Eur J Clin Pharmacol (2007) 63:725–731 DOI 10.1007/s00228-007-0324-2

SPECIAL ARTICLE

Potentially inappropriate medications in the elderly: a French consensus panel list Marie-Laure Laroche & Jean-Pierre Charmes & Louis Merle

Received: 15 February 2007 / Accepted: 9 May 2007 / Published online: 7 June 2007 # Springer-Verlag 2007

Abstract Objective To evaluate drug-related problems in the elderly, various lists of potentially inappropriate medications have been published in North America. Unfortunately, these lists are hardly applicable in France. The purpose of this study was to establish a list of inappropriate medications for French elderly using the Delphi method. Method A two-round Delphi method was used to converge to an agreement between a pool of 15 experts from various parts of France and from different backgrounds (five geriatricians, five pharmacologists, two pharmacists, two general practitioners, one pharmacoepidemiologist). In round one, they were sent a questionnaire based on a literature review listing medications and clinical situations. They were asked to comment on the potential inappropriateness of the criteria proposed using a 5-point Likert scale (from strong agreement to strong disagreement) and to suggest therapeutic alternatives and new criteria. In round two, the experts confirmed or cancelled their previous answers from the synthesis of the responses of round one. After round two, a final list of potentially inappropriate drugs was established.

Results The final list proposed 36 criteria applicable to people≥75 years of age. Twenty-nine medications or medication classes applied to all patients, and five criteria involved medications that should be avoided in specific medical conditions. Twenty-five medications or medication classes were considered with an unfavourable benefit/risk ratio, one with a questionable efficacy and eight with both unfavourable benefit/risk ratio and questionable efficacy. Conclusion This expert consensus should provide prescribers with an epidemiological tool, a guideline and a list of alternative therapies.

Keywords Inappropriate medications . Elderly . Delphi method . France Abbreviations NSAID nonsteroidal antiinflammatory drug SNRI serotonin and noradrenaline reuptake inhibitor SSRI selective serotonin reuptake inhibitor IMs inappropriate medications

Introduction M.-L. Laroche : L. Merle (*) Department of Pharmacology-Toxicology, Centre of Pharmacovigilance, University Hospital Dupuytren, 87042 Limoges, France e-mail: [email protected] M.-L. Laroche e-mail: [email protected] J.-P. Charmes Department of Geriatrics, Hospital Rebeyrol, 87042 Limoges, France e-mail: [email protected]

Drug-related problems are a major public health issue, especially as many adverse drug reactions are considered preventable. Among these adverse effects, those linked to unsuitable medications are of particular interest. These drugs named IMs have an unfavourable benefit-to-risk ratio when safer or equally effective alternatives are available. Therefore, IMs appear as a risk factor for preventable drugrelated illnesses. Identifying these medications is of paramount importance when treating the elderly, who often receive many drugs.

726

A way to establish a list of IMs in the elderly is the use of an expert consensus to develop explicit criteria when clinical information is lacking. Explicit criteria have already been proposed in the United States and Canada. In 1991, Beers et al. developed the first list of criteria for determining IM use in nursing home residents [1]. In 1997, this criteria list was expanded so as to include drugs whose risk may outweigh their benefit in all patients older than 65 years, whatever their dwelling place [2]. A last updating of the Beers criteria was published by Fick et al. in 2003 [3]. In Canada, McLeod et al. proposed another IM list because of a disagreement with some medications identified by Beers [4]. The Canadian explicit criteria identified medications in the context of drug–drug and drug–disease interactions. In these various cases, the explicit criteria were obtained through a literature review and a questionnaire evaluated by national experts involved in geriatric care, clinical pharmacology, psychopharmacology, clinical pharmacy and ambulatory care. These North American criteria are not adapted to the European situation. Availability of drugs, clinical practice, socioeconomic levels and health system regulations are different from those prevailing in the United States and Canada and are even different between European countries [5, 6]. Nevertheless, we think the discrepancies between European countries are smaller than those that can be identified when comparing North American and European practices. To our knowledge, up to now, no criteria for IMs have been developed for European countries. In France, an IM list derived from the 1997 Beers criteria was elaborated in 2001 by nine French experts (five geriatricians, four pharmacologists) and used in three studies [7–9]. Most of the Beers criteria were included in this list, except drugs not available in France, drugs necessitating dose information and drugs that should not be used in the elderly under specific medical conditions. Three criteria were added: concomitant use of two (or more) nonsteroidal anti-inflammatory drugs, concomitant use of two (or more) psychotropic drugs from the same therapeutic class and use of any medications with anticholinergic properties other than those listed by Beers. This list has some limits: lack of a consensus method, exclusion of some criteria (dose, drugs in specific medical conditions), obsolescence of the list as a consequence of marketing and removal of drugs from the pharmaceutical market since 2001. Therefore, this list is hardly usable in other pharmacoepidemiological studies or in public health intervention for minimising drug-related problems. Besides, none of these lists (neither North American nor that presently used in France) suggested any alternative drugs to replace inappropriate medications. The aim of this study was to establish a list of IMs for the French elderly population using the Delphi method and to propose safer, effective alternatives.

Eur J Clin Pharmacol (2007) 63:725–731

Methods The Delphi method, developed by the Rand Corporation in the 1950s, is a research method allowing a consensus opinion to be reached among experts, using questionnaires, through an iterative process known as rounds [10]. We used two rounds in our study. The responses from the first round were collected and analysed; a revised questionnaire based on the results of this analysis was then submitted to the same experts to converge to an agreement from the average responses. The study was organised in six phases: (a) creation of the preliminary questionnaire of IMs from a literature review, (b) recruitment of the experts, (c) mailing of the round-one questionnaire, (d) analysis of the answers and creation of the new questionnaire, (e) mailing of this round-two questionnaire based on round-one synthesis and (f) final analysis. A preliminary questionnaire about IM in the elderly was constructed from the Beers lists (1991, 1997, 2003), the Canadian criteria (1997), the criteria adapted to French practice (2001), and the guidelines of the French Medicine Agency (Agence Française de Sécurité Sanitaire des Produits de Santé, AFFSaPS) on medication prescribing in the elderly (June 2005) [1–3, 4, 7, 11]. The questionnaire was composed of criteria that covered two categories: (a) medications or medication classes that should generally be avoided in the elderly as being either ineffective or prone to induce a risk when a safer alternative is available and (b) medications that should be avoided in specific medical conditions. Fifteen experts were invited to participate. The panel was composed of five pharmacologists, one pharmacoepidemiologist, five geriatricians, two pharmacists (one from hospital and one from community practice) and two general practitioners with clinical geriatric qualification (one from urban and one from rural areas). They were all selected from different geographic parts of France. All the experts accepted and participated in all the rounds of the study. In April 2006, the round-one questionnaire was sent to the panel of experts together with information on how to fill in the forms. This preliminary questionnaire included 37 criteria: 30 drugs or drug therapeutic classes independently from the diagnoses and seven criteria linked to specific medical conditions. Each criterion was to be evaluated by the experts using a 5-point Likert scale [12]; a score of 1 meant a strong agreement on the inappropriateness, 3 meant an equivocal opinion and 5 meant a strong disagreement about inappropriateness. Experts were also invited to suggest safer alternative therapeutics and to add other criteria. The responses were used to create the second questionnaire. Items with a median score of 1 or 2 were retained in the IM list; items with a median score of 4 or 5 were excluded from the list. Items with a median score of 3 were

Eur J Clin Pharmacol (2007) 63:725–731

resubmitted. A synthesis was elaborated, based on the experts’ arguments and on the criteria suggested and then sent to the experts as round-two questionnaire. In August 2006, the round-two questionnaire was sent to the panel of experts. This second questionnaire also included a synthetic table grouping the results of roundone scores, enabling the experts to compare their answers to those of the members of their expert panel. The new criteria and the criteria necessitating a consensus were evaluated with the same 5-point Likert scale. Experts confirmed the proposed therapeutic alternatives and indicated the reasons of criteria inappropriateness: (a) unfavourable benefit-torisk ratio, (b) questionable efficacy and (c) unfavourable benefit-to-risk ratio and questionable efficacy. At the end of the last round, a final list of inappropriate drugs was retained. In this second round, experts also indicated from what age this list of IMs could be applied.

Results The expert panel considered this list was applicable to people 75 years of age and older. Younger elderly patients were considered as quite similar to middle-aged adults. In the first round, 37 criteria were submitted to the panel of experts. Thirty criteria were considered by these experts as inappropriate in the elderly (median score: 1 or 2). One criterion was not selected: amiodarone (median score: 4). For six criteria, no consensus was obtained, so they were to be submitted again during the second round of evaluation (dextropropoxyphene-paracetamol, fluoxetine, long-term prescription of NSAIDs, long-term prescription of NSAIDs for patients with a history of hypertension, long-term prescription of NSAIDs for patients with a history of renal failure and antispasmodic drugs with anticholinergic properties). Three criteria were proposed by the experts (association of drugs with anticholinergic properties, concomitant use of drugs with anticholinergic properties together with anticholinesterase drugs and myorelaxants with anticholinergic properties). After the issue of the second round of evaluation, among 39 criteria, five were definitely eliminated by the French experts (dextropropoxyphene-paracetamol, fluoxetine, long-term prescription of NSAIDs, long-term prescription of NSAIDs for patients with a history of hypertension and long-term prescription of NSAIDs for patients with a history of renal failure). The final list contained 34 criteria: 29 medications or medication classes to be used in all older people and five criteria related to medications that should be avoided in specific medical conditions (Table 1). Among these 34 criteria, 25 were considered inappropriate as their benefit-to-risk ratio was unfavourable, 1 was considered with questionable efficacy and eight with both an unfav-

727

ourable benefit-to-risk ratio and a questionable efficacy (Table 1).

Discussion We propose a list of explicit criteria for identifying the potentially IM use in the population 75 years of age and older in France. These criteria were identified from conditions with, (a) an unfavourable benefit to risk ratio, (b) a questionable efficacy or (c) an unfavourable benefit-to-risk ratio together with a questionable efficacy. Safer therapeutic alternatives were also indicated for each criterion. This first French list of potentially IMs, based on a consensus of experts, regrouped the opinion of practitioners commonly involved in the management of drugs given to elderly people. The geographical dispersion and the various practice modes give a large overview of medical and pharmaceutical practices in France. The French experts considered that the list could be applicable in the the population aged 75 years and older, as from this age on, pharmacokinetic and pharmacodynamic changes are significant enough to markedly alter the response to medications [13]. The French list of potential IMs proposed here retains some of the Beers criteria. Several drugs or therapeutic drug classes were not selected, as they are not available on the French market (pentazocine, trimethobenzamide, flurazepam, meperidine, oorphenadrine, guanadrel, isoxsuprine, doxazosin, thioridazine, mesoridazine, ethacrynic acid, desiccated thyroid) or are no longer available, as they are judged harmful and ineffective, such as barbiturates (except phenobarbital). Several criteria were not considered for the consensus, as they differed from common medical practices: unopposed estrogens, for instance, are not prescribed in France without association with progesterone in the treatment of menopause. Phenylbutazone, dropped from the 2003 Beers list, was integrated in the French list because of the serious haematological effects it could induce. Among the criteria added in the 2003 Beers list, two drugs were not judged as inappropriate by French experts: fluoxetine, which was not considered as more prone to induce problems than any other selective serotonin reuptake inhibitor, and amiodarone, which was judged as effective and as susceptible to inducing adverse effects as do the other available antiarrhythmics [14]. Conversely, the French experts, like their North American counterparts, considered nitrofurantoin, short-acting nifedipine and stimulant laxatives as inappropriate in the elderly. Lastly, the obtained consensus allowed confirmation of criteria added in the previous list and adapted to French practice: concomitant use of two (or more) nonsteroidal antiinflammatory drugs, concomitant use of two (or more) psychotropic drugs from the same therapeutic class and use

Anticholinergic hypnotic drugs: doxylamine, aceprometazine, alimemazine

Dose of short- or intermediate-life benzodiazepine≤half the dose given in young subjects

Protracted activity, increased likelihood of adverse effects occurrence (drowsiness, fall...)

The aged are more sensitive to sedation, hypotension, bradycardia, Other antihypertensive drugs, except syncope short-acting calcium-channel blockers and reserpine Postural hypotension, myocardial infarction or stroke Other antihypertensive drugs, except centrally acting antihypertensives and reserpine Drowsiness, depression, GI disturbance Other antihypertensive drugs, except shortacting calcium-channel blockers and centrally acting antihypertensives

Enhanced adverse effects

Trospium or other drugs with less anticholinergic activity No association

Muscarinic-blocking drugs. To be avoided when possible

Muscarinic-blocking drugs. Sedation, drowsiness

Muscarinic-blocking drugs. Cognition impairment

SSRIs, SNRIs

Muscarinic-blocking agents with cardiotoxicity when overdosed. Tricyclics at times more active than SSRIs, but their benefit/risk ratio is less favourable in the elderly. Second-choice drugs Muscarinic-blocking drugs. Second choice drugs

Atypical antipsychotics with less anticholinergic activity (clozapine, risperidone, olanzapine, amisulpride, quetiapine), meprobamate Dose of short- or intermediate half-life hypnotic benzodiazepine ≤ half the dose given to young subjects Cetirizine, desloratadine, loratadine...

NSAIDs except phenylbutazone NSAIDs except indomethacin Use only one NSAID

Alternative drugs

Severe CNS adverse effects. Second-choice drug Severe haematological adverse effects. To be avoided No enhancement of efficacy, albeit increase of adverse effect risk

Reasons

Antiarrhythmics 14 Digoxin>0.125 mg/day or digoxin serum concentration>1.2 ng.ml−1 Increased sensitivity of the elderly. The dose should remain≤0.125 Digoxin≤0.125 mg/day or serum concentration mg/day or preferably should be adapted to maintain serum between 0.5 and 1.2 ng.ml−1 concentration < 1.2 ng.ml−1

13 Reserpine

12 Short-acting calcium-channel blockers: nifedipine, nicardipine

Anticholinergic antihistamines: promethazine, mequitazine, alimemazine, carbinoxamine, hydroxyzine, brompheniramine, dexchlorpheniramine, dexchlorpheniramine-betamethasone, cyproheptadine 8 Anticholinergic muscle relaxants and antispasmodic drugs: oxybutynin, tolterodine, solifenacin 9 Concomitant use of drugs with anticholinergic properties Sedative or hypnotic drugs 10 Long-acting benzodiazepines (half-life≥20 h : bromazepam, diazepam, chlordiazepoxide, prazepam, clobazam, nordazepam, loflazepate, nitrazepam, flunitrazepam, clorazepate, clorazepateacepromazine, aceprometazine, estazolam Antihypertensives 11 Centrally acting antihypertensives: methyldopa, clonidine, moxonidine, rilmenidine, guanfacine

7

6

Unfavourable benefit/risk ratio Analgesics 1 Indomethacin 2 Phenylbutazone 3 Concomitant use 2 or more NSAIDs Drugs with anticholinergic properties 4 Anticholinergic antidepressants: clomipramine, amoxapine, amitriptyline, maprotiline, dosulepin, doxepin, trimipramine, imipramine 5 Antipsychotic drugs: chlorpromazine, fluphenazine, propericiazine, levomepromazine, pipotiazine, cyamemazine, perphenazine

Criteria

Table 1 Final list of potentially inappropriate medications (IMs) for the French population 75 years of age and older

728 Eur J Clin Pharmacol (2007) 63:725–731

Proton-pump inhibitors and other H2 antagonists: ranitidine, famotidine, nizatidine, Osmotic laxatives

Short- or intermediate-acting sulfonylureas, insulin, metformin, alpha-glucosidase inhibitors

Confusion. More interactions than with other H2-blocking drugs

Protracted hypoglycaemia

Acute-angle glaucoma risk increased

Urinary retention risk increased

Drowsiness, amnesia, fall

Thiocolchicoside, mephenesine

Clopidogrel, aspirin

Blood and liver adverse effects

Worsening of irritable bowel syndrome

Amiodarone, other antiarrhythmics

Alternative drugs

Heart failure, anticholinergic effect

Reasons

Aggravation of urinary incontinence, postural hypotension 24 Dementia: drugs with anticholinergic properties (criteria 4–9, 15, 29, Aggravation of cognitive impairment 30, 34), trihexyphenidyl, tropatepine, biperiden, neuroleptics except olanzapine and risperidone, benzodiazepines 25 Chronic constipation: drugs with anticholinergic properties (criteria Bowel-occlusion risk, postural hypotension 4 –9, 15, 29, 30, 34) , centrally acting antihypertensives (criteria 11) Questionable efficacy 26 Cerebral vasodilators: dihydroergocristine, dihydroergocryptine, No really proven efficacy while postural hypotension and fall risks Therapeutic abstention dihydroergotoxine, ginkgo-biloba, moxisylyte, naftidrofuryl, are increased with most vasodilators nicergoline, pentoxifylline, piracetam, piribedil, raubasinedihydroergocristine, troxerutin-vincamine, vinburnine, vincamine, vincamine-rutoside Unfavourable benefit/risk ratio and questionable efficacy Sedative or hypnotic drugs 27 Dose of short- or intermediate- half-life benzodiazepines > half the No proven improvement of efficacy when the daily dose is above Dose of short- or intermediate- half-life dose given in young subjects: lorazepam>3 mg/j, oxazepam>60 half that prescribed to young adults and increase of adverse effects benzodiazepine≤ half the dose given in mg/j, alprazolam>2 mg/j, triazolam>0,25 mg/j, temazepam>15 young subjects mg/j, clotiazepam>5 mg/j, loprazolam>0,5 mg/j, lormetazepam> 0,5 mg/j, zolpidem>5 mg/j, zopiclone> 3,75 mg/j Gastrointestinal drugs 28 Meprobamate for gastro-intestinal dysfunction Drowsiness, confusion 29 Gastrointestinal antispasmodic drugs with anticholinergic properties: No proven efficacy. Muscarinic-blocking agents Mebeverine, phloroglucinol association with belladonna, clidinium bromure-chlordiazepoxide, dihexyverine, diphenoxylate-atropine, scopolamine, tiemonium

Other muscle relaxants 20 Methocarbamol, baclofen, tetrazepam With clinical conditions 21 Prostate adenoma, chronic urinary retention: drugs with anticholinergic properties (criteria 4–9, 15, 29, 30, 34) 22 Closed-angle glaucoma : drugs with anticholinergic properties (criteria 4–9, 15, 29, 30, 34) 23 Urinary incontinence: urapidil, prazosin

18 Stimulant laxatives: bisacodyl, docusate, castor oil, sodium picosulfate, cascara, sennosides, aloe... Hypoglycaemic 19 Long-acting sulfonylureas: carbutamide, glipizide

15 Disopyramide Antiplatelet drugs 16 Ticlopidine Gastrointestinal drugs 17 Cimetidine

Criteria

Table 1 (continued)

Eur J Clin Pharmacol (2007) 63:725–731 729

No association

Antibiotics with renal elimination according to the antibiogram Can induce renal insufficiency, pneumopathy, peripheral neuropathy, allergic reaction. Bacterial resistance in case of protracted use

No association

Antiplatelet drugs except ticlopidine Less efficient than aspirin. Vasodilation and postural hypotension

Drug–drug associations 33 Concomitant use of two or more psychotropic drugs from the same No improved efficacy but increase of adverse effects therapeutic class 34 Concomitant use of anticholinesterase drugs and drugs with Illogical association of two antagonistic mechanisms anticholinergic properties

Nausea: domperidone Cough: clobutinol, olexadine Drowsiness: acetyl-leucine, betahistine, Rhinitis: saline

Other drugs with anticholinergic properties 30 Antiemetics, cough suppressants, nasal decongestants, or antidrowsiness drugs with anticholinergic properties: alizapride, buclizine, dimenhydrinate, diphenhydramine, meclozine, metopimazine, oxomemazine, pheniramine, pimethixene, promethazine, association triprolidine, chlorphenamine... Antiplatelet drugs 31 Dipyridamole Antimicrobial 32 Nitrofurantoin

Reasons Criteria

Table 1 (continued)

No proven efficacy. Muscarinic-blocking agents. Confusion, sedation

Eur J Clin Pharmacol (2007) 63:725–731

Alternative drugs

730

of any medications with anticholinergic properties other than those listed by Beers. The Delphi method is a consensus technique used and validated in various health domains such as nursing, clinical practice or education [15, 16]. This technique has some limitations. The reliance on intuitive judgements is obviously not an accurate method and depends mostly on the panel of experts chosen. Every expert would summarise their experience in a single answer. However, a complex situation such as the study of the appropriateness of drugs given to elderly patients with multiple adverse-effectfacilitating factors can hardly be solved. Participants could change opinions between rounds, which may introduce some biased responses, as the results of the first round were made available to all participants who could thus compare their opinions to that expressed by their group. Experts were consulted separately and were not able to debate; this would allow for more clear-cut opinions but would complicate the path towards a consensus. Nevertheless, the Delphi method allows experts to express their opinions independently and confidentially without the pressures that may occur during a face-to-face meeting [17]. Finally, the consensus obtained is the average of the experts’ responses due to the convergence of opinions throughout the successive rounds. The Delphi method allowed the proposal of explicit criteria for assessing prescribing quality to older people. However, explicit criteria tools are stringent and do not take into account the clinical context of prescribing [18]. For instance, the NSAIDs criterion was much debated between experts with respect to dose, administration duration, continuity or not of treatment, co-prescriptions, age and renal and cardiovascular functions. Finally, NSAIDs were not classified as inappropriate, but the use of these drugs with neither adaptation to the clinical condition nor surveillance was considered harmful and inappropriate. Therefore, for several criteria, the French list encompasses some clinical conditions, especially those most frequently encountered in geriatric practice. However, their number is limited, as considering the clinical condition of the patient together with the drugs given adds complexity and hampers the epidemiological analysis. As Beers emphasised, these criteria do not allow the identification of all cases of the use of potentially IMs [2]. Only the main ones are mentioned in this study. They are based on expert opinions and are not drawn from an evidence-based methodology [19, 20]. This list of criteria is a general guide for assessing the potential inappropriateness of medications. Nevertheless, for a given patient, a benefit-to-risk ratio has also to be assessed due to evaluation of the clinical condition, comorbidities, functional status, drugs received and prognosis. This set of criteria should not be used as a substitute for

Eur J Clin Pharmacol (2007) 63:725–731

the clinical evaluation, which could lead reasonably, at times, to the appropriate prescription of supposedly inappropriate and second-choice drugs such as indomethacin, tricyclic antidepressants, or antipsychotics with anticholinergic properties (Table 1). This list should not be seen by people without adequate clinical expertise as an absolute prohibition against prescribing certain medications [19]. Care should also be taken to prevent this list from being considered solely as an opposable reference for economic purposes, for instance, by the social security system: its first aim is epidemiological; its second aim is the supply of clinical guidelines. It is not devised for the economic regulation of care but can, however, help reduce the cost of drug misuse.

Conclusion This French list of potentially IMs should be seen as a reference because it is derived from an expert consensus. Its use may help to reduce the occurrence of adverse drugrelated problems in the elderly. It could also be used as a teaching tool for training medical students and doctors in the use of appropriate drugs in the elderly. This is a public health tool, the impact of which could be measured by epidemiological studies provided it is reviewed periodically to ensure that it remains up to date and suited to the evolution of drug use. In addition, it could serve as a basis for initiating a European list based on a consensus between experts from different countries. Acknowledgements We acknowledge the following individuals for contributing their expertise to this study as panel members: Martine Alt, Jean-Pierre Charmes, Claire Dessoudeix, Jean Doucet, Annie Fourrier, Philippe Gaertner, Marie-Claude Guelfi, Alain Jean, MarieJosèphe Jean-Pastor, Claude Jeandel, Jean-Pierre Kantelip, Louis Merle, Jean-Louis Montastruc, François Piette, Jean-Marie Vetel.

References 1. Beers MH, Ouslander JG, Rollinger I, Brooks J, Reuben D, Beck JC (1991) Explicit criteria for determining IM use in nursing homes. Arch Intern Med 151:1825–1832 2. Beers MH (1997) Explicit criteria for determining potentially IM use by the elderly. An update. Arch Intern Med 157:1531–1536 3. Fick DM, Cooper JW, Wade WE, Waller JL, Maclean JR, Beers MH (2003) Updating the Beers criteria for potentially IM use in older adults. Results of a US consensus panel of experts. Arch Intern Med 163:2716–2724

731 4. McLeod PJ, Huang AR, Tamblyn RM, Gayton DC (1997) Defining inappropriate practices in prescribing for elderly people: a national consensus panel. Can Med Assoc J 156:385–391 5. Fialova D, Topinkova E, Gambassi G, Finne-Soveri H, Jonsson PV, Carpenter I, Schroll M, Onder G, Sorbye LW, Wagner C, Reissigova J, Bernabei R, AdHOC Project Research Group (2005) Potentially inappropriate medication use among elderly home care patients in Europe. JAMA 293:1348–1358 6. Gallagher P, Barry P, O’Mahony D (2007) Inappropriate prescribing in the elderly. J Clin Pharm Ther 32: 113–121 7. Lechevallier-Michel N, Gautier-Bertrand M, Alperovitch A, Berr C, Belmin J, Legrain S, Saint-Jean O, Tavernier B, Dartigues JF, Fourrier-Reglat A, The 3C Study Group (2005) Frequency and risk factors of potentially inappropriate medication use in a communitydwelling elderly population: results from the 3C study. Eur J Clin Pharmacol 60:813–819 8. Laroche ML, Charmes JP, Nouaille Y, Fourrier A, Merle L (2006) Impact of hospitalisation on inappropriate medication use in the elderly. Drugs Aging 23:49–59 9. Laroche ML, Charmes JP, Nouaille Y, Picard N, Merle L (2007) Is inappropriate medication use a major cause of adverse drug reactions in the elderly? Br J Clin Pharmacol 63:177–186 10. Dalkey NC (1969) The Delphi method: an experimental study of a group opinion. Rand Corporation, Santa Monica 11. http://agmed.sante.gouv.fr/htm/10/iatro/iatro.pdf 12. Matell MS, Jacoby J (1971) Is there an optimal number of alternatives for Likert scale items? I: reliability and validity. Educ Psychol Measure 31:657–674 13. Merle L, Laroche ML, Dantoine T, Charmes JP (2005) Predicting and preventing adverse drug reactions in the very old. Drugs Aging 22:375–392 14. Lafuente-Lafuente C, Mouly S, Longas-Tejero MA, Mahe I, Bergmann JF (2006) Antiarrhythmic drugs for maintaining sinus rhythm after cardioversion of atrial fibrillation : a systematic review of randomized controlled trials. Arch Intern Med 166: 719–728 15. Caplin DA, Rao JK, Filloux F, Bale JF, Van Orman C (2006) Development of performance indicators for the primary care management of paediatric epilepsy: expert consensus recommendations based on the available evidence. Epilepsia 47:2011–2019 16. Jones J, Hunter D (1995) Consensus methods for medical and health services research. BMJ 311:376–380 17. Stevens B, McGrath P, Yamada J, Gibbins S, Beyene J, Breau L, Camfield C, Finley A, Franck L, Howlett A, Johnston C, McKeever P, O’Brien K, Ohlsson A (2006) Identification of pain indicators for infants at risk for neurological impairment: A Delphi consensus study. BMC Pediatrics 6:1 18. Shelton PS, Fritsch MA, Scott MA (2000) Assessing medication appropriateness in the elderly. A review of available measures. Drugs Aging 16:437–450 19. Swagerty D, Brickley R (2005) American medical directors association and American society of consultant pharmacists joint position statement on the Beers list of potentially inappropriate medications in older adults. J Am Med Dir Assoc 6:80–86 20. Egger SS, Bachmann A, Hubmann N, Schlienger RG, Krähenbühl S (2006) Prevalence of potentially inappropriate medication use in elderly patients. Comparison between general medical and geriatric wards. Drugs Aging 23:823–837