Subselecting Patients at Risk for Pulmonary Embolism

To the Editor. —Several misconceptions may result from the recent report of pulmonary embolism in women and men in the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study.1In this study, risk factors were compared for men and women suspected of having pulmonary embolism. It is valid to compare the presence of risk factors in men and women who have the disease (eg, positive angiogram). However, when one compares these risk factors for those suspected of a disease, the results are biased by the physician's threshold for suspicion of disease, which includes the risk factors under study. The comparison of the frequency of a risk factor in men and women with definite pulmonary embolism with those suspected of pulmonary embolism but with a negative angiogram is invalid owing to selection bias. The authors acknowledge the problem of a biased denominator and concede that suspicion of pulmonary embolism and, hence,