Background Clinical decision rules for suspected pulmonary embolism are proposed to identify patients suitable for discharge without radiological investigation. Their use varies between institutions. Aims To quantify unnecessary radiological investigations for suspected pulmonary embolism (PE) as defined by a newly proposed three‐tiered clinical decision rule incorporating the revised Geneva score, Pulmonary Embolism Rule‐Out Criteria and D‐dimer. To quantify missed diagnosis of PE if the proposed clinical decision rule were followed. Methods A retrospective audit was conducted; applying the proposed clinical decision rule to 584 emergency department (ED)‐based encounters at the Royal Adelaide Hospital from May to November 2015. Encounters were confined to emergency presentations where suspected acute PE was investigated with computed tomography pulmonary angiography or ventilation‐perfusion scanning; inpatient and follow‐up studies were excluded. Sensitivity, specificity, positive predictive value and negative predictive value of the proposed clinical decision rule within the studied population were calculated. Results Data were obtained for 584 patient encounters where suspected PE was investigated radiologically. Applied retrospectively, the proposed clinical decision rule had a negative predictive value of 97.7% and a sensitivity of 98.5% for radiologically proven PE; 9.2% of scans could have been avoided. One case of PE would have been missed; a false‐negative rate of 1.5%. Conclusion Retrospective application of the proposed clinical decision rule to the studied cohort indicates at least 9% of radiological investigations were unnecessary. A prospective study is needed to assess the safety and cost‐effectiveness of applying such a pathway to all patients presenting to ED with suspected PE.