Most patients with gallstone disease have non-specific abdominal symptoms. The aim of this study was to determine the prevalence of gallstone disease in patients attending a gastroenterology outpatient clinic with abdominal pain, and to assess what symptoms are of greatest diagnostic value. Two hundred and eight-two consecutive outpatients presenting with abdominal pain were studied prospectively. A proforma recording 11 different symptoms was completed at the first visit, and abdominal ultrasound was subsequently carried out in all patients: gallstones were diagnosed in 41 out of 282 (14.5% prevalence); and biliary colic had the highest specificity (98%), but low sensitivity (22%). Among the other symptoms, only radiation of pain through to the back provided a high sensitivity (83%) and specificity (74%), giving a positive predictive value significantly greater than gallstone prevalence (36% compared with 14.5%, P < 0.001). The diagnostic value of pain radiating to the back for peptic ulcer disease was low (sensitivity 25%, specificity 69% and positive predictive value 18%). The positive predictive value for fatty dyspepsia (16%) was not significantly different from gallstone prevalence. Abdominal ultrasonography to detect gallstones is therefore recommended in all outpatients with abdominal pain radiating through to the back.