Background: The prognostic significance of chest pain is well established in patients with coronary artery disease, but still ill defined in primary prevention. Therefore, the aim of our analysis was to assess the prognostic value of different forms of chest pain in a large cohort of primary care subjects under the conditions of contemporary modalities of care in primary prevention, including measurement of serum levels of the biomarker NT-pro-BNP. Design: We carried out a post-hoc analysis of the prospective DETECT cohort study. Methods: In a total of 5570 unselected subjects, free of coronary artery disease, within the 55,518 participants of the cross-sectional DETECT study, we assessed chest pain history by a comprehensive questionnaire and measured serum NT-pro-BNP levels. Three types of chest pain, which were any chest pain, exertional chest pain and classical angina, were defined. Major adverse cardiovascular events (MACEs=cardiovascular death, myocardial infarction, coronary revascularization procedures) were assessed during a 5-year follow-up period. Results: During follow-up, 109 subjects experienced a MACE. All types of reported chest pain were associated with an approximately three-fold increased risk for the occurrence of incident MACEs, even after adjusting for cardiovascular risk factors. Any form of reported chest pain had a similar predictive value for MACEs as a one-time measurement of NT-pro-BNP. However, adding a single measurement of NT-pro-BNP and the information on chest pain resulted in reclassification of approximately 40% of subjects, when compared with risk prediction based on established cardiovascular risk factors. Conclusions: In primary prevention, self-reported chest pain and a single measurement of NT-pro-BNP substantially improve cardiovascular risk prediction and allow for risk reclassification of approximately 40% of the subjects compared with assessing classical cardiovascular risk factors alone.