Introduction Proton pump inhibitors (PPIs) are one of the most frequently prescribed classes of drug in the world, and their use has doubled in the last 8 years. Although highly effective in the management of peptic oesophagitis and ulcer disease, they are being increasingly employed for diverse indications, and studies have demonstrated that 25–70% of prescriptions are possibly inappropriate. Although generally regarded as safe, concerns have been raised regarding potential hazard, including possibly harmful interactions with cardiovascular medications. Furthermore, rebound acid hypersecretion following PPI use has been shown to induce the symptoms that these drugs are designed to treat. We aimed to review the characteristics of patients prescribed PPI in a primary care population.

Methods Data was obtained from all patients in one GP practice who had been started on a PPI in 2008 who had no prior PPI use in the previous year. Computerised records of prescription and patient encounters, and correspondence with secondary care were analysed. Patients receiving PPI as part of eradication therapy for H pylori were excluded.

Results 1292 patients (15% of practice population) were prescribed a PPI in 2008. 418 of these patients (4.8% of practice population) had not received a PPI in the previous year. 268/418 (64%) had never previously received PPI therapy. 80% of prescriptions were initiated in primary care and twenty different indications were recorded. The commonest documented indications were gastro-protection in NSAID use (20%), dyspepsia (18%) and reflux disease (16%). 44/418 (10.5%) patients were prescribed PPI for unlicensed indications and in 66/418 (16%) no indication was discernible. 48/81(59%) of patients prescribed a PPI with NSAID therapy had no risk factors for ulcer disease. 347/418 (83%) of patients received PPI on acute prescription and 4% of these patients remained on repeat prescription for PPI 2 years later.

Conclusion PPI prescribing in primary care is highly prevalent and most commonly initiated by GPs. A quarter of patients received PPIs for unlicensed or unknown indications, and this was commonest for prescriptions initiated in hospital. Gastro-oesophageal reflux disease accounted for only 16% of prescriptions. The single commonest indication for PPI was gastro-protection in NSAID therapy, but most patients in this category had no evidence of risk factors for NSAID enteropathy. Despite the high incidence of PPI prescribing, prevalence of continued use at 2 years in this cohort was low.