OBJECTIVE: Radiography of the chest remains one of the most commonly performed examinations, although many of these procedures could be avoided if specific criteria were considered at the time of request. The aim of this study was to check if the quality of the clinical examination (physical examination) was associated with clinically significant abnormal findings in chest x-ray films. MATERIALS AND METHODS: In this transversal cohort study, 142 outpatients from an university hospital (Salvador, BA, Brazil), for whom a chest x-ray was requested, were included. RESULTS: Clinically significant abnormalities were seen in less than half (46.5%) of the patients submitted to chest x-rays. Patients aged 50 years or above showed a prevalence (P) 2.8 fold higher of presenting abnormalities on chest x-ray examinations. This figure was similar (P = 2.7) in patients coming from the outpatients of a cardiothoracic or medical clinic or in patients who were on therapy for their condition (P = 2.0). However, the variable utilized for assessment of the quality of medical care showed no association with the presence of chest x-ray abnormalities or with the purpose of the request and the type of complaint reported by the patient. CONCLUSION: The results indicate the need to conduct additional studies using a different methodological approach in order to determine the impact of clinical history and physical examination accuracy on request forms of chest x-ray examinations.

Keywords: Chest radiography, Clinical examination, Accuracy, Efficiency