Changing incidence of peptic ulcer--facts or artefacts? A cohort study from Tromsø.

STUDY OBJECTIVE The aim was to estimate the age and sex specific incidences of duodenal and gastric ulcers censored for death and migration. DESIGN A population based cohort was followed seven years through records of x ray examinations, endoscopies, and operations in the only hospital serving the area. SETTING The study was conducted in the municipality of Tromsø, Northern Norway, where all men aged 20 to 54 years and women aged 20 to 49 years in 1980, a total of 21,440, were included. MAIN RESULTS We found an incidence of 1.47 (95% CI 1.21-1.76) and 0.88 (0.67-1.14) per 1000 person-years for gastric and 1.98 (1.69-2.31) and 0.85 (0.64-1.11) for duodenal ulcers in men and women, respectively. A small and insignificant sex difference for gastric ulcer was noted. The duodenal preponderance in men was only recognised among the ulcers identified by x ray, not among ulcers diagnosed by endoscopy. CONCLUSIONS This population based study has, in contrast to recent studies from other areas, revealed an unchanged high incidence of gastric and duodenal ulcer in both sexes. Compared to earlier studies from this area it indicates a preponderance of duodenal ulcers. The study also shows the impact of attendance rate, death, and migration, and of diagnostic methods on the incidence estimates. Ignoring these potential biases may lead to conclusions on peptic ulcer trends that reflect artefacts rather than real changes in peptic ulcer occurrence.