Introduction Efforts have been focused on the prevention of overt upper gastrointestinal bleeding (UGIB), using acid inhibitors, particularly in users of NSAIDs and anti-thrombotic therapy (ATT). Little is known about the impact of such efforts on occult GIB which might also originate from sites not protected by acid inhibition. We aimedto measure the incidence of both overt and occult GIB in a well-defined geographical region over a 5-year period, 2007–2012; assess the use of NSAIDs, ATT, PPIs, and iron therapy; and compare the outcomes of patients with occult vs. overt GIB.

Method The incidence was analysed in 2007, 2010 and 2012. Overt UGIB included haematemesis and melena. A sample of 300 patients (100 from each of the 3 index years) with occult GIB was randomly selected and their outcomes and details were compared with those of patients with overt UGIB (N = 869) using the Mann-Whitney and Fisher’s exact tests. Trends with time in incidence were assessed using the chi-squared test for trend. Numbers of prescriptions issued per 1000 population were recorded for each year from 2007 to 2012 and trends were assessed using Pearson correlation analysis.

Results The incidence of overt UGIB and occult GIB and details of drug use are shown in Table 1. Demography: Compared with patients with occult GIB (N = 300) randomly selected in the 3 index years, those with overt UGIB (N = 869) had a median (IQR) age of 63 (47–77) years vs. 67 (53–75) in the occult group (NS), 62.5% males vs. 39.3% (P < 0.001), 33.3% smokers vs. 28.0% (NS), and 25.0% excessive alcohol drinkers vs. 11.0% (P < 0.001), respectively. Both groups were comparable in the use of NSAIDs and ATT, but overt UGIB patients had higher Charlson comorbidity scores, 2 (1–3) vs. 1 (0–2) (P < 0.001). Transfusion and 30-day mortality: 3.8% in the overt UGIB group died vs. 3.0% (NS) in the occult group, and 34.3% vs. 13.0% (P < 0.001) were transfused with 3 (2–6) blood units vs. 2.5 (2–4) (P = 0.024), respectively in the occult GIB group.

Conclusion (1) The incidence of occult GIB continues to rise while that of overt UGIB continues to fall. (2) This is accompanied by a rise in the use of NSAIDs, PPIs, and iron therapy. (3) Patients with occult GIB have similar 30-day mortality to overt UGIB but are less likely to require blood transfusion. (4) An alternative approach to acid inhibition might be needed to prevent occult GIB.

Disclosure of interest None Declared.