Background: Most patients with gastro‐oesophageal reflux disease (GERD), regardless of endoscopic status, suffer symptomatic relapse within 6 months of stopping acid suppressant therapy. Aim: To assess the efficacy of ‘on‐demand’ treatment of GERD with esomeprazole, the first proton pump inhibitor developed as an optical isomer. Methods: In this multicentre, double‐blind study, 342 endoscopy‐negative GERD patients demonstrating complete resolution of heartburn during the final week of a 4‐week treatment period with esomeprazole 20 mg or omeprazole 20 mg once daily were randomized to receive esomeprazole 20 mg or placebo on demand (maximum of one dose per day) for a further 6 months. Use of rescue antacids was permitted. Results: All 342 patients (191 males), aged 19–79 (mean 49) years, were evaluable in the intention‐to‐treat analysis. The proportion of patients who discontinued treatment due to insufficient control of heartburn was significantly higher among placebo compared to esomeprazole recipients (51% vs. 14%; P < 0.0001). Patients randomized to esomeprazole on‐demand therapy remained in the study longer than those in the placebo group (mean 165 vs. 119 days). Over 50% took the study medication for periods of 1–3 consecutive days (esomeprazole) or 4–13 consecutive days (placebo). Use of antacids was > 2‐fold higher among placebo recipients. The frequency of adverse events was similar in the two groups, when adjusted for time spent in the study, as were the clinical laboratory profiles. Conclusions: On‐demand therapy with esomeprazole 20 mg is effective and well tolerated in maintaining symptom control in endoscopy‐negative GERD.

The disease course and prognosis of UC appears better than previously described in the literature. The frequency of surgery was low, and only a minority of the patients had symptoms that interfered with their everyday activities 5 years after diagnosis.

In this study one of the highest incidences of UC in the world has been found. The classification 'indeterminate colitis' seems reasonable to use in some of the cases to prevent misclassification at the initial stage of diagnosis.

The 5-year clinical course in an unselected cohort of CD patients was mostly mild. The frequency of surgery was lower than that observed in other studies and only a minority of the patients had symptoms that interfered with everyday activities 5 years after the initial diagnosis. The Vienna classification predicted the risk of surgery, but did not predict symptoms at 5 years, relapses during the observation period or the course of disease as described by the patients.

Unemployment or sick leave is more common in IBD patients than in the Norwegian background population. The number of patients receiving DP is significantly increased in women with CD but not in the other patient groups. Unemployment, sick leave, and DP are related to the patient's HRQOL in a negative way, but this effect is most pronounced in patients reporting IBD-related sick leave.