The risk for upper gastrointestinal (UGI) bleeding in patients taking nonselective nonsteroidal anti-inflammatory drugs (NSAIDs) can be reduced by concomitant gastroprotective therapy. However, many patients do not comply with this approach.
To evaluate adherence to gastroprotective therapy and its effect on UGI events, investigators conducted a nested case-control study using three large national databases from the UK, the Netherlands, and Italy. Among a cohort of NSAID users aged 
50 years who took gastroprotective agents, patients with UGI events were identified. Adherence to gastroprotective therapy was determined using pharmacy data and compared between patients with and without UGI events.
Investigators documented 117,307 episodes of NSAID use with gastroprotective cotherapy. Patient adherence was rated as low (<20%) among 4.9% of the group and high (>80%) among 68.1%. A total of 339 UGI events occurred. The risk for such events was greater in those with low versus high adherence (odds ratio, 2.39 for all events; and OR, 1.89 for bleeding).
Comment: The authors conclude that nonadherence to gastroprotective therapy is associated with an increased risk for upper gastrointestinal events and bleeding and that strategies to improve adherence should be developed. Despite the limitations of retrospective database studies, this study's notable strengths were inclusion of patient databases from three European countries (each of which showed the same result) and use of a cohort taking both NSAIDs and gastroprotective agents, which eliminated the possibility of confounding through channeling of patients with higher risk of incident symptoms. However, the results compare only the patients with lowest (<20%) and highest (>80%) adherent, whereas many patients are occasionally adherent. We cannot conclude from this study if a threshold adherence rate lower than 80% exists that should be targeted clinically.
— David J. Bjorkman, MD, MSPH (HSA), SM (Epid.)
Published in Journal Watch Gastroenterology July 22, 2011
