Medical evidence and health policy: a marriage of convenience? The case of proton pump inhibitors

Journal of Evaluation in Clinical Practice ISSN 1356-1294 Medical evidence and health policy: a marriage of convenience? The case of proton pump inhibitors Mieke L. van Driel MD MSc,1 Robert Vander Stichele MD PhD,2 Jan De Maeseneer MD PhD,3 An De Sutter MD PhD4 and Thierry Christiaens MD PhD5 1 Researcher, 3Professor of Primary Heath Care, Head of Department, Department of General Practice and Primary Health Care, Ghent University, Ghent, Belgium 2 Professor of Pharmacology, Heymans Institute of Pharmacology, Ghent University, Ghent, Belgium 4 Lecturer in General Practice, 5Professor of General Practice, Department of General Practice and Primary Health Care, Ghent University, Ghent, Belgium; Heymans Institute of Pharmacology, Ghent University, Ghent, Belgium Keywords evidence-based medicine, health policy, pharmaceutical economics, doctor’s practice patterns, proton pump inhibitors, quality of care Correspondence Mieke L. van Driel Department of General Practice and Primary Health Care Ghent University, De Pintelaan 185, UZ-1K3 B-9000 Ghent, Belgium E-mail: mieke.vandriel@ugent.be Accepted for publication: 28 November 2006 Abstract Rationale In Belgium, several policies regulating reimbursement of acid suppressant drugs and evidence-based recommendations for clinical practice were issued in a short period of time, creating a unique opportunity to observe their effect on prescribing. Aims and objectives To describe the evolution of prescriptions for acid suppressants and explore the interaction of policies and practice recommendations with prescribing patterns. Method Monthly claims-based data for proton pump inhibitors (PPIs) and H2-antihistamines by general practitioners, internists and gastroenterologists were obtained from the Belgian national health insurance database (1997–2005). The evolution of reimbursed defined daily doses and expenses after introduction of reimbursement regulations and dissemination of practice recommendations was explored. Results Recommendations had no impact on prescribing. All changes can be related to concomitant policies. Lifting reimbursement restrictions for cheaper products did not control growth or save costs in the long term. Only restricting reimbursement of all PPIs managed to curb the growth. We observed an unintended increase of non-omeprazole PPIs by gastroenterologists. Conclusions Reimbursement policies influence prescribing, but their effect can be unintended. A dialogue between policymakers and guideline developers, and evidence-based policies that are linked to clinical guidelines, could be an effective way to pursue both costcontainment and quality of care. Introduction Improving quality of care through integrating the best research evidence into clinical practice is the cornerstone of evidence-based health care [1]. The development of clinical guidelines, protocols and algorithms has provided clinicians with useful tools for implementing evidence-based recommendations aimed at improving quality of care. However, implementation in clinical practice has not been very successful. There is evidence that merely distributing guidelines has very little or no impact on how doctors prescribe [2]. Restrictions imposed by reimbursement regulations could be a more powerful instrument to change prescribing habits. But these regulations are often inspired by cost-containment agendas and are usually not directly linked to scientific evidence from clinical practice guidelines. 674 Internationally, the volume of prescriptions for proton pump inhibitors (PPIs) is increasing. This adds to the growth of public spending on medication and raises concerns for policymakers about cost-containment. Several examples of successful policy strategies regulating reimbursement in order to counter the growth of PPI spending have been published [3–6]. In Belgium, in a short period of time the national government issued various regulations in an attempt to control growing expenses for acid suppressants. Also, recommendations for rational use were distributed. This created an opportunity to closely study their impact on prescribing. We evaluated national prescribing data for acid suppressants over a period of 9 years (from 1997 to 2005) and explored the impact of several potential drivers of prescribing, including availability of practice recommendations, introduction of new products on the market, and national reimbursement pol- © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd, Journal of Evaluation in Clinical Practice 13 (2007) 674–680 Reimbursement policies and EBM M.L. van Driel et al. icy regulations. By using a multifaceted approach, we may identify clues to integrate these drivers in comprehensive strategies to improve rational prescribing, quality of care and containment of health care spending. Methods The Farmanet database Our analyses are based on the nationwide claims-based prescription database (Farmanet) of the Belgian National Institute for Sickness and Invalidity Insurance (RIZIV-INAMI). Farmanet covers all medication that is reimbursed by the RIZIV-INAMI for approximately 9.3 million people (90% of the Belgian population in 2004). Over-the-counter products are not included in this database. We extracted monthly reimbursement data for PPIs (ATC classes A02BC01/02/03/04/05) and H2-antihistamines (ATC classes A02BA01/02/03/04/07), and analysed these data for the three main groups of prescribers (i.e. general practitioners, general internal medicine specialists and gastroenterologists). Interventions We made an inventory of policy regulations concerning the availability and reimbursement strategies for PPIs and H2antihistamines by consulting the official publications of the Belgian government, the Belgian State Monitor. We studied the evidence report of the national consensus conference on rational use of acid suppressant drugs [7]. Data analysis We generated graphs with SPSS 12.0 software for the total reimbursed volume of PPIs and H2-antihistamines in defined daily doses (DDD) and expenses. We also analysed omeprazole and non-omeprazole PPIs separately, as policy regulations were different for each of these medication groups. Prescribing behaviour of general practitioners and general internal medicine specialists was compared with gastroenterologists. Results Policy regulations All relevant policy regulations that were issued in the studied period (1997–2005) are listed in Table 1. Until 2001 PPIs and H2antihistamines were only reimbursed after authorization by a consultative doctor of the insurance agencies. Authorization was based on the report of an endoscopic examination. In March 2001, two H2-antihistamines (cimetidine and ranitidine) became available without authorization (‘open benefit’). Two years later (April 2003), omeprazole (following its patent expiry) and rabeprazole, as the only non-omeprazole PPI, were granted the same type of open-benefit access (i.e. without authorization or endoscopy). For reimbursement of all other PPIs, authorization by a consultative doctor (including endoscopy) was still required. Initially, only small packages with low doses (20 mg) were reimbursed, but 1 year later (April 2004) packages with higher doses (40 mg) were included. This liberal access came to a halt in July 2005, when, for all PPIs, including omeprazole, authorization requiring an endoscopy report was reintroduced. The regulations, however, were slightly more complex now and differentiated between different groups of PPIs, and between generics and brand-name products. Authorization for reimbursement of generic forms of omeprazole and the brand-name product of rabeprazole could be obtained by making a report with the patient’s history and the results of a clinical examination (registered in the patient record) available upon request from the insurance agencies. For all other PPIs, reimbursement would only be granted with a priori authorization by a consultative doctor, based on a recent endoscopy report. An exception to this rule was made for PPIs co-prescribed with a nonsteroidal anti-inflammatory drug (NSAID), in which case endoscopy was not required. Evidence-based recommendations Concerned about the rising consumption and costs of acid suppressant drugs, the Belgian National Institute for Sickness and Invalidity Insurance organized a consensus conference on rational Table 1 Summary of policy regulations and publication of the evidence report on rational use of acid suppressant drugs (H2-antihistamines and PPIs) in Belgium (1997–2005) Date Policy regulation March 2001 April 2002 April 2003 Cimetidine and ranitidine ‘open benefit’ Omeprazole patent expires Omeprazole 20 mg (generic) small package ‘open benefit’ Pariet® (rabeprazole) ‘open benefit’ Omeprazole 40 mg small package ‘open benefit’ April 2004 September 2004 March 2005 July 2005 Evidence report Publication of Consensus Report ‘Acid suppressant drugs’ Omeprazole new large packages de novo in ‘open benefit’ Cimetidine (generic), ranitidine (generic) and Zantac®: increased out-of-pocket payment by the patient All other H2-antihistamines ‘restricted’ Omeprazole ‘restricted’ and increased out-of-pocket payment for low doses/small packages All other PPIs ‘restricted’ PPI, proton pump inhibitor. © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd 675 Reimbursement policies and EBM M.L. van Driel et al. use of acid suppressant drugs in May 2003. The resulting evidence report was based on a review of the available scientific literature and consensus of a multidisciplinary jury. It contains recommendations for clinical practice and was distributed to all doctors and pharmacists in Belgium in September 2004 [7] (see Box 1). Prescribing of PPIs and H 2-antihistamines Figure 1 shows that the total reimbursed volume of PPIs and H2antihistamines has been increasing since 1997, from 4.5 million DDD (500 DDD per 1000 inhabitants) in 1997 to 16.7 million DDD (1800 DDD per 1000 inhabitants) in 2005. Although the price per DDD has decreased over time by 56% (from €1.68 per DDD in 1997 to €0.74 in 2005), the public expenses for anti-ulcer drugs grew from €74.4 million in 1997 to €123.8 million in 2005. The patient’s out-of-pocket payments per DDD are 26% lower in 2005 compared with in 1997 (€0.22 versus €0.30), but owing to the growing dispensed volume, the total of patients’ payments has increased. Two distinct changes in total prescribed volume can be distinguished: a sudden rise in March 2001 and a sharp decrease in July 2005 (Fig. 1). Both changes can be linked to reimbursement policies. In March 2001, cimetidine and ranitidine were switched from ‘restricted’ to ‘open benefit’ (without authorization, i.e. endoscopy was no longer required). The drop in total volume of acid suppressants reimbursed in July 2005 can be linked to the introduction of a policy restricting the use of omeprazole again. The publication and distribution of evidence-based recommendations concerning rational use of acid suppressants in September 2004 did not have a distinguishable impact on the prescribed volume. Box 1 Recommendations of the consensus report on rational use of acid suppressant drugs, September 2004 [7] Patients with non-ulcer dyspepsia without alarm symptoms can be treated empirically using a step up approach starting with nonpharmacological treatment, followed by H2-antihistamines and then proton pump inhibitor (PPIs) for a limited period of time (4 weeks). Patients with gastro-oesophageal reflux disease (GORD) can be treated with either a step-up or a step-down strategy. Patients treated with NSAIDs and at high risk of gastrointestinal complications require association with a PPI in standard dose or double dose of H2-antihistamines. In the above-mentioned indications, PPIs and H2-antihistamines are both effective. PPIs are slightly more effective than H2-antihistamines (number needed to treat 1–6), but PPIs are not first choice. There is no evidence for superiority of one PPI over another. public expenses 20 000 000 out of pocket DDD DDD/Euro 15 000 000 10 000 000 5 000 000 0 200511 200509 200507 200505 200503 200501 200411 200409 200407 200405 200403 200401 200311 200309 200307 200305 200303 200301 200211 200209 200207 200205 200203 200201 200111 200109 200107 200105 200103 200101 200011 200009 200007 200005 200003 200001 199911 199909 199907 199905 199903 199901 199811 199809 199807 199805 199803 199801 199711 199709 199707 199705 199703 199701 Year/Month Figure 1 Consumption of acid suppressants in Belgium from 1997 to 2005. Reimbursed volume in defined daily doses (DDD), public and patients’ out-of-pocket expenses in Euro. 676 © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd Reimbursement policies and EBM M.L. van Driel et al. Figure 2 shows the evolution of the reimbursed DDD for omeprazole, non-omeprazole PPIs and H2-antihistamines separately. Omeprazole accounts for the majority of prescriptions for acid suppressants, peaking at approximately 12 million DDD (1300 DDD per 1000 inhabitants per month) in the first semester of 2005. The data show that, following the lifting of restrictions for H2antihistamines in 2001, the growth of omeprazole consumption came to a halt. For the first time since 1997 the consumption of H2antihistamines exceeded that of PPIs. Until mid-2001 three PPIs were available on the Belgian market: omeprazole, lansoprazole and pantoprazole. Rabeprazole was introduced in 2001 and esomeprazole in 2002. Omeprazole accounted for the majority of prescribed DDD until mid-2002, when non-omeprazole PPI volumes surpassed the omeprazole volume. This trend is reversed in the first quarter of 2003, after a new policy lifted the restrictions on the use of omeprazole and rabeprazole. The result was a marked growth of omeprazole consumption. The sharp drop in total prescribed volume in July 2005 (Fig. 1) can be linked to the drop in consumption of omeprazole following the reintroduction of restrictions for omeprazole (Fig. 2). In Fig. 3, the prescribing pattern of gastroenterologists illustrates the shift from omeprazole to non-omeprazoles that occurred in the period from 2001 to 2003. In this group of prescribers, since the first quarter of 2002 and continuing throughout the study period, the volume of prescribed non-omeprazole PPIs exceeds the volume of omeprazole. The 2003 policy, allowing unrestricted reimbursement of omeprazole and rabeprazole, did not reverse this trend. The volume of H2-antihistamines prescribed by gastroenterologists remained very low and stable throughout the study period. Discussion This descriptive analysis of national claims-based prescribing data for PPIs and H2-antihistamines in Belgium over a period of 9 years shows that policy regulations are an important driver of prescribing, but the effects of these policies can be unintended. Publication of an evidence-based report with recommendations for rational use of acid suppressants did not have a clear impact on prescribing. These findings may have important implications, particularly for policymakers and evidence-based guideline developers. Throughout the study period, several different reimbursement policies concerning acid suppressants have been implemented, but they did not always result in the desired or intended effect on consumption and costs. The first policy in March 2001 made two H2-antihistamines available without authorization. By facilitating the use of cheaper molecules, attention could be drawn away from the more expensive PPIs and lead to cost saving. Indeed, the consumption of H2-antihistamines increased, but not at the expense of PPI consumption. The net effect of this policy was a rise of the total consumption of acid suppressants without marked cost savings. The patent expiry of omeprazole in 2002 created opportunities for price reductions and cost-saving policies. By making the cheapest PPI (omeprazole) available without restrictions, the growth of the more expensive non-omeprazole PPIs could be stopped. However, the result was not as expected. In anticipation of this policy, which had been discussed in the media before it came into effect, consumption of non-omeprazole PPIs, predominantly prescribed by gastroenterologists, had increased. The consumption of non-omeprazole PPIs even exceeded the volume of 12 000 000 omeprazole Cimetidine ranitidine open benefit Omeprazole restricted non-omeprazole PPI H2-antihistamines 10 000 000 Arrows indicate introduction of new Omeprazole rabeprazole open benefit Sum of DDD 8 000 000 non-omeprazole PPIs 6 000 000 4 000 000 2 000 000 Evidence report 0 200511 200509 200507 200505 200503 200501 200411 200409 200407 200405 200403 200401 200311 200309 200307 200305 200303 200301 200211 200209 200207 200205 200203 200201 200111 200109 200107 200105 200103 200101 200011 200009 200007 200005 200003 200001 199911 199909 199907 199905 199903 199901 199811 199809 199807 199805 199803 199801 199711 199709 199707 199705 199703 199701 Year/Month Figure 2 Evolution of reimbursed volume (DDD) of omeprazole, non-omeprazole PPIs and H2-antihistamines in Belgium for general practitioners, internists and gastroenterologists from 1997 to 2005. DDD, defined daily doses; PPI, proton pump inhibitor. © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd 677 Reimbursement policies and EBM M.L. van Driel et al. 250 000 Omeprazole restricted omeprazole non-omeprazole PPI Cimetidine ranitidine open benefit H2-antihistamines 200 000 Sum of DDD Omeprazole rabeprazole open benefit 150 000 100 000 Evidence report 50 000 0 200511 200509 200507 200505 200503 200501 200411 200409 200407 200405 200403 200401 200311 200309 200307 200305 200303 200301 200211 200209 200207 200205 200203 200201 200111 200109 200107 200105 200103 200101 200011 200009 200007 200005 200003 200001 199911 199909 199907 199905 199903 199901 199811 199809 199807 199805 199803 199801 199711 199709 199707 199705 199703 199701 Year/Month Figure 3 Evolution of reimbursed volume (DDD) of omeprazole, non-omeprazole PPIs and H2-antihistamines for gastroenterologists in Belgium from 1997 to 2005. DDD, defined daily doses; PPI, proton pump inhibitor. omeprazole. Once the policy was introduced in March 2003, the consumption of non-omeprazole PPIs stabilized and the volume of prescribed omeprazole increased more rapidly than before. Owing to important price reductions for all acid suppressants, public expenses ceased to increase, but still remained high. Failure of these policies to lead to meaningful cost reductions inspired the last and most restrictive policy that was introduced in July 2005. By reversing the unrestricted reimbursement of omeprazole and reintroducing the need for authorization (hence an endoscopy), both total reimbursed volume and public expenses dropped. Several factors, linked to the structure of the health care system and the process of care, can explain the lack of success of the policies [8,9]. The number of available packages of acid suppressants (per product, dosage and formulation), registered as National Drug Codes (NDCs), increased from 24 in January 1997 to 147 in November 2005: 153 new NDCs were introduced for reimbursement and only 30 were withdrawn. Controlling this rapidly expanding market is a difficult task. Others have shown that, when the number of products on the market rises, consumption increases [10]. Mere availability of PPIs may have boosted consumption. Successful marketing by the pharmaceutical industry of new molecules, mainly the non-omeprazole PPIs, will have contributed to their rising popularity [11,12]. During a period when reimbursement regulations for omeprazole and non-omeprazole PPIs were the same, the prescribed volume of the more expensive nonomeprazole PPIs managed to exceed the volume of omeprazole. Patent expiry of omeprazole may have facilitated this shift. A 678 Dutch study also demonstrated that, after its patent expired in April 2002, more patients switched from omeprazole to another PPI [13]. There is no scientific evidence to underpin this switch. Another factor related to the pay-for-performance structure of the Belgian health care system could also have facilitated this observed preference for non-omeprazole PPIs. Before March 2003, PPIs were only reimbursed if an endoscopy report was available. Thus, patients were obliged to consult a gastroenterologist. In Belgium, doctors are remunerated on a fee-for-service basis. Therefore, making omeprazole available without the need of an endoscopic examination had a direct and tangible impact on the income of gastroenterologists. After March 2003, authorization was only required for non-omeprazole PPIs. Although the volume prescribed by gastroenterologists is small compared with the total volume of acid suppressants, they have an important influence on other prescribers. They are opinion leaders and role models, and their choice of medication is often continued in primary care [14,15]. Finally, the growing popularity of PPIs over the past decade cannot be explained solely by factors related to reimbursement. An increase of the incidence of specific pathologies or indications for acid suppressant treatment could justify the observed nearly threefold increase. A limitation of our study was that the prescription data are not linked to clinical diagnoses. Therefore, we collected data from other sources in order to understand the observed phenomena. A Belgian survey estimated that, in 1991, the prevalence of patients using reimbursable anti-ulcer medication was 30.5 per 1000 inhabitants per year, and that 80% of these patients were © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd Reimbursement policies and EBM M.L. van Driel et al. treated for a gastric or duodenual ulcer [16]. This epidemiological pattern has changed over the past decade. In the Netherlands, Post et al. observed a 50% decline in the incidence of gastric ulcer between 1992 and 2003, that they attribute to the reduced incidence of H. pylori infection and increased use of PPIs [17]. The latter may account for part of the growth in prescribed volume, but more frequent indications for PPI use in clinical practice nowadays are gastro-oesophageal reflux disease (GORD) or non-ulcer dyspepsia [18,19]. The prevalence of heartburn or regurgitation in western countries is estimated at 10% to 20% [20,21] and, for dyspepsia reports, range from 7% to 63% [22]. As for ulcer-related diseases in the past, nowadays psychological factors are linked to morbidity in dyspepsia and GORD and, consequently, to the need for PPIs [22]. In addition, co-prescription of PPIs with NSAIDs may account for some of the growth [23–25]. PPIs seem to have found a new role as a ‘comfort’ or ‘lifestyle’ drug in a stressful society [26,27]. Guidelines and recommendations underpinned by the best available scientific evidence are a cornerstone of evidence-based practice, but implementing them in clinical practice has been disappointing [2]. Interventions directly aimed at doctors [25–31] or patients [32] can change prescribing behaviour, but publication of practice recommendations without further reinforcement has little effect [2]. Our data describe that policy regulations, and not the publication of practice recommendations for acid suppressants, influenced prescribing. It is possibly also because the practice guidelines were published after the reimbursement regulations were issued. These findings make a case for exploring the role of policies in supporting implementation of evidence-based recommendations and the role of scientific evidence in designing policy strategies. More studies with other classes of drugs and diagnostic groups are needed to study these interactions between medical evidence and policy regulations and their potential for success. Could introducing the policy of 2003 (lifting reimbursement restrictions for omeprazole) together with a clinical guideline have been more successful in controlling consumption and costs? Could integrating reimbursement policies with clinical guidelines support implementation of rational prescribing? Reimbursement policies are a strong driver of prescribing, but their effect can be unintended and undesired. Piloting regulations on a small experimental scale or in specific target groups could help detect unintended effects and be useful in designing adequate and efficient policies [33]. Likewise, publishing recommendations for clinical practice without taking into account the policy context and harmonizing them with incentives, will not lead to rational prescribing and better quality of care. Conclusions Although quality of care should rely on scientific evidence underpinning practice guidelines and recommendations, in our study, prescribing follows policy regulations rather than clinical practice recommendations, with several unintended effects. The policies in our survey do not discriminate between rational and irrational use of medication, and therefore cannot be regarded as a sufficient instrument to improve quality of care. Ideally, policy regulations should be designed according to an evidence-based methodology and be able to guide, stimulate and facilitate the implementation of guidelines. Stakeholders should make use of the available evi- © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd dence to justify their positions. Integrating medical and policy evidence may be a more effective and efficient way to achieve better quality of care and public health. Acknowledgement The prescribing data were provided by the Belgian National Institute for Sickness and Invalidity Insurance (RIZIV-INAMI). No funding was received for this study. References 1. Evidence-Based Medicine Working Group (1992) Evidence-based medicine. A new approach to teaching the practice of medicine. Journal of the American Medical Association, 268, 2420–2425. 2. Grimshaw, J. M., Thomas, R. E., MacLennan, G., et al. (2004) Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technology Assessment, 8 (6). Available at: http:// www.hta.nhsweb.nhs.uk/project/994.asp (last accessed 29 January 2007). 3. Schneeweis, S., Maclure, M., Dormuth, C. R., Glynn, R. J., Canning, C. & Avorn, J. (2006) A therapeutic substitution policy for proton pump inhibitors: clinical and economic consequences. Clinical Pharmacology and Therapeutics, 79, 379–388. 4. Huskamp, H. A., Deverka, P. A., Epstein, A. M., Epstein, R. S., McGuigan, K. A. & Frank, R. G. (2003) The effect of incentive-based formularies on prescription-drug utilization and spending. New England Journal of Medicine, 349, 2224–2232. 5. Bursey F., Crowley M., Janes C. & Turner C. J. (2000) Cost analysis of a provincial drug program to guide the treatment of upper gastrointestinal disorders. Canadian Medical Association Journal, 162, 817–823. 6. Marshall, J. K., Grootendorst, P. V., O’Brien, B. J., Dolovich, L. R., Holbrook, A. M. & Levy, A. R. (2002) Impact of reference-based pricing for histamine-2 receptor antagonists and restricted access for proton pump inhibitors in British Columbia. Canadian Medical Association Journal, 166, 1655–1662. 7. RIZIV/INAMI (2004) The Rational Use of Acid Suppressant Drugs in Gastro-oesophageal Reflux Disease and Dyspepsia [Dutch/French]. Report of the Consensus Conference 15 May 2003. Brussels: RIZIV/ INAMI. Available at: http://riziv.fgov.be/drug/nl/statistics-scientificinformation/consensus/index.htm (last accessed 29 January 2007). 8. van Driel, M. L., De Sutter, A. I., Christiaens, T. C. M. & De Maeseneer, J. M. (2005) Quality of care: the need for medical, contextual and policy evidence in primary care. Journal of Evaluation in Clinical Practice, 11, 417–429. 9. van Weel, C. & Rosser, W. W. (2004) Improving health care globally: a critical review of the necessity of family medicine research and recommendations to build research capacity. Annals of Family Medicine, 2 (Suppl. 2), S5–S16. 10. Monnet, D. L., Ferech, M., Frimodt-Moller, N. & Goossens, H. (2005) The more antibacterial trade names, the more consumption of antibacterials: a European study. Clinical Infectious Diseases, 41, 114–117. 11. Adair, R. F. & Holmgren, L. R. (2005) Do drug samples influence resident prescribing behavior? A randomized trial. American Journal of Medicine, 118, 881–884. 12. Jones, M. P. (2004) Education, research, and industry. Patients first? Journal of Clinical Gastroenterology, 38, 471–474. 13. Klok, R. M., Boersma, C., Oosterhuis, I., Visser, S. T., de Jongvan den Berg, L. T. W. & Postma, M. J. (2006) Switch patterns before and after patent expiry of omeprazole: a case study in The Netherlands. Alimentary Pharmacology and Therapeutics, 23, 1595–1600. 679 Reimbursement policies and EBM 14. Jones, M. I., Greenfield, S. M., Jowett, S., Bradley, C. P. & Seal, R. (2001) Proton pump inhibitors: a study of GPs’ prescribing. Family Practice, 18, 333–338. 15. Krag, A., Teglbjerg, L. S., Malchow-Møller, A., Hallas, J. & Bytzer, P. (2006) Prescribing of acid suppressive therapy: interactions between hospital and primary care. Alimentary Pharmacology and Therapeutics, 23, 1713–1718. 16. Vander Stichele, R. H. & Petri, H. (1995) Utilization patterns of subsidized and nonsubsidized reimbursable peptic ulcer medication in Belgium. Pharmacoepidemiology and Drug Safety, 4, 213–224. 17. Post, P. N., Kuipers, E. J. & Meijer, G. A. (2006) Declining incidence of peptic ulcer but not of its complications: a nation-wide study in The Netherlands. Alimentary Pharmacology and Therapeutics, 23, 1587– 1593. 18. El Serag, H. B. & Sonnenberg, A. (1998) Opposing time trends of peptic ulcer and reflux disease. Gut, 43, 327–333. 19. Malfertheiner, P., Fass, R., Quigley, E. M. M., et al. (2006) Review article: from gastrin to gastro-oesophageal reflux disease – a century of acid suppression. Alimentary Pharmacology and Therapeutics, 23, 683–690. 20. Dent, J., El-Serag, H. B., Wallander, M. A. & Johansson, S. (2005) Epidemiology of gastro-oesophageal reflux disease: a systematic review. Gut, 54, 710–717. 21. Nocon, M., Keil, T. & Willich, S. N. (2006) Prevalence and sociodemographics of reflux symptoms in Germany – results from a national survey. Alimentary Pharmacology and Therapeutics, 23, 1601–1605. 22. Agréus, L. (2002) Natural history of dyspepsia. Gut, 50 (Suppl. IV): iv2–iv9. 23. Moynihan, R., Heath, I. & Henry, D. (2002) Selling sickness: the pharmaceutical industry and disease mongering. British Medical Journal, 324, 886–891. 24. Smith, R. (2002) In search of ‘non-disease’. British Medical Journal, 324, 883–885. 680 M.L. van Driel et al. 25. Teeling, M., Bennett, K. & Freely, J. (2003) Have COX-2 inhibitors influenced the co-prescription of anti-ulcer drugs with NSAIDs? British Journal of Clinical Pharmacology, 57, 337–343. 26. Murthy, S. K., Kaldher, S. & Targownik, L. E. (2006) Physicians’ approaches to the use of gastroprotective strategies in low-risk nonsteroidal anti-inflammatory drug users. Alimentary Pharmacology and Therapeutics, 23, 1365–1372. 27. RIZIV/INAMI (2003) Pharmaceutical indicators [Dutch/French]. Available at: http://riziv.fgov.be/drug/nl/statistics-scientificinformation/pharmanet/pharmaceutical-tables/pdf/2003/tables2003.pdf (last accessed 29 January 2007). 28. Valori, R. M., Brown, C. M., Strangeways, P. & Bradburn, M. (2001) Reducing community dyspepsia drug costs: a controlled trial. Gut, 49, 495–501. 29. Freemantle, N., Johnson, R., Dennis, J., Kennedy, A. & Marchment, M. (2000) Sleeping with the enemy? A randomized controlled trial of a collaborative health authority/industry intervention to influence prescribing practice. British Journal of Clinical Pharmacology, 49, 174– 179. 30. Lucas, L. M., Gerrity, M. S. & Anderson, T. (2001) A practice-based approach for converting from proton pump inhibitors to less costly therapy. Effective Clinical Practice, 4, 263–270. 31. Brufsky, J. W., Ross-Degnan, D., Calabrese, D., Gao, X. & Soumerai, S. (1998) Shifting physician prescribing to a preferred histamine-2-receptor antagonist: effects of a multifactorial intervention in a mixed-model health maintenance organization. Medical Care, 36, 321–332. 32. Krol, N., Wensing, M., Haaijer-Ruskamp, H., et al. (2004) Patientdirected strategy to reduce prescribing for patients with dyspepsia in general practice: a randomized trial. Alimentary Pharmacology and Therapeutics, 19, 917–922. 33. Jowell, R. (2003) Trying It Out. The Role of ‘Pilots’ in PolicyMaking. London: Cabinet Office. © 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd