Introduction Non-steroidal anti-inflammatory drugs (NSAIDs) and aspirin remain the most common cause of life-threatening iatrogenic pathology. Risks are age-dependent and can be reduced by switching to a selective cyclooxygenase (COX)-2 inhibitor or co-prescription of a proton pump inhibitor (PPI). The UK National Institute for Health and Clinical Excellence (NICE) has recommended that PPI co-prescription is sufficiently cost effective to be justified in all users of NSAIDs or COX-2 inhibitors. However, use of both PPIs and COX-2 inhibitors is widely discouraged by UK pharmacy advisers.
Methods We conducted an analysis of prescribing trends in patients aged 45 or over in UK general practices, recorded in the QResearch electronic database of general practice records. A patient was considered to be receiving a drug in any year if they had received >2 prescriptions in the 6 months prior to 1 April.
Results There were 459 practices which met the inclusion criteria. Overall there were 1 825 406 patients with at least 12 months of medical records and 1 594 369 were >45 years old. This resulted in 4 875 250 person-years of observations. Prescribing of aspirin rose from 9.9% in 2004 to 14.01% in 2007, while NSAID prescribing fell slightly from 5.57% to 5.49%. A PPI was co-prescribed with aspirin in only 11.7% of aspirin users in 2004 and 20.2% in 2007. For NSAIDs the values were 8.9% in 2003 and 23.1% in 2007. During this time prescription of PPIs in patients not on aspirin or NSAID rose from 4.4% to 7.8%, suggesting that increased use in aspirin and NSAID users was largely untargeted. Even among high-risk patients aged 80–84, protective strategies were limited, with 23.6% of male and 23.1% of female aspirin users receiving a PPI in 2007. In this age group, only 29.0% of men and 31.3% of women receiving an NSAID were co-prescribed a PPI. Of patients on aspirin, 17.7% in 2004 and 11.1% in 2007 were also prescribed and NSAID despite a potentially harmful pharmacodynamic interaction. Use of COX-2 inhibitors was limited (1.33% in 2003 and 0.33% in 2007), with co-prescription of a PPI in 25.0%.
Conclusion Most prescribing of anti-inflammatory drugs in the UK is on an unprotected background even in the highest risk patients and NICE advice to co-prescribe a PPI is not followed. Given evidence that the combination of a COX-2 inhibitor and a PPI may virtually eliminate the risk of ulcer complications associated with NSAIDs, our data suggest UK patients are exposed to unacceptable levels of avoidable risk. Whether deficiencies in prescribing standards are due to financial pressures of a need for education requires further evaluation.