated hypotheses.