Objectives: To determine the prevalence and predictors of undiagnosed chronic obstructive pulmonary disease (COPD) in Norway. Methods: An age and gender stratified random sample of all adults aged 47-48 and 71-73 years in Bergen, Norway, were invited. The 3506 participants filled in questionnaires including symptoms of COPD, smoking, socio-economic status, self-rated health and cardiac co-morbidity. Spirometry was performed before and after inhalation of 400 mu g of salbutamol. COPD was defined as post-bronchodilator forced expiratory volume in 1 s (FEV1) / forced vital capacity (FVC) < 0.7 whereas diagnosed COPD was defined as having received treatment for obstructive lung disease the last year. Results: Three hundred-three persons (9%) were classified as having COPD, and the undiagnosed fraction was 66%. In multiple logistic regression analysis, including multiple imputation, predictors of undiagnosed COPD were absence of COPD symptoms [odds ratio (OR) 6.92, P = 0.001], and self-report of being in good/excellent health (OR 2.39, P = 0.005). When post-bronchodilator FEV1 was added to the analysis, undiagnosed disease was predicted by pack years [OR 1.21 (1.01-1.47) per 10 pack-year increase, P = 0.043], and close to normal lung function [OR 1.48 (1.22-1.80) per 10% increase in post-bronchodilator FEV1 % predicted, P < 0.001]. Anthropometrical variables, socio-economic status and cardiac co-morbidity were not associated with having undiagnosed COPD. Conclusion: Two out of three COPD patients in Norway are undiagnosed. Risk factors for being undiagnosed are moderate reduction in lung function, absence of COPD symptoms and self-report of being in good health. Please cite this paper as: Hvidsten SC, Storesund L, Wentzel-Larsen T, Gulsvik A and Lehmann S. Prevalence and predictors of undiagnosed chronic obstructive pulmonary disease in a Norwegian adult general population. The Clinical Respiratory Journal 2009; DOI:10.1111/j.1752-699X.2009.00137.x.