Diagnosis of complex pulmonary disease (MAC-PD) requires positive culture of expectorated sputum or specimens acquired by bronchoscopy. Whether patients diagnosed using bronchoscopy have milder disease and milder progression than those diagnosed using sputum remains uncertain. To clarify whether disease severity and progression differ according to the diagnostic method. We retrospectively analysed 92 patients with MAC-PD. We compared characteristics of patients and disease progression according to the diagnostic methods used: sputum or bronchoscopy. Additionally, we investigated the impact of these methods on disease progression using multivariate analysis. Patients diagnosed using sputum were younger than those diagnosed using bronchoscopy; however, there were small differences from the viewpoint of clinical practice in disease severity, and estimated progression-free survival rate did not differ significantly. The predictors of disease progression were disease forms other than non-cavitary nodular/bronchiectatic disease, hypoalbuminemia and severe radiographic scores. The diagnostic methods had no significant impact on disease severity and disease progression of MAC-PD. If the diagnosis cannot be established by sputum culture or if sputum cannot be obtained in the patients with risk factors for disease progression, bronchoscopy would be useful to provide opportunity of treatment for MAC-PD.