The aim of this 3-month follow-up prospective pragmatic study was to evaluate the implementation of a pulmonary embolism (PE) diagnostic strategy in clinical practice. One thousand and one hundred thirty-four consecutive in- and outpatients with clinically suspected PE were enrolled into a sequential diagnostic algorithm in which vascular medical unit plays a pivotal role in advising physicians and suggesting the most appropriate tests according to the diagnostic algorithm. In this observational study, patients that followed the proposed work-up were attributed to a so-called "conform group". Patients in whom diagnostic work-up was not according to protocol were attributed to a "non-conform group". Nine hundred and ninety-seven patients (87.9%) had a conform work-up, and 137 patients a non-conform work-up according to the proposed diagnostic algorithm. The non-conform work-up directly increased in relation to the age of the referred patients. PE was ruled out in 907 (80%) patients of whom 787 (86.8%) were in the conform group. Of the 797 patients who did not receive anticoagulant drugs, follow-up was obtained in 792 (99.4%). Among these patients, the incidence of acute thromboembolic events during the 3-month follow-up period was different in the group of patients that had a conform work-up (1%, [95% CI, 0.5-2.1%]) from the non-conform group patients (4.5%, [95% CI, 2-10.2%]. Therefore patients from the non-conform group have an independent increased risk to develop a thromboembolic event during the follow-up, adjusted odds ratio 3.3 [1.1-10, 95% CI]. Therefore we demonstrated that a non-conform diagnostic management strategy is associated with a higher risk of thrombotic event occurrence.