PURPOSE: To explore the effect on prescribing of policies that lift reimbursement restrictions on selected H2-antihistamines and proton pump inhibitors (PPI) and of practice recommendations. METHODS: Monthly claims based data for proton pump inhibitors (PPIs) and H2-antihistamines were obtained from the national health insurance database (Farmanet 1997-2005). Two policies were issued. In March 2001 two H2-antihistamines and in March 2003 two PPIs became available without restrictions. An evidence report was distributed in September 2004. Periods before and after implementation of the interventions were compared. Interrupted time series with segmented regression analysis was used to assess and compare time trends. RESULTS: The first policy resulted in an increase of volume (1.6 million DDDs; 95% CI 1.4-1.8 million; p < .001) and expenditure (euro 637,744; 95% CI 177 052-1 098 437; p = .026) for H2-antihistamines, but consumption of PPIs continued to grow. After the second policy use of selected PPIs also increased (4.7 million DDDs increase; 95% CI 3.9-5.5 million; p < .001), but more than the desired shift toward selected PPIs. Although total expenditure stabilized at a lower level, there was no significant change of trend. Publication of the evidence report did not have any impact on prescribing. CONCLUSIONS: Policies that lift reimbursement restrictions did not result in meaningful changes in utilization or cost saving. They may even have unintended effects. Collaboration between policymakers and guideline developers and linking policies to evidence-based guidelines could be a more effective way to pursue cost-containment and better quality of care. Copyright (c) 2008 John Wiley & Sons, Ltd.
PURPOSE: To explore the effect on prescribing of policies that lift reimbursement restrictions on selected H2-antihistamines and proton pump inhibitors (PPI) and of practice recommendations. METHODS: Monthly claims based data for proton pump inhibitors (PPIs) and H2-antihistamines were obtained from the national health insurance database (Farmanet 1997-2005). Two policies were issued. In March 2001 two H2-antihistamines and in March 2003 two PPIs became available without restrictions. An evidence report was distributed in September 2004. Periods before and after implementation of the interventions were compared. Interrupted time series with segmented regression analysis was used to assess and compare time trends. RESULTS: The first policy resulted in an increase of volume (1.6 million DDDs; 95% CI 1.4-1.8 million; p < .001) and expenditure (euro 637,744; 95% CI 177 052-1 098 437; p = .026) for H2-antihistamines, but consumption of PPIs continued to grow. After the second policy use of selected PPIs also increased (4.7 million DDDs increase; 95% CI 3.9-5.5 million; p < .001), but more than the desired shift toward selected PPIs. Although total expenditure stabilized at a lower level, there was no significant change of trend. Publication of the evidence report did not have any impact on prescribing. CONCLUSIONS: Policies that lift reimbursement restrictions did not result in meaningful changes in utilization or cost saving. They may even have unintended effects. Collaboration between policymakers and guideline developers and linking policies to evidence-based guidelines could be a more effective way to pursue cost-containment and better quality of care. Copyright (c) 2008 John Wiley & Sons, Ltd.