Abstract
A population sample, selected at random after stratification for the presence of pulmonary disease, was examined for benign oesophageal disease by means of a mailed questionnaire, which has been described in a previous report. Eight hundred and nine subjects without pulmonary disease, 264 with chronic bronchitis, and 248 with bronchial asthma answered the questionnaire. Subjects answering affirmatively to a discriminating combination of questions, as well as some of the participants with single symptoms and randomly selected controls without oesophageal symptoms, were invited to a clinical examination (n = 346). One hundred and seventy-five subjects accepted an invasive investigation, 86 without pulmonary disease and 89 with chronic obstructive pulmonary disease (COPD). Endoscopy of the upper gastrointestinal tract was performed in 169 subjects, while 168 underwent pressure measurements of the oesophagus, and 113 had oesophageal 12-h pH measurements taken. On the basis of accepted definitions we found 114 subjects with benign oesophageal disease. The diagnostic sensitivity and specificity of the discriminating combination in the non-COPD and COPD groups were 73.9%/42.9% and 89.5%/47.1% respectively. The predictive accuracy of the questionnaire in the non-COPD and COPD group was 0.41 and 0.56, respectively. Prevalence rates of benign oesophageal disease in the non-COPD and the COPD groups were 34.5% (20–49%) and 44.5% (34–55%) respectively (P < 0.001). We estimated the prevalence rate in the general population to be about 30%.