The purpose of this study was to delineate the etiological factors for iatrogenic pneumothorax in the era of increased use of invasive procedures and to determine its impact on morbidity. Between 1983 and 1987 the Department of Veterans Affairs conducted a cooperative study at 13 medical centers to assess the utility of the intrapleural instillation of tetracycline for the prevention of recurrent pneumothorax. Since all patients with any type of pneumothorax were screened at each medical center, it was decided to collect data on all the iatrogenic pneumothoraces during that period. During the study period, the number of reported iatrogenic pneumothoraces were 538. Because of incomplete data, 3 iatrogenic pneumothoraces reported from one center were excluded. The leading causes of iatrogenic pneumothorax were transthoracic needle aspiration (128), subclavicular needle stick (119), thoracentesis (106), transbronchial biopsy (54), pleural biopsy (45) and positive pressure ventilation (38). Most patients required treatment for 4-7 days; however, hospitalization was prolonged due to this treatment in only 8% of patients, presumably because of their underlying disease which required long hospitalization. Patients with underlying chronic obstructive pulmonary disease required significantly longer duration of treatment than those without. We conclude that in our patient population, the three leading causes of iatrogenic pneumothorax are transthoracic needle aspiration, subclavicular needle stick and thoracentesis. Hospitalization is prolonged in only a small percentage of the patients who required treatment for the iatrogenic pneumothorax.