Two methods of assessing the pre-test probability of pulmonary embolism (PE) are compared in this paper. The Well's score is an established “prediction rule” based on six clinical parameters while the “Geneva score” uses only objective measurements such as age and arterial oxygen tension to assess pre-test risk, both systems attributing patients to low, medium, or high risk groups. However, when clinical judgement disagreed with the objective Geneva score, a third diagnostic strategy—the “Geneva rule with clinical override”—was invoked. Two hundred and seventy seven patients admitted with suspected PE were assessed using the two scoring systems and the diagnosis of PE was made by a combination of d-dimer, venous ultrasonography, helical CT scanning, and pulmonary angiography with a positive diagnosis in 77 cases. The Geneva score overridden by implicit clinical evaluation performed significantly better than either score alone.

Despite developments in diagnostic methodology, the effective diagnosis of PE remains reliant upon the clinical skills of the attending physician.