Aim: The objectives of this study were to know the incidence of acute nonvariceal upper gastrointestinal bleeding (UGIB) in a Mediterranean region, and to analyze variations due to sociodemographic (age, gender and Primary Care district) and time factors (trend and seasonality).
Material and methods: Longitudinal study during 4 years (April 1995 to March 1999) including all cases of nonvariceal UGIB admitted to the reference hospital of a Health authority area of the Valencian Community, divided into 17 Primary Care districts. Incidence was modeled by Poisson regression.
Results: Incidence for the whole period was 108 cases per 100,000 persons-year (95% CI: 100-115). Almost half of the cases (44.9%) were associated to the previous use of nonsteroidal anti-inflammatory drugs (NSAIDs). Incidence was significantly higher for males and older people. Variations of incidence by Primary Care Districts were only significant for NSAID-related UGIB. In the time series analysis, we found a negative trend and a clear seasonality with higher incidence in autumn and winter compared with spring and summer. When relating the seasonality with UGIB, a significant seasonal pattern was only found for UGIB caused by duodenal ulcer, and disappeared in subjects with previous use of NSAIDs.
Conclusions: In our region, nonvariceal UGIB has a significant incidence, but slightly below that reported by other studies carried out in our country. The highest incidence of UGIB was observed in males, and increased sharply with age. In the time series analysis, a negative trend and a clear seasonal pattern that is absent in cases of previous use of NSAIDs, were found.