Eur J Clin Pharmacol (2012) 68:1123–1138

DOI 10.1007/s00228-012-1238-1

REVIEW ARTICLE

Potentially inappropriate medications in the elderly:

a comprehensive protocol
Suzana Mimica Matanović & Vera Vlahovic-Palcevski

Received: 19 November 2011 / Accepted: 31 January 2012 / Published online: 24 February 2012
# Springer-Verlag 2012

Abstract Elderly patients are at increased risk of drug-related Keywords Potentially inappropriate medications . Elderly
morbidity and mortality. Avoiding the use of potentially inap- patients . PIM screening tools . Drug-prescribing behavior .
propriate medications (PIMs) is one of the strategies that has Drug–drug interactions
been widely adopted to reduce the harmful consequences of
drug use. There are several PIM screening tools available. In
this review, we provide an overview of existing screening tools
to detect PIMs in the elderly, emphasizing the advantages and Introduction
disadvantages of each. Combining previously published and
adopted tools (adjusted Beers list, French consensus panel, The percentage of the total elderly population is increasing
McLeod’s list, and Lindblad’s list of clinically important in most countries, and it is estimated that by 2050 almost
drug–disease interactions), we develop a new comprehensive 30% of the population in developed countries will be over
tool that also includes the adjusted Hanlon’s and Malone’s lists 65 years of age [1]. Elderly patients consume approximately
of potentially serious drug–drug interactions in the elderly. In 30% of all healthcare resources and, therefore, the growth of
addition to listed PIMs and clinically important drug–drug this population group will have significant implications on
interactions, alternative therapeutic solutions are suggested. future healthcare budgets [2]. Elderly individuals often have
The new protocol differentiates: drugs with an unfavorable many chronic diseases and are consequently taking multiple
benefit/risk ratio (to be avoided regardless of the underlying medications. They also have increased risk for adverse drug
disease/condition), drugs with a questionable efficacy, and reactions (ADRs) due to age-related changes in the pharma-
drugs to be avoided with certain diseases/conditions, and pro- codynamics and pharmacokinetics of drugs, co-morbidities,
vides a list of potentially serious drug–drug interactions. A tool and polypharmacy [3]. It may thus be anticipated that this
consisting of PIMs and potential drug–drug interactions within increase in the numbers of elderly people will also lead to
the same protocol provides more comprehensive quality as- higher drug-related morbidity and mortality [2, 4].
sessment of drug-prescribing behavior to the elderly, which in Suboptimal or inappropriate prescribing in elderly patients
turn may lead to better prescribing practices. pose the risk of drug-related morbidity and mortality. Inappro-
priate prescribing includes the prescribing of medications with
S. Mimica Matanović (*)
potentially serious drug–drug interactions or the underuse,
Clinical Pharmacology Unit, University Hospital Center Osijek, overuse, and misuse of drugs. Misuse encompasses the use of
Osijek, Croatia potentially inappropriate medications (PIMs), inappropriate
e-mail: mimica-matanovic.suzana@kbo.hr dose, or inappropriate duration of treatment. PIMs are defined
as drugs with a potential risk that is higher than their potential
S. Mimica Matanović
Department of Pharmacology, Medical School Osijek, benefit to the patient, particularly when safer alternative thera-
University J.J. Strossmayer Osijek, pies exist for the same condition [4, 5].
Osijek, Croatia Several screening tools for detecting PIMs in the elderly
V. Vlahovic-Palcevski
have been developed in the USA, Canada, and European
Clinical Pharmacology Unit, University Hospital Center Rijeka, countries [6–16]. Their role is to optimize the appropriateness
Rijeka, Croatia of prescribing behavior and to reduce negative outcomes,
1124 Eur J Clin Pharmacol (2012) 68:1123–1138

including preventable adverse drug effects (ADEs). These combining their clinically most useful parts together, it
screening tools encompass lists of drugs that should generally should be possible to develop a new comprehensive tool.
be avoided by the elderly and drugs that should be avoided with In evaluating the advantages of previous tools, we focused
certain diagnoses or conditions. While most of the protocols are on the potential clinical applicability of the criteria and on
explicit (criterion-based), there is only one implicit (judgement- PIMs having common clinical consequences in the elderly
based) protocol, the Medication Appropriateness Index (MAI), [e.g., focusing on non-steroid anti-inflammatory drugs
which was developed by Hanlon and colleagues [17] . (NSAIDs), associated with multiple possible adverse
The aim of this review is to critically evaluate available outcomes].
protocols for detecting PIMs in the elderly and summarize these The new protocol is developed by combining the adjust-
into a new comprehensive and widely applicable protocol. ed Beers list, the French consensus panel, McLeod’s list,
and Lindblad’s list of clinically important drug–disease
interactions, with the addition of several new drugs [8,
Overview of the existing screening tools for detection 10–12]. As part of the protocol, a list of clinically important
of PIMs in the elderly drug–drug interactions in the elderly is developed by mod-
ifying Malone’s and Hanlon’s lists and adding four new
A literature search was performed within the MEDLINE, drug–drug interactions [35, 36]. A clinically based approach
PubMed, OVID, and Google Scholar databases using MeSH was used to build the protocol. The new criteria have been
terms “aged,” “inappropriate prescribing,” “clinical proto- shown by many researchers to be associated with adverse
cols,” “medication errors,” and “polypharmacy.” Articles clinical and healthcare outcomes, thus confirming their rel-
published between January 1991 and December 2010 were evance [27–34]. By combining parts of the North American
selected if they contained explicit criteria adressing poten- and European tools together, we assume that new combined
tially inappropriate prescribing in the elderly. tool will be widely applicable in the elderly population
Table 1 summarizes the protocols and screening tools for (defined as 65 years or older).
detection of PIMs prescribed to the elderly, published in The protocol groups PIMs into: (1) those with an unfa-
chronological order. For each tool in the table, data on the vorable benefit/risk ratio (drugs to be avoided regardless of
authors, year, country of origin, scope, main content, and the underlying disease/condition); (2) drugs with a question-
main advantages/disadvantages are presented. One of the able efficacy, and (3) drugs to be avoided with certain
tools was developed by Barry et al. [13] to evaluate under- diseases/conditions. Those three major subgroups are de-
prescribing, i.e., for detecting omission of evidence-based fined similarly in the French consensus panel and on Beers
medications. list. For each criterion listed, a possible alternative solution
We considered the potential clinical applicability of the is given [8, 12, 16]. Part (4) of the protocol presents the list
criteria and PIMs resulting in common clinical consequen- of potentially serious drug–drug interactions in the elderly
ces as well as the wide applicability of a protocol to different population.
healthcare settings and different geographical regions to be
advantageous. Limitations or disadvantages of a protocol Drugs with unfavorable benefit/risk ratio
were considered to be a lack of clinical assessment and (part 1 of new protocol)
evaluation or validation of the protocol, incomplete drug
listings, and/or obsolete drugs listed. The list of drugs with an unfavorable benefit/risk ratio is
According to different authors, newer criteria offer cer- based on a combination of the adjusted Beers list and the
tain innovations and improvements compared to those on French consensus panel, as shown in Table 2, and consists
the original Beers list and address drug–drug interactions, of 33 criteria or individual drugs [10, 12]. Drugs defined
underuse of drugs, drugs with questionable efficacy, among solely by the French panel include antipsychotic drugs with
others [18–20]. However, while several new tools need to be anticholinergic properties and concomitant use of ≥2
evaluated and assessed in clinical studies, Beers criteria NSAIDs, clonidine, and moxonidine among the centrally
have been the most widely used and an association of the acting antihypertensives and dipyridamole (according to
listed drugs with adverse outcomes has been shown by Beers list, dipyridamole is inappropriate medication only
number of authors [21–34]. in patients receiving anticoagulant therapy or in those with
blood clotting disorders). These drugs are included in the
new protocol because their use can lead to severe ADRs
The new comprehensive protocol [37–42] .
Drugs defined solely by the Beers list include: doxazosin,
Taking into account the reported advantages and disadvan- amiodarone, fluoxetine, thioridazine, ferrous sulfate >325 mg/
tages of existing screening tools, it may be assumed that by day, estrogens only (oral), methyltestosterone, long-term use
Table 1 Characteristics of existing screening tools for potentially inappropriate medications

Authors Year Country Scope Main content Pro’s and con’s

Beers et al. [6] 1991 USA Nursing home residents ≥65 years Delphi consensus based. Thirty criteria for drugs to be Pros: the first tool developed for PIM screening
avoided in the elderly in the elderly
Cons: many drugs from the list are unavailable in other
countries. Developed for nursing home residents but
also used in studies with other patient populations
Beers [7] 1997 USA All patients ≥65 years Delphi consensus based; updated and expanded Pros: more generally applicable (for ambulatory
version. Fifteen drugs omitted from the original patients)
version. Contains 28 drugs or drug classes to be
avoided in ambulatory elderly independent of Cons: many drugs from the list are unavailable in other
diagnosis and 35 drugs or drug classes to be countries. Drug–drug interactions and duplication of
Eur J Clin Pharmacol (2012) 68:1123–1138

avoided in patients with certain disease or condition treatments are not evaluated
McLeod et al. [8] 1997 Canada All patients ≥65 years Delphi consensus based. Thirty-eight inappropriate Pros: Nine inappropriate practices address prescribing of
practices (grouped into cardiovascular, psychotropic, NSAIDs, including long-term prescription in patients
analgesic and miscellaneous drugs) with a history of pepticulcer, hypertension, chronic
• drug generally
renal, or congestive heart failure. Drug–drug interactions
contraindicated (18)
• drug–disease interactions (16)• drug–drugs
addressed. Alternative therapy for each criterion
interactions (4) suggested
Cons: some of the criteria obsolete (e.g., beta blockers
in patients with asthma or COPD or beta blockers in
patients with congestive heart failure)
Naughler et al. [9] 2000 Canada All patients ≥70 years Derived from McLeod’s list. Fourteen inappropriate Pros: simple and easily applicable tool
combinations of drugs and diseases
Cons: some of the criteria obsolete (e.g., beta blockers in
patients with asthma or COPD or beta blockers in
patients with congestive heart failure). Three of the
criteria involve today uncommonly used tricyclic
antidepressants
Fick et al. [10] 2003 USA All patients ≥65 years Delphi consensus based updated version of Beers list. Pros: the most widely cited explicit criteria. Association
Sixty-eight criteria: 48 drugs or drug classes generally with adverse healthcare outcomes shown
to be avoided; 20 diseases or conditions with drugs
to be avoided Cons: many drugs from the list are unavailable in other
countries. Drug–drug interactions and duplication of
treatments are not evaluated. Appropriateness of some
drugs from the list still subject of debate (e.g.,
amiodarone or amitriptyline). Only four inappropriate
practices associated with the use of NSAIDs addressed
Lindblad et al. [11] 2006 USA All patients ≥65 years Delphi consensus based. Twenty-eight clinically Pros: simple and easily applicable tool. Introduces
important drug–disease interactions, involving 14 new criteria not defined by Beers’ or McLeod’s list
diseases or conditions. Eleven drug-disease interactions
included on Beers list and 5 included in McLeod’s list Cons: drugs to be avoided regardless of a disease or
condition are not included
Laroche et al. [12] 2007 France All patients ≥75 years Delphi consensus based, first European screening tool. Pros: Alternative drugs or therapeutic abstention for
Thirty-four criteria for inappropriateness. PIMs grouped each criterion suggested. The first tool to address
into those with unfavorable benefit/risk ratio (25 criteria), drugs with questionable efficacy as potentially
questionable efficacy (one criterion), and both
unfavorable benefit/risk ratio and questionable efficacy
1125
Table 1 (continued)
1126

Authors Year Country Scope Main content Pro’s and con’s

(7 criteria). Twenty-nine criteria are independent of a inappropriate (cerebral vasodilators). Duplication

disease or condition and five are linked to disease or of treatments addressed as potentially inappropriate
condition
Cons: underuse of drugs is not addressed. Needs to
be assessed and confirmed in clinical studies
Barry et al. [13] 2007 Ireland All patients ≥65 years Delphi consensus based. START (Screening Tool to Alert Pros: first tool evaluating underuse of drugs
Doctors to Right Treatment). Twenty-two medications
included. Underprescribing or omission of clinically
indicated, evidence-based medications evaluated (e.g.,
ACE inhibitors in chronic heart failure or beta blocker in
chronic stable angina, if no contraindications).
Gallagher et al. [14] 2008 Ireland All patients ≥65 years Delphi consensus based. STOPP (Screening Tool of Older Pros: innovative approach introduced (first tool to
Persons’ Prescription). Sixty-five criteria for PIMs address inappropriate use of PPI for peptic ulcer and
evaluation arranged according to relevant physiological use of aspirin without history of coronary, cerebral
systems.
or peripheral vascular symptoms or occlusive
event). Seven criteria address prescribing of
NSAIDs, including prescription in patients
hypertension, chronic renal or congestive heart failure.
Drug class duplication and drug–drug interactions
addressed as potentially inappropriate. Sensitive for
identifying patients with potential to suffer PIMs-
related ADRs
Cons: drugs with questionable efficacy not addressed.
Needs to be assessed and confirmed in clinical studies.
Rognstad et al. [15] 2009 Norway Patients ≥70 years in general practice Delphi consensus based. Twenty-one explicit criteria for Pros: Lists inappropriate single drugs (i.e., theophylline
single drugs and 15 criteria for drug–drug interactions. or sotalol) and drug combinations [e.g., combination
of NSAIDs (or coxib) and ACE inhibitors (or ARBs)
which may increase the risk of renal failure, or
combination of NSAIDs and diuretics, resulting in
reduced diuretic effect] not addressed by previous
tools
Cons: aimed at patients in general practice. Needs to be
assessed and confirmed in clinical studies
Holt et al. [16] 2010 Germany All patients ≥65 years Delphi consensus based. Lists a total of 83 drugs to be Pros : easily applicable tool. Names main concerns,
avoided regardless of the underlying disease/condition, possible therapeutic alternatives and precautions to
contained in 18 drug classes. be taken. Lists many drugs not addressed by previous
tools (e.g., ketoprofen, meloxicam, prasugrel, flecainide,
metildigoxin, haloperidol >2 mg, zaleplon >5 mg,
phenobarbital). Drugs with questionable efficacy are
addressed (circulation-promoting agents)
Cons: aimed primarily at German elderly population.
Needs to be assessed and confirmed in clinical studies

ACE, Angiotensin converting enzyme; ADRs, adverse drug reactions; ARBs, angiotensin II receptor blockers; COPD, obstructive pulmonary disease; NSAIDs, non-steroid anti-inflammatory drugs;
PIMS, potentially inappropriate medications, PPI, proton pump inhibitor
Eur J Clin Pharmacol (2012) 68:1123–1138
Table 2 Drugs with unfavorable benefit/risk ratio

Drug Possible adverse effects Possible therapeutic solutions

Analgesics
Indomethacin Severe CNS side effects Short-term use of a weak NSAID (e.g., ibuprofen) or acetaminophen
or a weak opioid (e.g., tramadol)
Concomitant use of 2 or more NSAIDs No enhancement of efficacy, increased risk of ADRs Short-term use of only one weak NSAID (e.g., ibuprofen)
Long-term use of full-dosage, longer half-life NSAIDs: naproxen, Increased risk of GI bleeding, renal failure, high blood pressure Short-term use of a weak NSAID (e.g., ibuprofen) or use of
piroxicam and heart failure acetaminophen or a weak opioid (tramadol, codeine)
Drugs with anticholinergic properties
Antidepressants: amitriptyline, maprotiline Muscarinic-blocking side-effects, cardiotoxicity when SSRIs (except flouxetine) or SNRIs
overdosed
Eur J Clin Pharmacol (2012) 68:1123–1138

Antipsychotic drugs: fluphenazine, levomepromazine Muscarinic-blocking side-effects Atypical antipsychotic drug with less anticholinergic activity (e.g.,
olanzapine, risperidone, quetiapine)
Antihistamines: diphenhydramine, dimenhydrinate Muscarinic-blocking side-effects, sedation, drowsiness Antihistamines without anticholinergic activity (e.g., cetirizine,
levocetirizine, loratadine, desloratadine)
Concomitant use of drugs with anticholinergic properties Enhanced anticholinergic ADRs Avoid drugs with anticholinergic activity in general
Sedative or hypnotic drugs
Long-acting benzodiazepines: diazepam, bromazepam, nitrazepam, Prolonged sedation and drowsiness, increased risk of falls Short-acting benzodiazepines given in dose ≤half the dose in
flurazepam younger adults
Short-acting benzodiazepines, dose >half the dose in younger Increased risk of ADRs without increased efficacy Short-acting benzodiazepines given in dose ≤half the dose in
adults (lorazepam >3 mg, oksazepam >60 mg, alprazolam >2 mg) younger adults
Meprobamat Very sedative properties, addictive with prolonged use Short-acting benzodiazepines given in dose ≤half the dose in
younger adults
Antihypertensives
Methyldopa Bradycardia, exacerbation of depression Other antihypertensive drugs, except the ones listed here [i.e.,
Clonidine Ortostatic hypotension diuretics, calcium channel blockers (except short-acting ones),
Moxonidine ACE inhibitors, AT1 blockers]
Headache, vertigo, asthenia
Nifedipine, short-acting Postural hypotension, myocardial infarction, stroke
Doxazosine Hypotension, dry mouth, urinary incontinence
Anti-arrhythmics
Amiodarone Prolonged QT interval, risk of “torsade de pointes”, reduced Other antiarrhythmics, depending on the type of arrhythmia (e.g.,
efficacy in the elderly propafenone, beta blockers, calcium channel blockers)
Disopiramide Negative inotropic and anticholinergic properties
Digoxin >0.125 mg Reduced renal clearance and increased risk of ADRs Digoxin <0.125 mg, with serum concentrations 0.5–1.2 ng/ml
Antiplatelet drugs and vasodilators
Ticlopidine Blood and liver adverse effect Clopidogrel, aspirin
Dipiridamole Vasodilation and postural hypotension, questionable efficacy
Drugs used to treat gastrointestinal disorders
Cimetidine CNS adverse events, confusion, common drug interactions Other H2-antagonists or PPIs
Scopolamine Muscarinic-blocking agent, no proven efficacy Mebeverine
Long-term use of stimulant laxatives: bisacodyl, sennosides Worsening of irritable bowel syndrome Osmotic laxatives (e.g., lactulose)
1127
Table 2 (continued)
1128

Drug Possible adverse effects Possible therapeutic solutions

Long-acting sulfonylureas
Chlorpropamide, glibenclamide Protracted hypoglycemia Short- or immediate-acting sulfonylureas (e.g., glipizide, gliclazide)
Muscle relaxants
Baclofen Drowsiness, amnesia, fall Thiocolchicoside, mephenesine
Opioid analgesics
Pentazocin More CNS adverse effects, including confusion and Other opioids, with more favorable risk/benefit profile (e.g.,
hallucinations; mixed agonist and antagonist tramadol, oxycodone)
Meperidine Not an effective oral analgesic in doses commonly used. May
cause confusion
Other
Ferrous sulfate >325 mg/day Increased incidence of constipation Dose <325 mg/day
Nitrofurantoin Can induce renal insufficiency, pneumopathy, peripheral Other type of antibiotics, depending on microbiology results
neuropathy, allergic reaction
Methyltestosterone Potential for prostatic hypertrophy and cardiac problems Avoid testosterone substitution
Estrogens only (oral) Carcinogenic potential and lack of cardioprotective effect If indicated, use combination of estrogens with progesterones. HRT
to be used for the shortest time possible.
Thioridazine Greater potential for CNS and extrapyramidal adverse effects Atypical antipsychotic drug with less extrapyramidal adverse effects
(e.g., olanzapine, quetiapine)
Daily fluoxetine Long half-life of drug and risk of producing excessive CNS Other SSRIs or SNRIs
stimulation, sleep disturbances, and increasing agitation

CNS, Central nervous system; GI, gastrointestinal; , SNRIs, serotonin and norepinephrine reuptake inhibitors; SSRI, selective serotonin reuptake inhibitors
Eur J Clin Pharmacol (2012) 68:1123–1138
Eur J Clin Pharmacol (2012) 68:1123–1138 1129

of naproxen and piroxicam, and use of the opiates pentazocine potential to cause ADRs in the elderly, such as postural
and meperidine. French experts did not find those drugs hypotension, falls, headache, or stomach upset [55–60].
relevant. However, since these drugs are commonly used The French panel recognizes a category of drugs with
and possible ADRs might have serious consequences, they both unfavorable benefit/risk ratio and questionable effica-
are included the new protocol [43–49]. For example, even cy, such as nitrofurantoin, anticholinergic antispasmodic
though the European Medicines Agency in 2007 recommen- scopolamine, or short/intermediate half-life benzodiazepines
ded restricted use of piroxicam, it seems that the extent of its in doses higher than half the dose given in the young
restricted prescribing among the elderly is not satisfactory patients. In the new protocol these drugs are classified
[50–52]. Amiodarone is a commonly prescribed antiarrhyth- among those with unfavorable benefit/risk ratio [12].
mic drug, partly due to the lack of similar antiarrhythmic drugs
on the market. However, not only is it associated with QT
prolongation in the elderly, but its complex pharmacokinetic Drugs to be avoided with certain diseases/conditions (part 3
properties, very long half-life, and possible multiple drug of new protocol)
interactions make it a drug that warrants high caution when
used [45–49]. This part of the new protocol includes 71 individual drug–
Drugs with unfavorable benefit/risk ratio listed in the disease interactions involving 28 diseases or conditions, as
new protocol that are on neither the French or Beerslist shown in Table 4. Criteria from the adjusted Beers list,
include glibenclamide, a long-acting sulphonylurea drug McLeod’s list, and Lindblad’s list of clinically important
associated with a higher risk of inducing prolongedhypo- drug–disease interactions are combined and respective data
glycemia, and promazine, an antipsychotic drug with strong updated [10–12, 61–69]. Some of the inappropriate drug–
anticholinergic properties [40, 53, 54]. disease combinations are defined by all three tools, such as
The remaining drugs with an unfavorable benefit/risk NSAIDs in peptic ulcer disease or tricyclic antidepressant in
ratio on the new list are in agreement with both the French benign prostate hyperplasia. However, only McLeod’s list
and Beers list as being potentially inappropriate. recognizes the long-term use of NSAIDs in patients with
heart failure and hypertension or long-term use of NSAIDs
to treat osteoarthritis regardless of the accompanying dis-
Drugs with a questionable efficacy (part 2 of new protocol) ease, as potentially inappropriate. NSAIDs are commonly
prescribed to the elderly, and their risk/benefit ratio should
The French list of PIMs in the elderly is the only available regularly be checked in individual patients. As a general
tool that contains drugs with questionable efficacy as poten- principle, the long-term use of NSAIDs should be avoided
tially inappropriate [12]. Medications defined as question- in patients with chronic renal or heart failure, hypertension,
ably efficacious are cerebral vasodilators, such as gingko and peptic ulcer disease, but also in older patients regardless
biloba, pentoxyphylline, piracetam, and dihydroergotoxine. of their underlying disease [38]. Not only is there always the
These four commonly used vasodilator drugs are included in possibility that NSAIDs can worsen renal or heart failure or
the new protocol, with the addition of two more drugs, cause gastrointestinal toxicity, but there are emerging data
betahistine and cinarizine (except in the indication of on the increased risk of stroke and myocardial infarction
Meniere syndrome and vestibular vertigo), which are often associated with their use [39]. Also, gastrointestinal bleed-
used as cerebral vasodilators in patients with cerebrovascu- ing is often the most common cause of ADRs related hos-
lar insufficiency (Table 3). Drugs used as cerebral vaso- pital admissions in the elderly [70–72]. Therefore, it seems
dilators do not have a proven efficacy but they do have a reasonable to include McLeod’s NSAIDs criteria in the new

Table 3 Drugs with question-

able efficacy Drug Possible Possible
adverse effects therapeutic solutions

Cerebral vasodilators
Dihydroergotoxine No really proven efficacy while Therapeutic abstention
Gingko biloba there is risk of postural
Pentoxifylline hypotension, falls, headache, or
stomach upset
Piracetam
Betahistine (except in the indication of
Meniere syndrome and vestibular vertigo)
Cinarizine (except in the indication of
Meniere syndrome and vestibular vertigo)
Table 4 Drugs to be avoided with certain diseases/conditions
1130

Disease or condition/drug Possible adverse effects Possible therapeutic solutions

Heart failure
Disopyramide Negative inotropic effect Antiarrhythmic drug without negative inotropic effect
High sodium content drugs (sodium and sodium salts Potential to promote fluid retention and exacerbation Avoid this type of drugs
[bicarbonate, biphosphate, citrate, phosphate, salicylate, and of heart failure
sulfate])
Calcium-channel blockers, except , dihydropyridines Negative inotropic effect. Avoid verapamil and dilatiazem in these patients. Depending
on the underlying diagnosis (hypertension, angina), drug
without negative inotropic effect should be used
Long-term prescription of NSAIDs Potential to promote fluid retention and exacerbation Acetaminophen or a weak opioid (e.g., tramadol).
of heart failure Monitoring of cardiovascular function.
Hypertension
Pseudoephedrine May produce elevation of blood pressure secondary Avoid over-the-counter cold and cough medications
to sympathomimetic activity containing this substance.
Long-term prescription of NSAIDs May produce elevation of blood pressure secondary Acetaminophen or a weak opioid (e.g., tramadol).
to salt and water retention Monitoring blood pressure.
Chronic renal failure
Long-term prescription of NSAIDs May reduce renal blood flow and worsen renal failure Acetaminophen or a weak opioid (e.g., tramadol).
Monitoring renal function.
Gastric or duodenal ulcers
NSAIDs May exacerbate existing or produce new ulcers Acetataminophen or a weak opioid (e.g., tramadol). If use of
NSAID is deemed unavoidable, use NSAID with less
gastrointestinal risk (e.g., ibuprofen) in combination with
proton pump inhibitors.
Aspirin May exacerbate existing or produce new ulcers Use in combination with proton pump inhibitors
Seizures or epilepsy
Clozapine, thioridazine May lower seizure threshold Atypical antipsychotic drug without proconvulsive effect
and with a favorable risk to benefit profile (e.g.,
olanzapine, risperidone, quetiapine)
Bupropion May lower seizure threshold SSRI (except fluoxetine) or SSNI
Blood clotting disorders or receiving anticoagulant therapy
NSAIDs Increased potential for bleeding For analgesia use acetaminophen or a weak opioid (e.g.,
tramadol)
Aspirin Increased potential for bleeding If the combination is deemed unavoidable, use with extreme
Clopidogrel caution and regular monitoring of the patient
Cimetidine (those taking warfarin) May elevate INR values and increase potential for Other H2-antagonist or proton pump inhibitor
bleeding
Bladder outflow obstruction
Anticholinergics May decrease urinary flow, leading to urinary Use drugs without anticholinergic activity
retention
Stress incontinence
Eur J Clin Pharmacol (2012) 68:1123–1138
Table 4 (continued)

Disease or condition/drug Possible adverse effects Possible therapeutic solutions

Alpha blockers (doxazosin, urapidil) May induce or worsen incontinence Use alternative antihypertensives (e.g., ACE inhibitors or
Anticholinergics AT1 blockers), antidepressants (SSRI or SNRI) and
Tricyclic antidepressants sedative/hypnotics (e.g., short-term use of short-acting
benzodiazepines). Avoid anticholinergics.
Long-acting benzodiazepines
Arrhythmias
Tricyclic antidepressants Concern due to proarrhythmic effects and ability to SSRI (except fluoxetine) or SSNI
produce QT interval changes
AV block
Tricyclic antidepressants May worsen heart block SSRI (except fluoxetine) or SSNI
Eur J Clin Pharmacol (2012) 68:1123–1138

Digoxine May worsen heart block Avoid digoxine, non-dihydropyridine calcium channel
Verapamil blockers, beta blockers and antiarrhythmics
Insomnia
Decongestants Concern due to CNS stimulant effects Short term topical application
Theophylline Inhaled brochodilators
Methylphenidate Avoid this type of drug
MAO inhibitors Another type of antidepressant
Parkinson disease
Metoclopramide Concern due to their antidopaminergic/cholinergic Another antiemetic drug
Conventional antipsychotics (fluphenazine, haloperidol) effects Atypical antipsychotics with less D2-blocking activity (e.g.,
quetiapine or clozapine)
Acetylcholinesterase inhibitors (donepezil) Memantine for treatment of dementia
Depression
Long-term benzodiazepine use May produce or exacerbate depression Short-term benzodiazepine use
Methylphenidate Avoid this type of drug
Sympatholytic agents: methyldopa and reserpin Another type of antihypertensive drugs (including beta
blockers)
Anorexia and malnutrition
CNS stimulants (methylphenidate) Concern due to appetite-suppressing effects Avoid this type of drugs
Fluoxetine Use cautiously another type of SSRI or SNRI with shorter
half-life
Syncope or falls
Short- to intermediate-acting benzodiazepines May produce ataxia, impaired psychomotor function, Avoid benzodiazepines
Long-acting benzodiazepines syncope, and additional falls Avoid benzodiazepines
Tricyclic antidepressants SSRI (except fluoxetine) and SRNI
Conventional antipsychotics (fluphenazine, haloperidol) Atypical antipsychotics with less alpha blocking activity (e.g.,
quetiapine, ziprasidone, aripiprazole)
1131
Table 4 (continued)
1132

Disease or condition/drug Possible adverse effects Possible therapeutic solutions

SIADH/hyponatremia
SSRIs May exacerbate or cause SIADH Most antidepressants and antipsychotics linked to SIADH.
Consider stopping the treatment or use a drug with a
different pharmacological profile and monitor serum
sodium levels. Consider concomitant treatment with
demeclocycline
Obesity
Olanzapine May stimulate appetite and increase weight gain Atypical antipsychotic drug with less effect on weight
gain (e.g., ziprasidone or aripiprazole)
COPD or asthma
Long-acting benzodiazepines May exacerbate or cause respiratory depression Short-term use of short-acting benzodiazepines
Beta blockers—nonselective Cardioselective beta-blockers
Chronic constipation
Calcium channel blockers May cause constipation Other types of antihypertensives, except centrally acting, or
another types of antianginal drugs (e.g., beta blockers,
nitrate)
Anticholinergics Avoid anticholinergics.
Centrally acting antihypertensives Other types of antihypertensives, except calcium channel
blockers
Opioid analgesics Concomitant use of osmotic laxatives
Gout
Thiazide diuretics May precipitate or worsen gout Use different type of antihypertensive (e.g., AT1 blocker
losartan or a calcium channel blocker amlodipine)
Diabetes
Corticosteroids May precipitate or worsen diabetes If corticosteroid therapy is unavoidable, use smallest dose
possible and monitor blood glucose
Narrow-angle glaucoma
Anticholinergics Acute-angle glaucoma risk increased Use drugs without anticholinergic activity
Osteoarthritis
Long-term prescription of NSAIDs May cause gastropathy, bleeding and salt and water Acetaminophen or a weak opioid (e.g., tramadol). NSAIDs
retention to be used for the shortest time possible (considering
gastrointestinal risk of an individual NSAIDs)
Extrapyramidal effects of antipsychotic drugs
Anticholinergics (e.g., biperiden) May cause agitation, delirium and impaired cognition Atypical antipsychotic drug with less extrapyramidal
adverse effects (e.g., olanzapine, quetiapine)
Dementia
Anticholinergics May worsen cognitive impairment Avoid all anticholinergics
Biperiden Avoid and use other antiparkinson drug
Eur J Clin Pharmacol (2012) 68:1123–1138
Eur J Clin Pharmacol (2012) 68:1123–1138 1133

PIMs screening tool. Preventing excessive NSAID use in

the elderly is one of the most important strategies to reduce

properties (e.g., olanzapine, risperidone, quetiapine)

Atypical antipsychotic drug without alpha blocking
drug-related morbidity and mortality.

Atypical antipsychotics: risperidone, olanzapine,

Lindblad’s list defines corticosteroid use in diabetic

AV, Atrioventricular; INR, international normalized ratio; MAO inhibitors, monoamine oxidase inhibitors; SIADH, syndrome of inappropriate antidiuretic hormone hypersecretion
Short-term use of low-dose benzodiazepines
patients or opiate use in patients with constipation as inap-
propriate [11]. Both of these drug–disease combinations are
considered clinically important and included in the new

SSRI (except fluoxetine) or SSNI

protocol.
Possible therapeutic solutions

Potentially serious drug–drug interactions (part 4 of new

Calcium channel blocker

Avoid this type of drugs

protocol)

In the quality assessment of drug-prescribing behavior to the

aripiprazole

elderly, an evaluation of potential drug–drug interactions lead-

ing to clinically important events is necessary. A list of 25
Delphi method-based drug–drug interactions with the greatest
clinical importance in the general population was developed
by Malone and colleagues in 2004 [35]. It was an important
expert-based attempt to assess the clinical importance of drug
interactions. These 25 interactions mostly involve drugs with
a narrow therapeutic index (i.e., warfarin, cyclosporine, or
digoxin) and generally are pharmacokinetic in their origin,
acting through inhibition or induction of hepatic drug metab-
May worsen the underlying condition

olism. Hanlon and Schmader [36] analyzed possible drug

May worsen postural hypotension

interactions in the elderly and suggested broadening Malone’s

list with seven clinically significant pharmacokinetic drug–
drug interactions involving antiarrythmics, 12 involving anti-
Possible adverse effects

epileptics, and 15 involving other drugs. They also suggested

adding nine clinically significant pharmacodynamic drug–
drug interactions, such as the combination of ACE inhibitors
with potassium-sparing diuretics or potassium supplements
[36].
In order to extend the evaluation of prescribing quality,
we included the adjusted Malone’s and Hanlon’s lists of
potentially serious drug–drug interactions in the new screen-
ing tool (Table 5) [35, 36]. As Malone’s list focuses on the
general population, we excluded drugs never or rarely pre-
scribed in the elderly (e.g, oral contraceptives, zidovudine,
or dexfenfluramine). Sibutramine was also excluded as it
was withdrawn from all markets in the European Union in
Raynaud disease or peripheral vascular disease

January 2010 due to increased cardiovascular risk [73]. The

original Malone’s list (25 drug–drug interactions) was re-
Long-term prescription of beta blockers

duced to 17 drug interactions overall. Among other potential

clinically important drug combinations assessed by the
Malone’s expert panel but not reaching the final list were
four combinations of drugs commonly used by the elderly:
Conventional neuroleptics

Tricyclic antidepressants
Disease or condition/drug

levodopa–monoamine oxidase inhibitors inhibitors; potassi-

Postural hypotension
Table 4 (continued)

um–potassium-sparing diuretics; HMG Co-A reductase

All benzodiapines

inhibitors–gemfibrozil; cytochrome P450 (CYP) 3A4-

Thioridazine
Barbiturates

metabolized HMG Co-A reductase inhibitors–macrolide

antibiotics [35]. In addition to these four combinations, the
new protocol includes another four important potential inter-
actions of drugs commonly prescribed to the elderly and
1134 Eur J Clin Pharmacol (2012) 68:1123–1138

Table 5 Potentially serious drug–drug interactions

1. Clinically significant pharmacokinetic drug-drug interactionsa

Antiarrhythmics
Disopyramide–Cimetidine ↑
Disopyramide–Macrolides (except azithromycin) ↑
Procainamide–Amiodarone ↑
Procainamide–Cimetidine ↑
Procainamide–Trimethoprim ↑
Quinidine–Cimetidine ↑
Quinidine–Fluvoxamine ↑
Antiepileptics
Carbamazepine–Danazol ↑
Carbamazepine–Diltiazem ↑
Carbamazepine–Macrolides ↑
Carbamazepine–Verapamil ↑
Phenytoin–Amiodarone ↑
Phenytoin–Cimetidine ↑
Phenytoin–Fluoxetine ↑
Phenytoin–Isoniazid ↑
Phenytoin–Omeprazole ↑
Quinidine–Phenytoin ↓
Theophylline–Phenytoin ↓
Warfarin Phenytoin–PT-INR ↓
Other drugs with low therapeutic index
Digoxin–Clarithromycin ↑
Digoxin–Amiodarone ↑
Digoxin–Propafenone ↑
Digoxin–Quinidine ↑
Digoxin–Verapamil ↑
Lithium–ACE inhibitors ↑
Lithium–Diuretics ↑
Lithium–NSAIDs ↑
Procainamide–Cimetidine ↑
Salicylates–Probenecid ↑
Theophylline–Cimetidine ↑
Theophylline–Erythromycin, clarithromycin ↑
Warfarin–Amiodarone ↑
Warfarin–Macrolides ↑
Warfarin–Quinolones ↑
Warfarin–Sulfamethoxazole ↑
2. Drug–drug interactions selected by Malone et al. as having greatest clinical importance (pharmacokinetic and pharmacodynamic)b
Benzodiazepines–Azole antifungal agents
Cyclosporine–Rifampin
Ergot alkaloids–Macrolide antibiotics (except azithromycin)
MAO inhibitors–Sympathomimetics (dopamine, ephedrine, phenylephrine, pseudoephedrine)
Meperidine–MAO inhibitors
Methotrexate–Trimethoprim
Nitrates–Sildenafil
SSRIs–MAO inhibitors
Theophyllines–Fluvoxamine
Theophyllines–Quinolones
Eur J Clin Pharmacol (2012) 68:1123–1138 1135

Thiopurines–Allopurinol
Warfarin–Fibric acid derivatives
Warfarin–NSAIDs
Warfarin–Cimetidine
Warfarin–Thyroid hormones
Warfarin– Barbiturates
3. Other clinically important drug–drug interactions (pharmacokinetic and pharmacodynamic)c
Atorvastatin/simvastatin–Amiodarone
Potassium– Potassium-sparing diuretics
Clopidogrel–PPIs
Levodopa–MAO inhibitors
SSRIs–Metoclopramide
SSRIs–Tramadol
HMG Co-A reductase inhibitors–gemfibrozil
Atorvastatin/simvastatinMmacrolide antibiotics
4. Clinically significant pharmacodynamic drug–drug interactionsd
Object drug/drug class Interacting drug/drug class Outcome
ACE inhibitors Potassium-sparing diuretics ↑ Potassium level
ACE inhibitors Potassium supplements ↑ Potassium level
Anticholinergic Anticholinergic ↑ Anticholinergic effect
Antihypertensive NSAIDs ↓ Antihypertensive effect
CNS agents (e.g., diazepam) CNS agents (e.g., codeine) ↑ CNS effect
Diuretics NSAIDs ↓ Diuretic effect
NSAIDs, Aspirin Corticosteroids ↑ Peptic ulcer risk
Verapamil Beta blockers ↓ Heart rate
Warfarin Antiplatelet agents ↑ Risk of bleeding
Antiplatelet agent Antiplatelet agent ↑ Risk of bleeding

Arrow indicates the influence of the second (interacting) drug on the concentration of the first, object drug
a
Clinically significant pharmacokinetic drug–drug interactions selected by Hanlon and Schmader [36]
b
Clinically significant drug-drug interactions selected by Malone et al. [35]
c
Other clinically important drug–drug interactions, not selected by Malone’s panel, including an additional four drug–drug interactions we added [35]
d
Clinically significant pharmacokinetic drug–drug interactions selected by Hanlon and Schmader, including additional two drug–drug interactions
we added [36]

possibly leading to serious adverse events: combination of defined by Hanlon and Schmader [36]. The new protocol
amiodarone with CYP3A4-metabolized HMG Co-A reduc- extends the list to potential interactions between aspirin and
tase inhibitors, i.e. either simvastatin or atorvastatin (acting corticosteroids, also leading to the increased risk of peptic
by inhibiting CYP3A4, amiodarone increases the concen- ulcer, and to potential interactions between two antiplatelet
tration of the two statins and leads to increased risk of agents, resulting in increased risk of bleeding [74].
rhabdomyolysis); combination of selective serotonin reup- In total, 70 potentially clinically important drug–drug
take inhibitors with either tramadol or metoclopramide (both interactions were included in the new protocol.
combinations could result in high serotonin levels and life-
threatening serotonin syndrome); combination of clopidog-
rel with proton pump inhibitors (PPIs) (acting through Discussion and conclusion
CYP2C19 inhibition, PPIs decrease the formation of the
clopidogrel-active metabolite and reduce its antiplatelet ac- There are numerous strategies aimed at improving drug-
tivity) (Table 5) [74–78]. prescribing behavior to the elderly, which is the population
Hanlon’s adjunct to Malone’s list contains overall 34 phar- most sensitive to adverse effects of drugs. These strategies
macokinetic and nine pharmacodynamic drug–drug interac- include reducing the underuse, overuse, and misuse of
tions in the elderly population. Among the pharmacodynamic drugs, as well as reducing potentially important drug–drug
interactions, a potential interaction between corticosteroids interactions. The overuse of drugs and polypharmacy, lead-
and NSAIDs (increased risk of peptic ulcer disease) was ing to increased risk of ADRs and drug–drug interactions,
1136 Eur J Clin Pharmacol (2012) 68:1123–1138

can be avoided if only drugs with a clear indication, proven comprehensive quality assessment of drug prescribing in
efficacy, and favorable risk-to-benefit profile are used [5, the elderly and facilitate the detection of adverse outcomes
18]. The overall number of drugs should be the smallest caused by either PIMs or drug interactions. Both outcomes
possible and the duration of treatment the shortest possible. may in turn result in better prescribing practices. We expect
Potentially inappropriate medications represent one form the new tool to be used internationally by prescribers and
of drug misuse. Numerous authors have concluded that pharmacists in ambulatory and clinical settings as well as in
avoiding PIMs reduces drug-related morbidity and mortality nursing homes. Combining North American and European
in elderly patients [2]. The use of explicit screening tools, as tools together makes the tool applicable worldwide in the
the most simple and easiest method of detecting PIMs, is elderly population.
expected to improve the quality of prescribing [4]. There are The sensitivity of our tool in detecting PIMs and in
several tools currently available, each with a different de- assessing PIM- or drug interaction-related morbidity and
gree of comprehensiveness and complexity and some with hospital admissions should be tested against other criteria
more advantages than others. As there are marked differ- in the real life environment.
ences between the availability of drugs in different
countries, we recommend that each country adapt the most
acceptable screening tool. For European countries, adapta- References
tion of the French or German screening tools together with
Beers criteria may be advisable.
1. United Nations Population Division. World Population Ageing: 1950–
However, no one screening tool can substitute for the 2050. Available at:http://www.un.org/esa/population/publications/
prescriber’s thorough consideration of each individual el- worldageing19502050/. Accessed 1 Oct 2011
derly patient. Appropriate prescribing in the elderly also 2. Gallagher P, Barry P, O’Mahony D (2007) Inappropriate prescribing
includes all measures aimed at improving the patient’s ad- in the elderly. J Clin Pharm Ther 32(2):113–121
3. Routledge PA, O’Mahony MS, Woodhouse KW (2004) Adverse
herence to the therapy, including good communication be-
drug reactions in elderly patients. Br J Clin Pharmacol 57(2):121–
tween the prescriber and the patient, but also among all 126
physicians involved in providing healthcare to that patient 4. Hamilton HJ, Gallagher PF, O’Mahony D (2009) Inappropriate
[3–5, 18, 79]. Reducing the use of PIMs is just one of the prescribing and adverse drug events in older people. BMC Geriatr
9:5
strategies that can be adopted for better and safer prescribing
5. Laroche ML, Charmes JP, Bouthier F, Merle L (2009) Inappropriate
in the elderly, but all useful strategies should be imple- medications in the elderly. Clin Pharmacol Ther 85(1):94–97
mented together. 6. Beers MH, Ouslander JG, Rollingher I, Reuben DB, Brooks J,
The new protocol reported here, which combines the Beck JC (1991) Explicit criteria for determinig inappropriate
medication use in nursing home residents. Arch Intern Med
French, Beers, McLeod, and Lindblad criteria, represents a
151:1825–1832
well-balanced and comprehensive tool that may be widely 7. Beers MH (1997) Explicit criteria for determining potentially
applied in the analysis of drug prescribing in the elderly inappropriate medication use by the elderly: an update. Arch Intern
(defined as 65 years or older). It could be argued that the Med 57(14):1531–1536
8. McLeod PJ, Huang AR, Tamblyn RM, Gayton DC (1997) Defin-
present protocol contains drugs that have not been found to
ing inappropriate practice in prescribing for elderly people: a
be potentially inappropriate by some other experts (i.e., national consensus panel. Can Med Assoc J 156(3):385–391
amiodarone), but it follows both the Beers’ criteria and the 9. Naugler CT, Brymer C, Stolee P, Arcese ZA (2000) Development
French panel. Although the McLeod criteria were developed and validation of an improved prescribing in the elderly tool. Can J
Clin Pharmacol 7:103–107
back in 1997, the definition of inappropriate long-term 10. Fick DM, Cooper JW, Wade WE, Waller JL, Maclean JR, Beers
NSAIDs use along with hypertension, chronic renal, or heart MH (2003) Updating the Beers criteria. Arch Intern Med 63
failure in these criteria is still relevant and should be part of (22):2716–2724
screening tools for PIMs. 11. Lindblad CI, Hanlon JT, Gross CR, Sloane RJ, Pieper CF, Hajjar
ER et al (2006) Clinically important drug–disease interactions and
Table 5, which lists potentially clinically important drug–
their prevalence in older adults. Consensus statement. Clin Ther 28
drug interactions, may seem complex and difficult to follow, (8):1133–1143
but there are many potentially important drug–drug interac- 12. Laroche ML, Charmes JP, Merie L (2007) Potentially inappropri-
tions and while reducing their number would make the tool ate medications in the elderly. A French consensus panel list. Eur J
Clin Pharmacol 63(8):725–731
easier to follow, it would result in very important adverse 13. Barry PJ, Gallagher P, Ryan C, O’Mahony D (2007) START
drug combinations being ignored. (screening tool to alert doctors to the right treatment)—an
In conclusion, the new protocol is able to detect poten- evidence-based screening tool to detect prescribing omissions in
tially clinically important drug–drug interactions and PIMs elderly patients. Age Ageing 36(6):632–638
14. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D (2008)
in the elderly and suggests alternative therapeutic solutions. STOPP (Screening Tool of Older Person’s Prescriptions) and
As PIMs and drug interactions are often overlaping, analyz- START (Screening Tool to Alert doctors to Right Treatment).
ing them within the same protocol will provide more Consensus validation. Int J Clin Pharm Ther 46(2):72–83
Eur J Clin Pharmacol (2012) 68:1123–1138 1137

15. Rognstad S, Brekke M, Fetveit A, Spigset O, Wyller TB, Straand J 33. Bongue B, Laroche ML, Gutton S, Colvez A, Guéguen R, Moulin JJ
(2009) The Norwegian General Practice (NORGEP) criteria for et al (2011) Potentially inappropriate drug prescription in the elderly
assessing potentially inappropriate prescriptions to elderly patients. in France: a population-based study from the French National Insur-
A modified Delphi study. Scand J Prim Health Care 27(3):153–159 ance Healthcare system. Eur J Clin Pharmacol 67(12):1291–1299
16. Holt S, Schmiedl S, Thürmann PA (2010) Potentially inappropriate 34. Lindblad CI, Artz MB, Pieper CF, Sloane RJ, Hajjar ER, Ruby CM
medications in the elderly: the PRISCUS list. Dtsch Arzteblatt Int et al (2005) Potential drug-disease interactions in frail, hospitalized
107(31–32):543–551 elderly veterans. Ann Pharmacother 39(3):412–417
17. Hanlon JT, Schmader KE, Samsa GP, Weinberger M, Uttech KM, 35. Malone DC, Abarca J, Hansten PD, Grizzle AJ, Armstrong EP,
Lewis IK et al (1992) A method for assessing drug therapy appro- Van Bergen RC et al (2004) Identification of serious drug-drug
priateness. J Clin Epidemiol 45(10):1045–1051 interactions: results of partnership to prevent drug–drug interactions.
18. Page RL 2nd, Linnebur SA, Bryant LL, Ruscin JM (2010) Inap- J Am Pharm Assoc 44(2):142–151
propriate prescribing in the hospitalized elderly patient: defining 36. Hanlon JT, Schmader ES (2005) Editorial. Drug–drug interactions
the problem, evaluation tools, and possible solutions. Clin Interv in older adults—which ones matter ? Am J Geriatr Pharmacother 3
Aging 5:75–87 (2):61–63
19. Levy HB, Marcus EL, Christen C (2010) Beyond the beers criteria: 37. Becker LC (1978) Conditions for vasodilator-induced coronary
a comparative overview of explicit criteria. Ann Pharmacother 44 steal in experimental myocardial ischemia. Circulation 57
(12):1968–1975 (6):1103–1110
20. Gallagher P, O’Mahony D (2008) STOPP (Screening Tool of Older 38. Vonkeman HE, van de Laar MA (2010) Nonsteroidal anti-
Persons’ potentially inappropriate Prescriptions): application to inflammatory drugs: adverse effects and their prevention. Semin
acutely ill elderly patients and comparison with Beers’ criteria. Arthritis Rheum 39(4):294–312
Age Ageing 37(6):673–679 39. Trelle S, Reichenbach S, Wandel S, Hildebrand P, Tschannen B,
21. Pitkala KH, Strandberg TE, Tilvis RS (2002) Inappropriate drug Villiger PM et al (2010) Cardiovascular safety of non-steroidal
prescribing in home-dwelling, elderly patients: a population-based anti-inflammatory drugs: network meta-analysis. Br Med J 342:
survey. Arch Intern Med 162(15):1707–1712 c7086. doi:10.1136/bmj.c7086
22. Onder G, Landi F, Cesari M, Gambassi G, Carbonin P, Bernabei R, 40. Mintzer J, Burns A (2000) Anticholinergic side-effects of drugs in
Investigators of the GIFA Study (2003) Inappropriate medication elderly people. J R Soc Med 93(9):457–462
use among hospitalized older adults in Italy: results from the Italian 41. Kostis JB, Rosen RC, Holzer BC, Randolph C, Taska LS, Miller
Group of Pharmacoepidemiology in the Elderly. Eur J Clin Phar- MH (1990) CNS side effects of centrally-active antihypertensive
macol 59(2):157–162 agents: a prospective, placebo-controlled study of sleep, mood
23. van der Hooft CS, Jong GW, Dieleman JP, Verhamme KM, van der state, and cognitive and sexual function in hypertensive males.
Cammen TJ, Stricker BH et al (2005) Inappropriate drug prescrib- Psychopharmacology (Berl) 102(2):163–170
ing in older adults: the updated 2002 Beers criteria–a population- 42. Prichard BN, Graham BR (2000) I1 imidazoline agonists. General
based cohort study. Br J Clin Pharmacol 60(2):137–144 clinical pharmacology of imidazoline receptors: implications for
24. Fialová D, Topinková E, Gambassi G, Finne-Soveri H, Jónsson the treatment of the elderly. Drugs Aging 17(2):133–159
PV, Carpenter I, AdHOC Project Research Group et al (2005) 43. Einhorn PT, Davis BR, Massie BM, Cushman WC, Piller LB,
Potentially inappropriate medication use among elderly home care Simpson LM, ALLHAT Collaborative Research Group et al
patients in Europe. JAMA 293(11):1348–1358 (2007) The Antihypertensive and Lipid Lowering Treatment to
25. Bonk ME, Krown H, Matuszewski K, Oinonen M (2006) Poten- Prevent Heart Attack Trial (ALLHAT) Heart Failure Validation
tially inappropriate medications in hospitalized senior patients. Am Study: diagnosis and prognosis. Am Heart J 153(1):42–53
J Health Syst Pharm 63(12):1161–1165 44. Cipriani A, Barbui C, Brambilla P, Furukawa TA, Hotopf M, Geddes
26. Barry PJ, O’Keefe N, O’Connor KA, O’Mahony D (2006) Inap- JR (2006) Are all antidepressants really the same? The case of
propriate prescribing in the elderly: a comparison of the Beers fluoxetine: a systematic review. J Clin Psychiatry 67(6):850–864
criteria and the improved prescribing in the elderly tool (IPET) in 45. Bardy GH, Lee KL, Mark DB et al (2005) Amiodarone or an
acutely ill elderly hospitalized patients. J Clin Pharm Ther 31 implantable cardioverter-defibrillator for congestive heart failure.
(6):617–626 N Engl J Med 352:225–237
27. Fick DM, Waller JL, Maclean JR et al (2001) Potentially inappro- 46. Coceani M, Mariotti R (2006) Is amiodarone safe in heart failure?
priate medication use in a Medicare managed care population: Br Med J 332:317–318
association with higher costs and utilization. J Manag Care Pharm 47. Zimetbaum P (2007) Amiodarone for atrial fibrillation. N Engl J
7(5):407–413 Med 356(9):935–941
28. Fu AZ, Liu GG, Christensen DB (2004) Inappropriate medication 48. Foley P, Kalra P, Andrews N (2008) Amiodarone—avoid the
use and health outcomes in the elderly. J Am Geriatr Soc 52 danger of torsade de pointes. Resuscitation 76(1):137–141
(11):1934–1939 49. Wyse DG (2009) Pharmacotherapy for rhythm management in
29. Lau DT, Kasper JD, Potter DE, Lyles A, Bennett RG (2005) elderly patients with atrial fibrillation. J Interv Card Electrophysiol
Hospitalization and death associated with potentially inappropriate 25(1):25–29
medication prescriptions among elderly nursing home residents. 50. European Medicines Agency (2007) European Medicines Agency
Arch Intern Med 165(1):68–74 recommends restricted use for piroxicam. Available at: http://www.
30. Fu AZ, Jiang JZ, Reeves JH, Fincham JE, Liu GG, Perri M 3rd ema.europa.eu/docs/en_GB/document_library/Press_release/2009/
(2007) Potentially inappropriate medication use and healthcare 11/WC500012655.pdf. Accessed 1 Oct 2011
expenditures in the US community-dwelling elderly. Med Care 51. Gulmez SE, Droz-Perroteau C, Lassalle R, Blin P, Bégaud B,
45(5):472–476 Rossignol M et al (2011) Are traditional NSAIDs prescribed
31. Jano E, Aparasu RR (2007) Healthcare outcomes associated with appropriately among French elderly with osteoarthritis? Results
beers’ criteria: a systematic review. Ann Pharmacother 41(3):438– from the CADEUS cohort. Eur J Clin Pharmacol 67(8):833–838
447 52. Agency for Medicinal Products and Medical devices of Croatia.
32. Fick DM, Mion LC, Beers MH, Waller JL (2008) Health outcomes (2010) Annual report on drug consumption for 2009. Available
associated with potentially inappropriate medication use in older at http://www.almp.hr/?ln0hr&w0publikacije&d0promet_
adults. Res Nurs Health 31(1):42–51 lijekova_2009. Accessed 1 Oct 2011
1138 Eur J Clin Pharmacol (2012) 68:1123–1138

53. Veitch PC, Clifton-Bligh RJ (2004) Long-acting sulfonylureas— 68. Larsson V, Engedal K, Aarsland D, Wattmo C, Minthon L, Londos
long-acting hypoglycaemia. Med J Aust 180(2):84–85 E (2011) Quality of life and the effect of memantine in dementia with
54. Greco D, Angileri G (2004) Drug-induced severe hypoglycaemia lewy bodies and Parkinson’s disease dementia. Dement Geriatr Cogn
in Type 2 diabetic patients aged 80 years or older. Diabetes Nutr Disord 32(4):227–234
Metab 17(1):23–26 69. Wu CS, Ting TT, Wang SC, Chang IS, Lin KM (2011) Effect of
55. Spagnoli A, Tognoni G (1983) ‘Cerebroactive’ drugs. Clinical benzodiazepine discontinuation on dementia risk. Am J Geriatr
pharmacology and therapeutic role in cerebrovascular disorders. Psychiatry 19(2):151–159
Drugs 26(1):44–69 70. Pirmohamed M, James S, Meakin S, Green C, Scott AK, Walley
56. Frampton JE, Brogden RN (1995) Pentoxifylline (oxpentifylline). A TJ et al (2004) Adverse drug reactions as cause of admission to
review of its therapeutic efficacy in the management of peripheral hospital: prospective analysis of 18 820 patients. Br Med J 329
vascular and cerebrovascular disorders. Drugs Aging 7(6):480–503 (7456):15–19
57. Hoyer S (1995) Possibilities and limits of therapy of cognition 71. Huic M, Macolic V, Vrhovac B, Francetic I, Bakran I, Giljanovic S
disorders in the elderly. Z Gerontol Geriatr 28(6):457–462 (1994) Adverse drug reactions resulting in hospital admission. Int J
58. Teive HA, Troiano AR, Germiniani FM, Werneck LC (2004) Clin Pharmacol Ther 32(12):675–682
Flunarizine and cinnarizine-induced parkinsonism: a historical 72. Franceschi M, Scarcelli C, Niro V, Seripa D, Pazienza AM, Pepe G
and clinical analysis. Parkinsonism Relat Disord 10(4):243–245 et al (2008) Prevalence, clinical features and avoidability of ad-
59. No authors listed (2010) Haemorrhage due to Ginkgo biloba? verse drug reactions as cause of admission to a geriatric unit: a
Prescrire Int 17(93):19 prospective study of 1756 patients. Drug Saf 31(6):545–556
60. Electronic Medicines Compendium. Available at: http://www. 73. European Medicines Agency (2010) European Medicines Agency
medicines.org.uk/EMC/medicine/2571/SPC/Serc-8mg/ Accessed recommends suspension of marketing authorisations for sibutamine.
1 Oct 2011 Available at: http://www.ema.europa.eu/docs/en_GB/document_
61. Andrus MR, Holloway KP (2004) Clark DB (2004) Use of beta- library/Press_release/2010/01/WC500069995.pdf. Accessed 1 Oct
blockers in patients with COPD. Ann Pharmacother 38(1):142– 2011
145 74. Baxter K (ed) (2005) Stockley’s drug interactions, 7th edn. Phar-
62. Verdoux H, Lagnaoui R, Begaud B (2005) Is benzodiazepine use a maceutical Press, London
risk factor for cognitive decline and dementia? A literature review 75. Dvir Y, Smallwood P (2008) Serotonin syndrome: a complex but
of epidemiological studies. Psychol Med 35(3):307–315 easily avoidable condition. Gen Hosp Psychiatry 30(3):284–287
63. Kotlyar M, Dysken M, Adson DE (2005) Update on drug-induced 76. Marot A, Morelle J, Chouinard VA, Jadoul M, Lambert M,
depression in the elderly. Am J Geriatr Pharmacother 3(4):288–300 Demoulin N (2011) Concomitant use of simvastatin and amiodar-
64. Raivio MM, Laurila JV, Strandberg TE, Tilvis RS, Pitkälä KH (2007) one resulting in severe rhabdomyolysis: a case report and review
Neither atypical nor conventional antipsychotics increase mortality or of the literature. Acta Clin Belg 66(2):134–136
hospital admissions among elderly patients with dementia: a two- 77. Laine L, Hennekens C (2010) Proton pump inhibitor and
year prospective study. Am J Geriatr Psychiatry 15(5):416–424 clopidogrel interaction: fact or fiction? Am J Gastroenterol
65. Weiden PJ (2007) EPS profiles: the atypical antipsychotics are not 105(1):34–41
all the same. J Psychiatr Pract 13(1):13–24 78. Siller-Matula JM, Jilma B, Schrör K, Christ G, Huber K (2010)
66. Kamble P, Chen H, Sherer J, Aparasu RR (2008) Antipsychotic Effect of proton pump inhibitors on clinical outcome in patients
drug use among elderly nursing home residents in the United treated with clopidogrel: a systematic review and meta-analysis. J
States. Am J Geriatr Pharmacother 6(4):187–197 Thromb Haemost 8(12):2624–2641
67. Maher AR, Maglione M, Bagley S, Suttorp M, Hu JH, Ewing B et 79. Spinewine A, Schmader KE, Barber N, Hughes C, Lapane KL,
al (2011) Efficacy and comparative effectiveness of atypical anti- Swine C et al (2007) Appropriate prescribing in elderly people:
psychotic medications for off-label uses in adults: a systematic how well can it be measured and optimised? Lancet 370
review and meta-analysis. JAMA 306(12):1359–1369 (9582):173–184