Appropriateness of proton pump inhibitor recommendations at hospital discharge and continuation in primary care

Background: Inappropriate prescriptions of proton pump inhibitors (PPI) in hospital and primary care have been widely reported. Recommendations from hospital have been implicated as one reason for inappropriate prescriptions of PPI in primary care. Objective: To quantify the amount of appropriate PPI recommendations in hospital discharge letters and the influence of these recommendations on general practitioners (GPs) PPI-prescriptions. Materials and Methods: This is an observational study in 31 primary care practices. We identified patients discharged from hospital with PPI recommendation between 2006 and 2007 and assessed practice records and PPI prescription six months prior and after hospital admission. Hospital recommendation for continuous PPI-treatment and continuation by GPs was classified as appropriate, inappropriate or uncertain. Logistic regression analysis was used to calculate factors associated with indicated and non-indicated PPI continuation. Results: In 263 (58%) out of 506 patients discharged from 35 hospitals with a PPI recommendation no indication could be found. Non-indicated PPIs were continued by GPs in 58% for at least 1 month. Indicated PPIs were discontinued in 33%. Two thirds of non-indicated PPIs were initiated in hospital. The strongest factor associated with non-indicated continuation was a PPI-prescription prior to hospital admission [OR: 3.0; 95% confidence interval (CI): 1.75.4]. This was also the strongest factor for continuation of an indicated PPI medication (OR: 3.2; 95% CI: 1.47.5). Conclusions: We found a strong influence of hospital recommendations and previous prescriptions on PPI prescriptions after discharge. Hospitals should critically review their practice of recommending PPI and document indications. GPs should carefully assess hospital recommendations and their medication prior to admission to avoid over- and under-prescribing.