Objectives: The objective of the study was to investigate the prevalence of undiagnosed dysglycaemia and the risk for type 2 diabetes using the Finnish Diabetes Risk Score (FINDRISC) in the working population of Belgium. Moreover, it was to evaluate performance and applicability of FINDRISC as a screening tool during occupational health surveillance. Material and Methods: A cross-sectional analysis was carried out over the years 2010–2011 among 275 healthy employees who underwent a health check including fasting plasma glucose and the FINDRISC questionnaire. The sensitivity, specificity and predictive value of different FINDRISC cut-off values to detect dysglycaemia was revised in the literature and then calculated. Results: The prevalence of unknown dysglycaemia was 1.8%. Twelve percent of the employees had a FINDRISC score of 12 to 14 corresponding to a moderate risk of 17% to develop diabetes within the next 10 years, and 5.5% had a score of 15 or more corresponding to a high – very high risk of 33% to 50%. All dysglycaemic individuals had a FINDRISC score of 12 or higher. The sensitivity and specificity for detecting dysglycaemia was respectively 100% and 84.1% for a FINDRISC cut-off value ≥ 12; and 80% and 95.9% for a cut-off value ≥ 15. Conclusions: A considerable number of workers had dysglycaemia or was at risk for developing type 2 diabetes. The questionnaire is a reliable, valuable and easy to use screening tool in occupational health surveillance.