Background: Barrett's oesophagus (BO) predisposes to oesophageal adenocarcinoma. Epidemiological data suggest that the incidence of BO is rising but it is unclear whether this reflects a true rise in incidence of BO or an increase in detection secondary to more upper gastrointestinal endoscopies performed. This study aimed to examine the changes in BO incidence relative to the number of upper gastrointestinal endoscopies performed in the general population. Methods: We conducted a cohort study using the Integrated Primary Care Information database. This general practice research database contains the complete and longitudinal electronic medical records of more than 500 000 persons. Results: In total, 260 incident cases of BO were identified during the study period. The incidence of BO increased from 14.3/100 000 person years in 1997 (95% confidence interval ( Conclusion:The incidence of diagnosed BO is increasing, independent of the number of upper gastrointestinal endoscopies that are being performed. This increase in BO incidence will likely result in a further increase in the incidence of oesophageal adenocarcinomas in the near future.

SUMMARY BackgroundProton pump inhibitors are widely used, but little is known about the usage pattern in different indications.

Background Gastro-protective agents (GPA) are co-prescribed with non-steroidal anti-inflammatory drugs (NSAID) to lower the risk of upper gastrointestinal (UGI) events. It is unknown to what extent the protective effect is influenced by therapy adherence. Aim To study the association between GPA adherence and UGI events among non-selective (ns) NSAID users. Methods The General Practice Research Database (UK 1998e2008), the Integrated Primary Care Information database (the Netherlands 1996e2007) and the Health Search/CSD Longitudinal Patient Database (Italy 2000e2007) were used. A nested case-control design was employed within a cohort of nsNSAID users aged $50 years, who also used a GPA. UGI event cases (UGI bleeding and/or symptomatic ulcer with/without obstruction/perforation) were matched to event-free members of the cohort for age, sex, database and calendar time. Adherence to GPA was calculated as the proportion of nsNSAID treatment days covered by a GPA prescription. Adjusted OR with 95% CI were calculated. Results The cohort consisted of 618 684 NSAID users, generating 1 107 266 nsNSAID episodes. Of these, 117 307 (10.6%) were (partly) covered by GPA, 4.9% of which with a GPA coverage <20% (non-adherence), and 68.1% with a GPA coverage >80% (full adherence). 339 patients experienced an event. Among non-adherers, the OR was 2.39 (95% CI 1.66 to 3.44) for all UGI events and 1.89 (95% CI 1.09 to 3.28) for UGI bleeding alone, compared to full adherers. Conclusions The risk of UGI events was significantly higher in nsNSAID users with GPA non-adherence. This underlines the importance of strategies to improve GPA adherence.Non-steroidal anti-inflammatory drugs (NSAID) have analgesic and anti-inflammatory properties, and are indicated mainly for pain management in musculoskeletal injury, osteoarthritis and rheumatoid arthritis. The use of NSAID may lead to upper gastrointestinal (UGI) symptoms such as dyspepsia, and to more severe events such as gastroduodenal ulcers or UGI bleeding. The incidence of such clinically significant UGI events during NSAID use has been estimated at 1e2.5/100 person-years 1 2 and is associated with substantial mortality.3 Non-selective (ns) NSAID inhibit the cyclooxygenase 1 enzyme more strongly than cyclooxygenase 2 selective inhibitors. 4 As cyclooxygenase 1 is involved in gastroprotection, nsNSAID are believed to increase the risk of UGI events to a higher degree than cyclooxygenase 2 inhibitors. 5To prevent UGI events during nsNSAID use, evidence-based guidelines recommend the concomitant use of gastroprotective agents (GPA), 6 mostly in nsNSAID users with one or more risk factors. The guidelines differ slightly in their definition of risk factors, but most consider advanced age, the history of a UGI event and the use of antiplatelet agents, anticoagulants or corticosteroids as risk factors.7e9 Some guidelines also mention other factors, including the use of selective serotonin Significance of this study What is already known about this subject?< The use of non-selective no...

SUMMARY BackgroundUpper gastrointestinal (UGI) complications are a well-recognized risk of NSAID treatment, requiring preventive measures in high-risk patients. Adherence to gastroprotective agents (GPAs) in NSAID users has been suggested to be suboptimal.

The present study indicates no association between PPI use and the risk of colorectal cancer. Larger numbers of long-term PPI users are needed to confirm the absence of a risk-increasing effect of long-term PPI exposure.