European Journal of Clinical Pharmacology

https://doi.org/10.1007/s00228-018-2446-0

REVIEW

Potentially inappropriate medication in the elderly: a systematic review

of validated explicit criteria
Fabiane Raquel Motter 1,2 & Janaína Soder Fritzen 1 & Sarah Nicole Hilmer 2 & Érika Vieira Paniz 3 & Vera Maria Vieira Paniz 1

Received: 4 January 2018 / Accepted: 14 March 2018

# Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract
Purpose Potentially inappropriate medication (PIM) use causes preventable adverse drug reactions in older patients.
Several assessment tools have been published to identify and avoid PIM use. In this systematic literature review, we
aim to provide summaries and comparisons of validated PIMs lists published between 1991 and 2017 internationally.
Methods In accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Statement
(PRISMA), we performed a systematic review of articles describing the development and validation of criteria for
identification of PIMs among older people published between January 1991 and April 2017. The searches were
conducted on PUBMED, AgeLine, Academic Search, Academic Search Premier, and CINAHL. We identified the
most common medications/classes described as PIM. We also identified the drug–disease interactions and drug–drug
interactions reported among criteria.
Results From 2933 articles screened, 36 met our inclusion criteria. The majority used the Delphi method to validate their criteria.
We identified 907 different medications/classes, 536 different drug disease interactions involving 84 diseases/conditions, and 159
drug–drug interactions. Benzodiazepines and nonsteroidal anti-inflammatory drugs were the medications most commonly re-
ported as potentially inappropriate for older people.
Conclusion Although approaches aimed at detecting inappropriate prescribing have intensified in recent years, we
observed limited overlap between different PIM lists. Additionally, some PIM lists did not provide special consid-
erations of use and alternative therapies to avoid PIMs. These facts may compromise the use of PIM lists in clinical
practice. Future PIM lists should integrate information about alternative therapies and special considerations of use in
order to help clinicians in the drug prescription.

Keywords Inappropriate prescribing . Potentially inappropriate medication list . Drug-related side effects and adverse reactions .
Aged

Electronic supplementary material The online version of this article

(https://doi.org/10.1007/s00228-018-2446-0) contains supplementary
material, which is available to authorized users.

* Fabiane Raquel Motter 1

Postgraduate Program in Collective Health, University of Vale do
[email protected] Rio dos Sinos (UNISINOS), Av. Unisinos, 920, Sao
Leopoldo, RS 93022-000, Brazil
Janaína Soder Fritzen
[email protected] 2
Department of Clinical Pharmacology and Aged Care, Kolling
Sarah Nicole Hilmer Institute, Royal North Shore Hospital and University of Sydney,
[email protected] Reserve Road, St Leonards, NSW 2065, Australia

Érika Vieira Paniz 3

Faculty of Medical Sciences, Federal University of Rio Grande do
[email protected]
Sul (UFRGS), Rua Ramiro Barcelos, 2400,4º andar Faculdade de
Vera Maria Vieira Paniz Medicin, Porto Alegre, RS 90035-003, Brazil
[email protected]
 Eur J Clin Pharmacol

Introduction this review, we aim to summarize and compare the validated

potentially inappropriate medications lists for older people
As the complexity of pharmacotherapy has increased with published in different countries between 1991 and 2017.
increasing medication use, particularly among older adults Additionally, we aim to summarize the medications and
with multiple morbidities [1], medication risk management drug–disease and drug–drug interactions listed in the different
has become an increasingly important area of research. In this potentially inappropriate medications lists.
field, potentially inappropriate medication (PIM) is a term
used to describe the use of a medicine for which the associated
risks outweigh the potential benefits, especially when more Methods
effective alternatives are available [2]. PIM use is an important
public health challenge, with high prevalence rates (from 18 to This review was performed according to a standard protocol
> 40%) across a variety of healthcare settings [3–6]. Notably, for systematic reviews, which was based on the methodolog-
elderly patients are more likely to be exposed to PIMs because ical manuals of the Preferred Reporting Items for Systematic
they often deal with age-related pharmacokinetic and pharma- Reviews and Meta-Analyses (PRISMA). The PRISMA
codynamic changes, which can result in increased adverse checklist is available in Appendix 1.
drug reactions and decreased efficacy [7, 8]. Additionally,
older patients often suffer from multiple chronic- Search strategy
degenerative diseases and therefore use a higher number of
drugs, compared to other age groups [9]. In this population, We systematically identified studies published between
PIM use can lead to avoidable adverse drug events (ADEs) [3, January 1991 and April 2017 without any language restriction
10], including falls, fractures, and delirium and is associated in the PUBMED and Academic Search Complete via
with hospitalization [11–13] and mortality [14–16]. EBSCOhost, Academic Search Premier via EBSCOhost,
In recent years, many strategies and tools have been devel- AgeLine via EBSCOhost, and CINAHL via EBSCOhost elec-
oped to assess the appropriateness of medication use in older tronic databases. We selected this time limit for publication
people [2, 17–51]. Explicit criteria (EC) focusing on a single inclusion because the first instrument for PIM assessment was
medication/medication class can support improvements to the published in 1991 by Beers et al. [22]. The search included
quality of drug therapy and help to monitor drug therapy. terms related to older adults or appropriate/inappropriate med-
However, the development of evidence-based PIM lists spe- ication. Details of the full search strategy are included in
cifically for older populations is problematic, as older people Appendix 2.
are typically underrepresented or excluded from most efficacy
and safety trials [52, 53]. Accordingly, some investigators Eligibility criteria
have used a consensus technique that synthesizes accumulated
expert opinion to develop EC that would facilitate the formu- Original studies describing the EC used to determine poten-
lation of recommendations for suitable treatments in older tially inappropriate medications were considered eligible for
people [54]. This consensus technique could be used to deter- inclusion in this review if they involved individuals aged
mine which statements from the literature are applicable in 65 years and older and described the development and valida-
clinical practice [22]. tion of the methods used in the PIM list. Interventions and
Many different expert panels, including pharmacists, geri- observational studies that evaluated PIMs were also retained
atricians, and other health professionals, have developed lists if the abstract described potentially relevant PIM lists.
of EC to identify PIM use among older people in different We applied the following exclusion criteria: medication
countries [2, 17–51]. Previous systematic reviews identified review techniques using implicit criteria to evaluate PIMs
7–25 different PIM lists applied to older people [55–57]. and lists of PIMs restricted to specific therapeutic classes or
Although Luchetti et al. [57] summarized and described all specific diseases. Additionally, we excluded studies of PIMs
medications classified as PIMs using 14 validated and not validated by expert consensus and guidelines or recom-
nonvalidated PIM lists published between 2006 and 2015, mendations for the assessment of inappropriate prescriptions,
the literature lacks a comprehensive evaluation of the most as well as letters, editorials, and duplicate studies.
common drug–disease and drug–drug interactions described
in these validated PIM lists. Notably, summaries of the items Study selection
proven valid by many consensus panels may facilitate a trans-
lational comparison of the processes and provide information Duplicate manuscripts were removed after exporting the
about the most important PIMs in clinical practice, which search results to Endnote, version X6 (Clarivate Analytics,
would inform the development of interventions aimed at im- Philadelphia, PA, USA). Subsequently, two reviewers inde-
proving the prescription of specific medications. Therefore, in pendently screened the titles and abstracts of the remaining
Eur J Clin Pharmacol

manuscripts to identify potentially relevant studies describing excluded according to the exclusion criteria. A manual search
the development and validation of PIM lists. Additional stud- from the reference lists of the included articles produced two
ies were identified by a manual search of the citation lists for relevant publications not found in the previous systematic
studies that detailed potentially relevant PIM lists. Finally, database search. Thus, 36 articles were included in this sys-
full-text copies of studies that described either the validation tematic review [2, 17–51].
or use of any of the potentially relevant measures were re- Table 1 describes the characteristics of the PIM lists eval-
trieved and considered for inclusion in this review. If a deci- uated in this review. Most studies were conducted in Europe
sion could not be reached regarding the ability of a manuscript [2, 23, 25, 26, 28–30, 35, 36, 38, 39, 42, 43, 47, 49] and North
to meet the inclusion criteria, a decision was reached during America [17–19, 21, 22, 27, 32, 37, 41, 45, 46, 48, 51].
the following selection round. However, other countries from Asia [24, 31, 33, 34, 40, 50],
Oceania [20], and South America [44] such as Taiwan [24],
Data extraction and synthesis Pakistan [40], South Korea [33, 34], Thailand [50], Japan [31]
Australia [20], and Chile [44] have also published lists of
Two authors (FRM and JSF) independently extracted the data, PIMs.
after which the first author checked the completeness by The majority of the PIM lists (23 PIM list, 63.9%) are
reviewing the extraction tables generated by the second author aimed at the general population aged 65 years and older.
and checking the extracted data in the full-text articles. The Norwegian General Practice (NORGEP) criteria [47]
Disagreements were resolved by discussion between the two and its adaptation for nursing home residents were designed
authors; if no agreement could be reached, a third author was especially for individuals aged 70 years and older [42] and the
consulted (VMV). French criteria [36] for those aged 75 years and older. Only
The following data were extracted from the selected arti- three PIM lists (8.3%) were developed for nursing home res-
cles: country of origin, source of data used, and validation idents [22, 32, 42], two (5.5%) for older hospitalized patients
method (consensus technique, expert panel, literature based). [28, 40] and one (2.7%) for use in community pharmacies
We also extracted aspects evaluated in the lists of PIMs (med- [49].
ications, dosage, duration of therapy, duplication, drug–dis- Some PIM lists, such as Beers (1991, 1997, 2003, 2012 and
ease interactions, drug–drug interactions). We also analyzed 2015) [17, 18, 21, 22, 27], STOPP (Screening Tool of Older
the medication/medication class names and drug–disease in- People’s Prescriptions) version 1 [29] (2008) and 2 (2015)
teractions (medication or medication dosage or medication [43], FORTA (Fit fOR The Aged) [35], Australian
duration with consideration of diagnosis) and drug–drug in- Prescribing Indicators Tool (2012) [20], Thailand criteria
teractions reported in all PIM lists. We considered all medica- (2008) [50], and Lindblad criteria (2006) [37], used the cur-
tions belonging to a class as inappropriate if the authors de- rent literature on efficacy and safety in older adults as an
scribed concerns about the medication class and did not de- evidence base to develop their own list of PIMs. Other PIM
scribe single medications. However, the anticholinergic med- lists, such as the McLeod criteria (1997) [41], Rancourt
ication class exhibited considerable variation in terms of the criteria (2004) [46], French criteria (2007) [36], NORGEP
selection of specific drugs. Therefore, we included anticholin- criteria (2009) [47], and PRISCUS (2010) [30], combined
ergic drugs described in a recent review of the literature for the ECs previously published with a review of current literature.
EC that did not specifically state which medicines were con- However, most PIM lists used previously published PIM lists
sidered anticholinergic [58]. Additionally, we also considered to develop their lists of PIMs [2, 19, 23–26, 28, 31–34, 38–40,
a medication class as inappropriate when the authors de- 42, 44–45, 48–49, 51]. Twenty-one (58.3%) of the 36 PIM
scribed single medications and raised concerns related to the lists were based on the Beers criteria and its updates,
medication class. ten(27.8%) on the STOPP criteria and its update, and seven
The data were entered into Excel (Microsoft Corp., (19.4%) on the McLeod criteria. The tool developed by
Redmond, WA, USA), and all individual medications reported Tommelein et al. [49] was based on items derived from 14
in the studies were subsequently grouped into Anatomical, different PIM lists (Table 1).
Therapeutic and Chemical (ATC) classes (five levels). Of the 36 studies identified, 19 (52.8%) used the Delphi
method and 14 (38.9%) used a modified Delphi method, to
validate their ECs. Two studies used the RAND/UCLA [20,
Results 49] (Research and Development/University of California, Los
Angeles) process and the Italian criteria [38] used the Nominal
The search strategy produced 2933 potentially relevant publi- Group Technique. The number of experts ranged from 4 to 62
cations (Fig. 1). After screening titles and abstracts, we and approximately 50.0% of the consensus panels included
retained 248 potentially relevant publications according to between 10 and 20 respondents. We observed a predominance
the inclusion criteria. After a full-text review, 214 articles were of physicians and pharmacists whose practices concentrate on
 Eur J Clin Pharmacol

Fig. 1 PRISMA flow diagram of References retrieved from PUBMED,

Identification
search strategy results EBSCOhost N=2933

Duplicates excluded
(N=241)

2444 Studies excluded after

Screening
analysis of titles and abstracts

Studies included for analysis of full text

(N= 248)

Full -text excluded (N=214)

Eligibility
Interventions (N=29)
Observational Studies (N=103)
Implicit Criteria (N=30)
Non-Validated (N=36)
Editorials (N=10)
Manual search in reference lists Duplicates(N=2)
(N= 02) Specific disease/condition (N=4)
Included

Studies included
(N= 36)

older adults and clinical pharmacology among the experts. condition. Benzodiazepines (29/33 PIM lists, 87.9%) and an-
Some studies included experts from different specialties such tihistamines (23/33 PIM lists, 69.7%) were the most common
as psychiatrists [24, 30, 33, 38], cardiologists [24, 38], medication classes reported followed by tricyclic antidepres-
pulmonologists, gastroenterologists [24, 38], and urologists sants (19/33 PIM lists, 57.6%) (Table 3). The commonest
[24, 38] (Table 1). medications included were diazepam, chlordiazepoxide (31/
Classification systems for PIMs varied between the studies. 33 PIM lists, 93.9%), amitriptyline, and chlorpheniramine
The majority of PIM lists provide an explicit listing of indi- (28/33 PIM lists, 84.8%) (Table 4).
vidual drugs. Eleven (30.6%) tools focused on PIMs to avoid Medications that can be avoided in individual diseases/
in older adults independent of disease/condition; 22 (61.1%) conditions are specified in 22 PIM lists (61.1%). The most
included PIMs to avoid in older adults for specific diseases or common medication classes implicated were NSAIDs (20/22
conditions, and 20 (55.6%) mentioned relevant drug–drug in- PIM lists, 90.9%), tricyclic antidepressants (19/ 22 PIM lists,
teractions. Fourteen (38.8%) tools presented alternative thera- 86.4%), followed by urologic spasmolytics, and long-acting
pies and 10 (27.8%) provided information about special con- benzodiazepines (18/22 PIM lists, 81.8%) (Table 3).
siderations of use. Furthermore, 29 PIM lists (80.5%) also Oxybutynin, diazepam, and chlordiazepoxide were the most
described doses or durations of medications, which should common medications; they were reported as PIMs in specific
not be exceeded. Avoiding unnecessary duplication was men- diseases/conditions in 20 (90.9%) of 22 PIM lists (Table 4).
tioned in eight PIM lists (22.2%) (Table 2). We identified 536 different drug–disease interactions involv-
The 36 PIM lists identified a total of 907 different ing 84 diseases/conditions. Among them, only 38 (7.0%) drug–
medications/ medication classes. Among them, only 4 classes disease interactions were cited in more than 25% of PIM lists.
and 44 medications were reported by more than 69% of PIM The most common conditions cited were constipation/chronic
lists. The most prevalent class of medication identified as in- constipation (42/536 drug–disease interactions, 7.8%),
appropriate was benzodiazepines, which were included in 33 dementia/cognitive impairment (41/536 drug–disease interac-
(91.7%) of the 36 PIM lists. Other medication classes of PIMs tions, 7.6%), insomnia (36/536 drug–disease interactions,
identified included nonsteroidal anti-inflammatory drugs 6.7%), lower urinary tract symptoms/benign prostatic hyperpla-
(NSAIDs) (28/36 PIM lists; 77.8%) followed by tricyclic an- sia (28/536 drug–disease interactions, 5.2%), heart failure (19/
tidepressants and antihistamines (27/36 PIM lists, 75.0%) 536 drug–disease interactions, 3.5%), and history of falls/
(Table 3). Regarding the medications, only diazepam, chlor- fractures (19/536 drug–disease interactions,3.5%). Table 5 sum-
diazepoxide, indomethacin, and amitriptyline were considered marizes the most common drug–disease interactions identified.
inappropriate by 35 (97.2%) of the 36 PIM lists (Table 4). The use of NSAIDs in patients with renal insufficiency (15/22
Similar results were observed when we considered the 33 PIM lists, 68.1%) and heart failure (13/22 PIM lists, 59.1%)) and
PIM lists (91.7%) that evaluated PIM independent of disease/ the use of metoclopramide in patients with Parkinson’s disease
Table 1 Characteristics of potentially inappropriate medication (PIM) lists

Author List name Country Population Validation Number Characteristics of experts Structure Categories Based
method of experts

Beers et al. 1991 Beers criteria USAa Nursing home Delphi method 13 Expertise in 30 criteria 19 medications or Literature review
Eur J Clin Pharmacol

[22] residents aged (two-round) psychopharmacology, statements medication

≥ 65 years pharmacoepidemiology, classes to avoid
clinical geriatric in the elderly; 11
pharmacology, general criteria
clinical geriatrics, long-term describing doses,
care frequencies, or
durations that
should not be
exceeded.
Stuck et al. 1994 Stuck criteria USAa and Community-residing Modified 13 Geriatricians and pharmacists 27 criteria Beers criteria 1991
[48] Canada older persons Delphi statements
≥ 65 years method
(two-round)
Beers et al. 1997 Beers criteria USAa Persons aged Modified 6 Expertise in general geriatrics, 63 criteria 28 criteria medications/class inde-
[21] ≥ 65 years Delphi clinical pharmacology, statements describing pendent of diagnoses
method pharmacoepidemiology, classified potentially 35 criteria describing
(two-round) clinical pharmacy, as having high inappropriate potentially inappropriate
psychopharmacology severity or not medications/class to be
avoided in 15 diseases/-
conditions
Literature
review
McLeod et al. Mcleod criteria Canada Persons aged Modified 33 7 clinical pharmacologists, 9 38 criteria 18 contraindicated Beers criteria 1991 and
1997 [41] ≥ 65 years Delphi geriatricians, statements drugs in specific literature
method 8 family practitioners, 8 conditions review about drug–drug
(two-round) pharmacists 4 drug–drug interactions
interactions
16 drug–disease
interactions
Zhan et al. 2001 Zhan criteria USAa Community-dwelling Modified 7 5 geriatricians, 1 33 drugs 11 drugs that should Beers criteria 1997
[51] persons aged Delphi pharmacoepidemiologist, 1 always be
≥ 65 years method pharmacist avoided
(two-- 8 drugs that are
rounds) rarely
appropriate
14 drugs that have
some indications
for use in an
elderly
population
Fick et al. 2003 Beers–Fick USAa Persons aged Modified 12 Expertise in 68 criteria 48 criteria Beers criteria 1997 and
[27] criteria ≥ 65 years Delphi psychopharmacology, statements describing literature review
method pharmacoepidemiology, classified as potentially
(three-- clinical geriatric having high or inappropriate
round) pharmacology, and clinical low severity medications
geriatric medicine. 20 diseases or
conditions and
medications to
be avoided in
these
diseases/-
conditions
Rancourt et al. Rancourt criteria Canada Modified 4 111 criteria Beers criteria 1991 and 1997,
2004 [46] Delphi Mcleod criteria
Table 1 (continued)

Author List name Country Population Validation Number Characteristics of experts Structure Categories Based
method of experts

Persons aged method 1 general practitioner with a 39 potentially 1997, and literature review
≥ 65 years in (two-round) geriatric inappropriate of
long-term care practice, medication/ class pharmacoepidemiological
1 family physician (LB), 15 potentially studies
1 clinical pharmacist and inappropriate
1 pharmacoepidemiologist duration
20 potentially
inappropriate
dosage
37 drug–drug
interactions or
drug duplication
Pugh et al. 2006 HEDISb USAa Persons aged Modified – – 42 criteria 3 categories: Beers–Fick criteria 2003
[45] ≥ 65 years Delphi Always Avoid,
method Rarely
Appropriate, and
Some
Indications.
Lindblad et al. Lindblad criteria USAa Persons aged Modified 9 2 geriatricians and 7 28 clinically Literature review
2006 [37] ≥ 65 years Delphi pharmacists important
method drug–disease
(two-round) interactions
Laroche et al. French criteria France Persons aged Delphi method 15 5 geriatricians, 5 34 inappropriate 29 medications or Beers criteria 1991 and 1997;
2007 [36] ≥ 75 years (two-round) pharmacologists, 2 practices in medication Beers–Fick criteria 2003,
pharmacists, 2 general prescribing classes that McLeod criteria 1997; the
practitioners, 1 should be criteria adapted to French
pharmacoepidemiologist avoided practice (2001) and the
5 drug–disease guidelines of the French
interactions Medicine Agency on
medication prescribing in
the elderly.
Imai et al. 2008 Japanese beers Japan Persons aged Modified 9 Expertise in 47 medications or Beers–Fick criteria 2003
[31] criteria ≥ 65 years Delphi psychopharmacology, medication
method pharmacoepidemiology, classes that
(three-- clinical geriatric should be
round) pharmacology, and clinical generally avoid
geriatric medicine for all elderly
patients
Gallagher et al. STOPPc Ireland Persons aged Delphi method 18 9 teaching hospital consultants 65 practice Literature review
2008 [29] version1 ≥ 65 years (two-round) in geriatric statements
medicine, 3 clinical
pharmacologists, 1
old age psychiatric, 2 senior
academic
primary care physicians, 3
senior hospital pharmacists
with interest in geriatric
pharmacotherapy
Winit-Watjana Thailand criteria Thailand Persons aged Delphi method 17/16 Geriatricians, geriatric 77 practice 33 medications or Literature review
et al. 2008 ≥ 65 years (three-- medicine lecturers or statements medication
[50] round) physicians working in the classes with
geriatrics area potential adverse
reactions
32 drug–disease
interactions
Eur J Clin Pharmacol
Table 1 (continued)

Author List name Country Population Validation Number Characteristics of experts Structure Categories Based
method of experts

12 drug–drug
Eur J Clin Pharmacol

interactions
d
Rognstad et al. NORGEP Norway Persons aged Delphi method 57/47 14 clinical pharmacologists, 17 36 criteria for 21 criteria Beers criteria 1991 and 1997,
2009 [47] ≥ 70 years in (three-- geriatricians, 16 general pharmacologi- concerning Beers–Fick criteria 2003,
general practice round) practitioners cally single drugs and and Swedish
inappropriate dosages recommendations,
prescribing in 15 criteria Norwegian studies and
general practice concerning drug literature.
combinations to
be avoided
Kim et al. 2010 Korean criteria Korea Persons aged Delphi method 14 14 geriatric specialists, 57 potentially Beers criteria 1991, and
[33] ≥ 65 years (two-round including 7 inappropriate 1997, Beers–Fick criteria
family medicine specialists, drugs for the 2003, Zhan criteria 2001
3 psychiatrists, 1 elderly, and one
neurologist, and 3 clinical independent of phamacoepidemiological
pharmacists diagnosis study
93 potentially
inappropriate
drugs in 29
diagnoses
Maio et al. 2010 Italian criteria Italy Persons aged A nominal 9 3 general practitioners, 1 23 potentially Beers–Fick criteria 2003
[38] ≥ 65 years group geriatrician, 1 clinical inappropriate
(outpatients) technique pharmacist, 2 psychiatrists, drugs
1 cardiologist, 1 director of
long-term care facilities
Holt et al. 2010 Priscus Germany Persons aged Modified 25/26 Experts represented eight 83 potentially Beers criteria 1997, Beers-
[30] ≥ 65 years Delphi different specialties: inappropriate Fick criteria 2003,
method geriatric medicine, clinical medications Mcleod criteria 1997;
(two-- pharmacology, French criteria 2007
rounds) general practice, internal and literature review
medicine, pain
therapy, neurology,
psychiatry, and pharmacy.
Basger et al. Australian Australia Persons aged RND/UCLA 15/12 Geriatricians/pharmacologists, 41 criteria Literature review
2012 [20] Prescribing ≥ 65 years (two-round) clinical pharmacists,
Indicators disease management
Tool—APIT advisors to organizations
that produce Australian
evidence based therapeutic
publications.
Mann et al. 2012 Austrian criteria Austria Persons aged Delphi method 8 A general practitioner, a 73 drugs to avoid in Priscus 2010
[39] ≥ 65 years (two-round) specialist in neurology, three older patients
specialists in internal because of an
medicine, a psychiatrist, and unfavorable
two clinical pharmacists benefit/risk
working in hospital
pharmacies
American Beers criteria USAa Persons aged Modified 13 Expertise in geriatric medicine, 63 criteria 34 medications or Literature review
Geriatrics ≥ 65 years Delphi nursing, pharmacy practice, statements medication
Society, 2012 method research and quality classified as classes to avoid
[17] (two-round) measures having high or in the elderly
low severity 14 diseases and
conditions and
medications to
Table 1 (continued)

Author List name Country Population Validation Number Characteristics of experts Structure Categories Based
method of experts

be avoided in
these conditions
13 medications to
be used with
caution in older
adults
Bachyrycz et al. New Mexico USAa Persons aged Delphi method 12 Clinical pharmacists, 72 drugs to be used Beers–Fich 2003
2012 [19] criteria ≥ 65 years (two-round) geriatricians, nurses, with caution in
managed care specialists, the elderly
and consumers
Chang et al. Taiwan criteria Taiwan Persons aged Modified 21 Geriatricians, neurologists, 36 criteria 24 drug or drug Beers–Fick 2003; McLeod
2012 [24] ≥ 65 years Delphi psychiatrists, cardiologists, classes to be criteria 1997;
method pulmonologists, generally Rancourt 2004; French
(two-round) gastroenterologist, avoided in older criteria 2007,
urologists, and clinical adults STOPPc version1;
pharmacists irrespective of NORGEPd 2009;
comorbidities, Thailand criteria 2008
12 chronic
conditions with 6
drug or drug
classes that
patients with
these conditions
should avoid.
Castillo-Páramo Castillo-Páramo Spain Persons aged Delphi method 19 Expertise in geriatric medicine 65 criteria STOPPc version 1
et al. 2013 criteria ≥ 65 years in (two-round) and pharmacotherapy in
[23] primary care older people
Clyne et al. Clyne criteria Ireland Persons aged Delphi method 5 2 general practitioners 34 criteria Mcleod 1997 improved
2013 [25] ≥ 65 years in (two-round) 2 pharmacists prescribing in the elderly
primary care + focus 1 physician tool (IPET), Beers criteria
group 2012, Prescription
Peer Academic Detailing
(Rx-PAD) study,
Assessing Care of
Vulnerable Elders
(ACOVE), and STOPPc
version 1
Fialova et al. Czech national Czech Republic Persons aged Modified 15 Experts from the fields of 121 criteria 74 criteria for Explicit criteria published
2013 [26] criteria - CNC ≥ 65 years Delphi geriatrics, internal medicine, medications between 1997
method general practitioners, potentially and 2011
(three clinical pharmacy, and inappropriate in
rounds) clinical pharmacology old age
46 criteria for
drug–disease
interactions
Kunh-Thiel FORTAe Germany and Persons aged Delphi method 20 17 geriatric internists and 3 225 drugs ranged A: 55 drugs that are Literature review
et al. 2014 Austria and ≥ 65 years (two-round) geriatric psychiatrists from from A indispensable
[35] Switzerland Germany and Austria (indispensable) B: 60 drugs that are
to D (avoid) de- beneficial
pending on the C: 67 drugs that are
state of evidence questionable
for safety, effica- D: 43 drugs that
cy and overall should be
age-- avoided.
appropriateness
Eur J Clin Pharmacol
Table 1 (continued)

Author List name Country Population Validation Number Characteristics of experts Structure Categories Based
method of experts

Galán- Retamal Galán- Retamal Spain Patients hospitalized Delphi method – Pharmacists and general 50 criteria Beers criteria 2012, STOPPc
Eur J Clin Pharmacol

et al. 2014 criteria ≥ 65 years (two-round) practitioners version 1 and Priscus

[28]
American Beers criteria USAa Persons aged Modified 13 Expertise in geriatric medicine, 101 criteria 40 medications or diseases/conditions and
Geriatrics ≥ 65 years Delphi nursing, statements medication medications to be avoided
Society method pharmacy practice, research classified classes to avoid in these conditions
2015 [18] (two-round) and quality as having high or in the elderly 16 medications/class medica-
measures low severity 12 tions to be used with cau-
tion
13 drug–drug interactions
20 medications inappropriate
based on kidney function
Literature
review
Renon- Guiteras EU (7) PIM listf Europe Persons aged Delphi method 27 14 geriatricians, 3 pharmacists, 282 drugs/drug Priscus 2010, French criteria
et al. 2015 [2] ≥ 65 years (two-round) 7 clinical pharmacologists, class and prepa- 2007, STOPPc version 1,
and other 9 medical rations were Beers 1997 e 2012,
specialists classified as Beers–Fick, Mcleod 1997
medication inap- and Micromedex
propriate
Kim et al. 2015 Kim criteria South Korea Persons aged Delphi method 20 14 Physicians and 6 Twenty-six drug Beers 2012, STOPPc version
[34] ≥ 65 years (two-round) pharmacists who had ingredients 2, and Priscus 2010
experience in a wide range belonging to
of specialties across the seven drug
internal medicine, mental classes
health, neurology,
gerontology, preventive
medicine, urology, family
medicine, rheumatology,
and clinical pharmacy
fields, and who have been
working in general hospitals
or universities as clinical
practitioners and/or re-
searchers.
Nyborg et al. NORGEP— Norway Nursing home Delphi method 62/49 Specialists in geriatrics or 34 criteria 11 single substance NORGEPd 2009
2015 [42] NHg residents aged (three-- clinical pharmacology, criteria
≥ 70 years round) physicians in nursing homes 15 drug–drug
and experienced combination
pharmacists criteria
8 criteria where
regular
consideration of
describing
O’Mahony et al. STOPPc version Europe Persons aged ≥ 65 Delphi method 19 Expertise in geriatric medicine 81 criteria Literature review
2015 [43] 2 (two-round) and pharmacotherapy in
older people
Passi et al. 2016 Chilean criteria Chile Persons aged Delphi method – Pharmacologists and 265 medications 19 drugs that Beers–Fick criteria 2003
[44] ≥ 65 years geriatricians should always be
avoided
47 drugs that are
rarely
appropriate
49 drugs that have
some indications
Table 1 (continued)

Author List name Country Population Validation Number Characteristics of experts Structure Categories Based
method of experts

for use in an
elderly
population and
150 drugs that can
be used in
elderly
population
Tommelein et al. GheOP3Sh Europe Persons aged RAND/UCLA Part1:12 Part 1: 4 clinical pharmacists, 5 83 criteria 31 potentially 14 different criteria, 3 explicit
2016 [49] ≥ 65 years— (two-round) Part 2: 7 geriatricians, 2 general, 2 inappropriate lists of prescribing
community practitioners, 2 academics, 1 drugs indicators and one review
pharmacy community pharmacist and independent of study about
1 physician. diagnosis drug–drug interactions
Part 2: 7 community 11 potentially
pharmacists (N = 7) inappropriate
drugs dependent
on diagnosis
29 drug–drug
interactions
6 PPO
6 general
care-related
items to be ad-
dresses in the
community
pharmacy
Khodykov et al. Khodykov USAa Nursing home Delphi method 17/11 5 pharmacists, 5 nurses, 4 24 criteria 22 potentially STOPPc version 2
2017 [32] criteria residents aged (three-- researchers, and 3 inappropriate
≥ 70 years round) physicians medications
criteria
2 underused
medications
criteria
Mazhar et al. Mazhar criteria Pakistan Patients hospitalized Delphi method 12 Part 1: geriatricians, resident 32 criteria Beers criteria 2015 and
2017 [40] ≥ 65 years (two-round) doctors, clinical STOPPc version 2
pharmacists,
pharmacotherapy specialists
and academic
pharmacologists
Part 2: specialists in geriatric
medicine

a
United States of America
b
Healthcare Effectiveness Data and Information Set
c
Screening Tool of Older Person's Prescriptions
d
Norwegian General Practice criteria
e
Fit fOR The Aged list
f
European list of potentially inappropriate medications for older people
g
Norwegian General Practice–Nursing Home criteria
h
Ghent Older People's Prescriptions community Pharmacy Screening
Eur J Clin Pharmacol
Eur J Clin Pharmacol

Table 2 Summary of evaluated aspects in the potentially inappropriate medication (PIM) lists

Special considerations
Duration of therapy
Independent of

Disease-Drug

Alternatives
interactions
Drug- Drug

Duplication

Therapies s
interaction
diagnoses
List name Year Country

Dosage

of use
Beers criteria 1991 USAa
Stuck criteria 1994 USA/ Canada
Beers-Fick criteria 2003 USAa
McLeod criteria 1997 Canada
Beers criteria 1997 USAa
Zhan criteria 2001 USAa
Rancourt criteria 2004 Canada
Lindblad criteria 2006 USAa
HEDISb 2006 USAa
Japanese Beers criteria 2008 Japan
French criteria 2007 France
Thailand criteria 2008 Thailand
STOPPc version1 2008 Ireland
NORGEPd 2009 Norway
Italian Criteria 2010 Italy
Priscus 2010 Germany
Korean criteria 2010 Korea
Taiwan criteria 2012 Taiwan
Austrian Criteria 2012 Austria
Australian Prescribing Indicators Tool 2012 Australia
New Mexico criteria 2012 USAa
Beers criteria 2012 USAa
Czech National criteria 2013 Czech Republic
Clyne et al. 2013 Ireland
Castillo-Paramo criteria 2013 Spain
FORTAe 2009 Germany
Galan - Retamal criteria 2014 Spain
STOPP version 2 2015 Europe
EU(7) PIM listf 2015 Europe
NORGEP- NHg 2015 Norway
Kim criteria 2015 Korea
Beers criteria 2015 USAa
GheOP3Sh 2016 Europe
Passi et al. 2010 Chile
Mazhar criteria 2017 Pakistan
Khodyakov criteria 2017 USAa
a
United States of America
b
Healthcare Effectiveness Data and Information Set
c
Screening Tool of Older Person’s Prescriptions
d
Norwegian General Practice criteria
e
Fit fOR The Aged list
f
European list of potentially inappropriate medications for older people
g
Norwegian General Practice—Nursing Home criteria
h
Ghent Older People’s Prescriptions community Pharmacy Screening

(13/22 PIM lists, 59.1%) were the most commonly reported. medications, warfarin was the most common medication report-
Other prevalent drug–disease interactions included anticholiner- ed; it was included in 18(90.0%) of 20 PIM lists that evaluated
gic drugs in those with dementia/cognitive impairment (12/22 drug–drug interactions. Table 6 describes the most common
PIM lists, 54.5%), benzodiazepines in those with a history of drug–drug interactions described. The concomitant use of war-
falls/fractures (11/22 PIM lists, 50.0%), and urologic spasmo- farin with NSAIDs (11/20 PIM lists, 55.5%) and aspirin (7/20
lytics in those with lower urinary tract symptoms/benign prostatic PIM lists, 35.0%) was the most common drug–drug interactions
hyperplasia (10/22 PIM lists, 45.4%). reported followed by the concomitant use of NSAIDs and ACE
We also identified 159 potential drug–drug interactions de- inhibitors (7/20 PIM lists, 35.0%) and the concomitant use of
scribed in 20 PIM lists. Among them, only 16 (10.1%) drug– beta blockers and verapamil (7/20 PIM lists, 35.0%).
drug interactions were cited in more than 20% of PIM lists. The
most common medication classes implicated were the NSAIDs
(19/20 PIM lists, 95.0%), tricyclic antidepressants (14/20 PIM Discussion
lists, 70.0%), followed by angiotensin-converting-enzyme inhib-
itors (12/20 PIM lists, 55.6%), and selective serotonin reuptake This systematic review presents data from 36 PIM lists (pub-
inhibitors (10/20 PIM lists, 50.0%). Regarding single lished between 1991 and April 2017) that developed and
 Eur J Clin Pharmacol

Table 3 Most common medication classes reported in potentially inappropriate medication (PIM) lists

Medication class All PIMa lists, N PIMa lists independent of disease/ Drug–disease interactions, Drug–drug interactions,
(%) condition, N (%) N (%) N (%)

Total 36 (100.0) 33 (100.0) 22 (100.0) 20 (100.0)

Alimentary tract and metabolism
Proton pump inhibitors 10 (27.8) 03 (9.1) 06 (27.3) 01 (5.0)
Gastrointestinal antispasmodics 24 (66.7) 14 (42.4) 15 (68.2) 06 (27.3)
Blood and blood forming organs
Vitamin K antagonists 09 (25.0) 01 (3.0) 03 (13.6) 09 (45.0)
Direct thrombin inhibitors 12 (33.3) 04 (12.1) 03 (19.0) 07 (35.0)
Direct Factor Xa inhibitor 08 (22.2) 01 (3.0) 02 (9.1) 07 (35.0)
Cardiovascular system
Alpha-adrenoreceptor antagonists 20 (55.5) 07 (21.2) 13 (59.1) 03 (15.0)
Thiazides diuretics 15 (41.7) 01 (3.0) 08 (36.4) 07 (35.0)
Loop diuretics 13 (36.1) 01 (3.0) 05 (22.7) 09 (45.0)
Potassium sparing diuretics 12 (33.3) 01 (3.0) 01 (4.5) 11 (55.5)
Beta blockers 15 (41.7) 00 (0.0) 09 (40.9) 10 (50.0)
Beta blockers (only nonselective) 17 (47.2) 01 (3.0) 11 (50.0) 10 (50.0)
Calcium channel blocker 16 (44.4) 00 (0.0) 12 (54.5) 07 (35.0)
Angiotensin-converting-enzyme 15 (38.9) 01 (3.0) 06 (27.3) 12 (55.6)
inhibitor
Angiotensin II receptor blockers 12 (33.3) 00 (0.0) 06 (27.3) 07 (35.0)
Genitourinary system and sex hormones
Estrogens 16 (44.4) 11 (33.3) 07 (31.8) 00 (0.0)
Urologic spasmolytics 20 (55.6) 06 (16.7) 18 (81.8) 07 (35.0)
Systemic hormonal preparations
Corticosteroids 19 (52.8) 00 (0.0) 16 (72.7) 06 (30.0)
Musculo-skeletal system
Nonsteroidal anti-inflammatory 28 (77.8) 13 (39.4) 20 (90.9) 19 (95.0)
drugs
Muscle relaxants 12 (33.3) 05 (15.1) 07 (31.8) 05 (25.0)
Nervous system
Opioid 12 (33.3) 04 (12.1) 06 (27.3) 08 (40.0)
Conventional antipsychotics 24 (66.7) 13 (36.4) 15 (68.2) 09 (45.0)
Atypical antipsychotics 19 (52.8) 04 (12.1) 06 (27.3) 05 (25.0)
Benzodiazepines long-acting 33 (91.7) 29 (87.9) 18 (81.8) 05 (25.0)
Benzodiazepines short-acting 29 (80.6) 25 (75.8) 16 (72.7) 05 (25.0)
Barbiturates 18 (50.0) 14 (42.4) 08 (36.4) 03 (15.0)
Nonbenzodiazepine hypnotics 12 (33.3) 07 (18.2) 05 (22.7) 03 (15.0)
Tricyclic antidepressants 27 (75.0) 19 (57.6) 19 (86.4) 14 (70.0)
Selective serotonin reuptake 17 (47.2) 04 (12.1) 09 (40.9) 10 (50.0)
inhibitor
Respiratory system
Antihistamines 27 (75.0) 23 (69.7) 16 (72.3) 07 (35.0)
a
Potentially inappropriate medication

validated EC for identification of PIMs. The aggregation and number of medications/classes in our review is justified by the
comparison of studies showed a wide variability of PIMs, and fact that we included more years and other baseline data in our
we identified different 907 medications/medication classes re- search strategies, and we did not exclude PIM lists for institu-
ported in all PIM lists. A previous systematic review identified tionalized or hospitalized patients or criteria that reported only
729 different medications/classes described in 14 different drug–disease interactions. Furthermore, aspects such as differ-
PIM lists published between 2006 and 2015 [57]. The higher ent settings and prescribing cultures, differences in medication
Eur J Clin Pharmacol

Table 4 Most common medications reported in potentially inappropriate medication (PIM) lists

Medication All PIMa lists, N PIMa lists independent of disease/condition, N Drug–disease interactions, N Drug–drug interactions, N
(%) (%) (%) (%)

Total 36 (100.0) 33 (100.0) 22 (100.00) 20 (100.0)

Benzodiazepines
Diazepam 35 (97.2) 31 (93.9) 20 (90.9) 10 (50.0)
Chlordiazepoxide 35 (97.2) 31 (93.9) 20 (90.9) 10 (50.0)
Flurazepam 32 (88.9) 27 (81.1) 18 (81.8) 05 (25.0)
Clorazepate 31 (86.1) 24 (72.2) 17 (77.3) 04 (20.0)
Triazolam 29 (80.6) 21 (61.1) 19 (86.4) 10 (50.0)
Alprazolam 26 (72.2) 21 (61.1) 14 (63.6) 06 (30.0)
Oxazepam 25 (69.4) 20 (60.6) 14 (63.6) 04 (20.0)
Quazepam 25 (69.4) 18 (54.5) 14 (63.6) 10 (50.0)
Clonazepam 25 (69.4) 16 (48.5) 14 (66.7) 04 (20.0)
Nitrazepam 25 (69.4) 18 (54.5) 15 (68.2) 03 (16.7)
Nonsteroidal anti-inflammatory drugs
Indomethacin 35 (97.2) 23 (69.7) 20 (90.9) 19 (95.0)
Piroxicam 31 (86.1) 19 (57.6) 20 (90.9) 19 (95.0)
Naproxen 29 (80.6) 16 (48.5) 19 (86.4) 18 (90.0)
Phenylbutazone 29 (80.6) 16 (48.5) 18 (81.8) 18 (90.0)
Meloxican 29 (80.6) 12 (36.4) 20 (90.9) 19 (95.0)
Ketoprofen 29 (80.6) 12 (36.4) 20 (90.9) 19 (95.0)
Ketorolac 28 (77.8) 15 (45.5) 20 (90.9) 18 (90.0)
Diclofenac 27 (75.0) 10 (30.3) 19 (86.4) 18 (90.0)
Celecoxibe 27 (75.0) 09 (27.3) 19 (86.4) 18 (90.0)
Ibuprofen 26 (72.2) 10 (30.3) 19 (86.4) 18 (90.0)
Tricyclic antidepressants
Amitriptyline 35 (97.2) 28 (84.8) 20 (90.9) 14 (70.0)
Doxepin 28 (77.8) 22 (66.7) 18 (81.0) 13 (65.0)
Imipramine 26 (72.2) 17 (51.5) 19 (86.4) 14 (70.0)
Clomipramine 25 (69.4) 16 (48.5) 16 (72.7) 13 (65.0)
Antihistamines
Chlorpheniramine 31 (86.1) 28 (84.8) 18 (81.0) 07 (35.0)
Promethazine 28 (77.8) 25 (75.8) 17 (77.3) 08 (40.0)
Hydroxyzine 27 (75.0) 24 (72.2) 11 (52.4) 08 (40.0)
Diphenydramine 27 (75.0) 24 (72.2) 16 (72.7) 07 (35.0)
Cyproheptadine 25 (69.4) 23 (69.7) 16 (72.7) 07 (35.0)
Gastrointestinal antispasmodics
Hyoscyamine 26 (72.2) 16 (48.5) 15 (71.4) 07 (35.0)
Antipsychotics
Thioridazine 26 (72.2) 16 (48.5) 17 (77.3) 09 (45.0)
Haloperidol 26 (72.2) 12 (36.4) 16 (72.7) 09 (45.0)
Urologic spasmolytics
Oxybutynin 26 (72.2) 13 (39.4) 21 (95.5) 07 (35.0)

Selective serotonin reuptake inhibitor

Fluoxetine 26 (72.2) 14 (42.4) 13 (59.1) 14 (70.0)
Opioid
Pethidine/meperidine 26 (72.2) 15 (45.5) 12 (54.5) 13 (65.0)
Antiarrhythmic drugs
Digoxin 27 (75.0) 19 (57.6) 08 (36.4) 05 (25.0)
Antihypertensive
Nifedipine 27 (75.0) 14 (42.4) 12 (54.5) 07 (35.0)
(short-acting)
Methyldopa 25 (69.4) 21 (63.6) 05 (23.8) 02 (10.0)

a
Potentially inappropriate medication

availability/formulary between countries, and excluded from well-designed clinical trials [52, 53]. Thus, a
ethnopharmacology may have contributed to these results. majority of the studies used prior PIM lists to develop their
Conversely, we observed that less than half of PIM lists own lists of PIMs [2, 19, 23–26, 28, 31–34, 38–40, 42, 44–45,
developed their own EC based on literature reviews. The de- 48–49, 51]. However, some of these authors have combined
velopment of evidence-based PIM lists is a dynamic and com- different PIM lists with drug references [2],
plex process, because older participants are commonly pharmacoepidemiologic studies [25, 28, 33, 44], or
Table 5 Most common drug–disease interactions described in 22 potentially inappropriate medication (PIM) lists

Medication class/medication Beers McLeod Beers–Fick Lindbad French criteria Thailand criteria STOPPe version Korean criteria Beers APITf
1997 1997 2003 2006 2007 2008 1 2010 2012 2012

Renal insufficiency
NSAIDsa x x x x xc x
Heart failure
NSAIDsa x x x x x x
Parkinson disease/Parkinsonism
Metoclopramide x x x x x x
Prochlorperazine x x
Peptic ulcer
Aspirin x x x x x x x
Non COX-2b selective x x x x x
NSAIDs
NSAIDsa x x x
Cognitive impairment/dementia
Anticholinergics drugs x x x x x x
Tricyclic antidepressants x x x
Urologic spasmolytics x
Benzodiazepines (all) x x x
Constipation/chronic constipation
Anticholinergics drugs x x x x x x
Tricyclic antidepressants x x x x x x
Calcium channel blockers x x x
Urologic spasmolytics x x x
Opioids x x x
Lower urinary tract symptoms, benign prostatic hyperplasia
Urologic spasmolytics x x x
Anticholinergics drugs x x x x x
Tricyclic antidepressants x x x x
Falls
Benzodiazepines (all) x x x x
Antipsychotics x x x x
Tricyclic antidepressants x x x x x
Blooding disorders
Aspirin x x x x x x
Dipyridamole x x x x x
Clopidogrel x x x x
NSAIDsa x x x x x
Glaucoma
Tricyclic antidepressants x x x x
Gout
Thiazide x x x
Hypertension
NSAIDs x x x x
COPD
Corticosteroids x
Nonselective beta blocker x x x
Theophylline x
Eur J Clin Pharmacol
Table 5 (continued)

Medication class/medication Beers McLeod Beers–Fick Lindbad French criteria Thailand criteria STOPPe version Korean criteria Beers APITf
1997 1997 2003 2006 2007 2008 1 2010 2012 2012

Diabetes
Eur J Clin Pharmacol

Corticosteroids x x x x x
Insomnia
Decongestants x x x x
Methylphenidate x x x x
Theophylline x x x x

Medication class/ CNCg Taiwan criteria Clyne Castillo-Páramo Galán-Retamal FORTAh Kim Beers STOPPe GheOP3Si Mazhar Khodyakov
medication 2012 2012 2013 2013 2014 2014 2015 2015 version 2 2016 2017 2017

Renal insufficiency
NSAIDsa x x x x x xc x x x
Heart failure
NSAIDsa x x x x x x x
Parkinson disease/Parkinsonism
Metoclopramide x x x x x x x
Prochlorperazine x x x x
Peptic ulcer
Aspirin x x x x x x
Non COX-2b selective x x x x x
NSAIDs
NSAIDsa x x x x x
Cognitive impairment/dementia
Anticholinergics drugs x x x x x x
Tricyclic antidepressants x x x x x
Urologic spasmolytics x x x x x x
Benzodiazepines (all) x x x
Constipation/chronic constipation
Anticholinergics drugs x x x x
Tricyclic antidepressants x x x x
Calcium channel blockers x x x x
Urologic spasmolytics x x x x
Opioids x x x
Lower urinary tract symptoms, benign prostatic hyperplasia
Urologic spasmolytics x x x x x x x
Anticholinergics drugs x x x x
Tricyclic antidepressants x x x x x
Falls
Benzodiazepines (all) x x x x x x x
Antipsychotics x x x x x
Tricyclic antidepressants x x x
Blooding disorders
Aspirin x x x
Dipyridamole x x x
Clopidogrel x x
Table 5 (continued)

Medication class/ CNCg Taiwan criteria Clyne Castillo-Páramo Galán-Retamal FORTAh Kim Beers STOPPe GheOP3Si Mazhar Khodyakov
medication 2012 2012 2013 2013 2014 2014 2015 2015 version 2 2016 2017 2017

NSAIDsa x
Glaucoma
Tricyclic antidepressants xd x x x xd
Gout
Thiazide x x x x x x
Hypertension
NSAIDs x x x
COPD
Corticosteroids x x x x x x
Nonselective beta blocker x x x
Theophylline x x x x x
Diabetes
Corticosteroids x
Insomnia
Decongestants x x
Methylphenidate x x
Theophylline x x

a
Nonsteroidal anti-inflammatory drug
b
Cyclo-oxygenase
c
Chronic kidney disease stages IV and V
d
Narrow-glaucoma
e
Screening Tool of Older Person’s Prescriptions
f
Australian Prescribing Indicators Tool
g
Czech national criteria
h
Fit fOR The Aged list
i
Ghent Older People’s Prescriptions community Pharmacy Screening
Eur J Clin Pharmacol
Table 6 Most common drug–drug interactions described in 20 potentially inappropriate medication (PIM) lists

Drug–drug interactions Mcleod Rancourt French criteria STOPPe version Thailand criteria NORGEPf Korean criteria APITg CNCh Taiwan criteria
1997 2004 2007 2008 2008 2009 2010 2012 2012 2012

Warfarin–NSAIDSa x x x x x
Eur J Clin Pharmacol

Warfarin + aspirin x x x x
NSAIDs + ACEIb x x x
Beta blocker + verapamil x x x
NSAIDs + diuretic x x
ACEI + potassium sparing x x x x
diuretics
Anticholinergic + x
anticolinergic
NSAIDs + corticoids x
NSAIDs + anticoagulants x x x
Aspirin + anticoagulants x x
NSAIDs + antiplatelet agent x
NSAIDs + SSRIc x
Warfarin–cimetidine x x x x

TCAd + calcium channel x x

blocker
TCA + opioids x x
Aspirin + NSAIDS x x x

Drug–drug interactions Castillo-Páramo Clyne Galán- Retamal Beers Kim NORGEP—NHi STOPPe version GheOP3Sj Mazhar Khodyakov
2013 2013 2014 2015 2015 2015 2015 2016 2017 2017

Warfarin–NSAIDSa x x x x x x
Warfarin + aspirin x x x
NSAIDs + ACEIb x x x x
Beta blocker + verapamil x x x x
NSAIDs + diuretic x x x x
ACEI + potassium sparing x x
diuretics
Anticholinergic + anticolinergic x x x x
NSAIDs + corticoids x x x x
NSAIDs + anticoagulants x x
Aspirin + anticoagulants x x x
NSAIDs + antiplatelet agent x x x
NSAIDs + SSRIc x x x
Warfarin–cimetidine
x
x x
 Eur J Clin Pharmacol

prescribing indicators [28, 49] in order to include some med-

Khodyakov ications and other instances of PIM use in older people (e.g.,
2017 drug–drug interactions, drug class prescription duplication,
special considerations of use, alternative therapies) which

x
were not described in prior PIM lists.
We also verified that the majority of studies were devel-
Mazhar
2017

oped for general practice. Few PIM lists focused on specific

populations such as nursing home residents [22, 32, 42] and
hospitalized patients [28, 40]. These were adaptations from
GheOP3Sj

existing PIM lists and included some new PIMs in their eval-
uation. For instance, some lists did not account for drugs fre-
2016

quently used during inpatient stays such as antibiotics. Thus,

this result suggests that more work is needed to develop PIM
lists for these populations and that some PIM lists originally
STOPPe version

designed for general practice could later be externally validat-

ed in these settings.
2015

The Delphi technique was used to validate EC in the ma-

jority of the studies. This method is defined as an exercise in
group communication that brings together and synthesizes the
NORGEP—NHi

knowledge of a group of geographically distributed partici-

pants who have never meet [59]. Although there is no agree-
ment on the definition of an expert, number of experts used,
2015

the number of rounds, and the consensus level in the literature,

Delphi technique has some advantages over other consensus
techniques such as the lack of discussion domination by any
2015
Kim

one panel member [60]. However, in this review, some studies

x

[17, 18, 21, 27] modified the Delphi technique; these studies
used a physical panel meeting at the end of consensus proce-
Beers
2015

dure in order to exchange views and resolve uncertainties.

We found that benzodiazepines and NSAIDs were the most
common drugs reported as PIMs for older adults in all PIM
Galán- Retamal

lists. Previous systematic review also verified that these med-

ication classes are among the most common reported in PIM
lists [57]. However, these authors considered the number of
2014

indications of each medication class in each PIM lists while

Ghent Older People’s Prescriptions community Pharmacy Screening

we evaluated the medication class included in each PIM lists.

Of the 36 PIM lists evaluated, 33 described benzodiazepines
Clyne
2013

as inappropriate. There is good observational data on the as-

The Norwegian General Practice—Nursing Home criteria

sociation between the use of benzodiazepine by older adults

and serious ADEs, including impaired cognitive function [61,
Castillo-Páramo

Screening Tool of Older Person’s Prescriptions

62], delirium [63], respiratory insufficiency [64], falls [65],

and fall-related injuries such as hip fractures [66]. Thus, they
Angiotensin-converting enzyme inhibitors

The Norwegian General Practice criteria

have the potential to create serious public health problems

Australian Prescribing Indicators Tool
2013

Selective serotonin reuptake inhibitor

Nonsteroidal anti-inflammatory drug

including hospitalization and death. Despite these risks, ben-

x

zodiazepines are commonly used in the treatment of anxiety,

depression, and insomnia in older patients around the world.
TCAd + calcium channel

Tricyclic antidepressant

Patients and providers hesitate to discontinue benzodiazepines

Czech national criteria
Drug–drug interactions

because of the fear of withdrawal symptoms or relapse [67].

Table 6 (continued)

Aspirin + NSAIDS

Studies show that there is a high prevalence of long-term use

TCA + opioids

of this class in this age group, ranging 12 to 43% [68, 69].

blocker

There was very limited overlap between the PIM lists that
we described in this study. Among all PIMs, only diazepam,
chlordiazepoxide, indomethacin, and amitriptyline were
b

h
a

j
Eur J Clin Pharmacol

considered inappropriate by 35 of the 36 PIM lists. third party if necessary. This reduced the risk of studies being
Furthermore, only 44 medications and 4 medication classes omitted and also reduced the risk of selection bias.
were present in 69.0% or more of PIM lists. Prior systematic
reviews also reported that only a few drugs are common to all Limitations
the lists of PIMs published [57]. The heterogeneity in the lists
of medications reflects the fact that medication management Our review had some important limitations. EC are limited in
in older adults is extremely complex with a very limited evi- that they do not address individual differences among patients
dence base to guide it. Additionally, health professionals from or the complexity or appropriateness of entire medication reg-
various fields were involved in the development of the PIM imens. Furthermore, they need to be regularly updated in line
lists and they would, therefore, have different approaches and with the evidence, and country-specific adaptations are neces-
attitudes. As a consequence, the list of medications can vary sary where countries differ in their guidelines, standards, and
widely. approved medications. It is important to recognize that a de-
We compiled all drug–disease interactions and drug–drug tailed description of the consensus method was not included in
interactions included in the different PIM lists. It is interesting some studies [26, 42, 43]. To our knowledge, there is no for-
to note that NSAIDs were the most common medication class mal method for quality assessment or risk of bias for consen-
in both types of drug interactions. Despite the consistent rec- sus studies, so a rigorous assessment of the quality/bias of
ommendations to avoid the use of this medication class in each study could not be performed as required by the
different situations, it is estimated that 40% of people aged PRISMA criteria [58].
65 years and older fill one or more prescriptions for a NSAIDs
each year [70] with additional users accessing NSAIDs over
the counter [71]. This, like the high utilization of benzodiaz- Conclusion
epines, may highlight the limited impact of the consensus on
PIMs or that, while potentially inappropriate, the benefit may Appropriate mediation management among older adults can
frequently be determined to outweigh the risk for the help prevent serious adverse drug events [3, 10] which are
individual. associated with the increase of hospitalization and mortality
We identified the drug–drug interactions described in 20 in this population. For this reason, approaches aimed at de-
PIM lists. Although a considerable proportion of adverse drug tecting inappropriate prescriptions have intensified in the last
reactions is caused by interactions between drugs [72, 73], decades with the development and validation of a number of
drug–drug interactions are still underreported in the criteria strategies, particularly PIM lists. These PIM lists are important
for assessing inappropriate prescriptions in older adults. Of educational tools and should be included in the comprehen-
the 159 drug–drug interactions identified, only 16 are de- sive assessment of every older patient who requires medica-
scribed in more than 20% of the PIM lists. The concomitant tion. We identified 36 different PIM lists. Different
use of NSAIDs and aspirin with warfarin was the most fre- medication/medication classes, drug–disease interactions,
quent drug–drug interaction described. Many studies have and drug–drug interactions were included in different lists,
provided an increased risk of hospitalization in elderly adults with limited overlap between the PIM lists presented. These
using this combination of drugs [72]. Additionally, the warfa- results demonstrate that the use of medications in older people
rin was the most common single medication reported among is complex field and that more evidence is required to be able
the drug–drug interactions lists. Despite this medication is to generate consistent expert recommendations and to imple-
highly effective in the prevention of stroke in atrial fibrillation, ment them.
it is known for its interaction with many drugs [72–73], which Our review highlights the most common PIMs, drug–dis-
is the leading cause of adverse drug event-related hospitaliza- ease interactions, and drug–drug interactions validated by ex-
tions in older adults and can lead to fatal outcomes in this pert consensus for over 26 years. These results can help health
population [74]. professionals to elaborate strategies to minimize use of PIMS
in many different settings. Although benzodiazepines and
Strengths NSAIDs were the most common medications classified as
being inappropriate, they are still commonly used in older
This is the first study that systematically compiled all drug– adults. Avoiding medication in which the risks outweigh the
disease interactions and drug–drug interactions included in benefits in the elderly patient continues to be a challenge for
validated PIM lists since 1991. This systematic review used health professionals. Some PIM lists are complex and did not
a comprehensive search strategy applied by the reviewers provide special considerations of use and alternative medica-
without language limitations. Furthermore, the study followed tions to avoid those considered potentially inappropriate. In
the PRISMA methodology, including study selection per- addition, few PIM lists provide information that supports safe-
formed by two independent reviewers with arbitration by a ly tapering or withdrawing PIM. These facts may compromise
 Eur J Clin Pharmacol

the use of PIM lists in clinical practice. Future PIM lists should prescriptions to elderly patients in the primary care setting: a sys-
tematic review. PLoS One 7(8):e43617. https://doi.org/10.1371/
integrate information about alternative therapies and special
journal.pone.0043617
considerations of use in order to help clinicians to make deci- 7. Corsonello A, Pedone C, Incalzi RA (2010) Age-related pharma-
sions about drug prescription. cokinetic and pharmacodynamic changes and related risk of ad-
verse drug reactions. Curr Med Chem 17(6):571–584. https://doi.
Acknowledgements We thank to the Coordination for the Improvement org/10.2174/092986710790416326
of Higher Education Personnel, National Council for Scientific and 8. Mangoni AA, Jackson SH (2004) Age-related changes in pharma-
Technological Development for the support that they are providing for cokinetics and pharmacodynamics: basic principles and practical
development of this study. applications. Br J Clin Pharmacol 57(1):6–14. https://doi.org/10.
1046/j.1365-2125.2003.02007.x
9. Hanlon JT, Shimp LA, Semla TP (2000) Recent advances in geri-
Author contributions FRM and VPM participated in all stages of this
atrics: drug-related problems in the elderly. Ann Pharmacother
project, from the design and interpretation of data to its final writing.
34(3):360–365. https://doi.org/10.1345/aph.19140
FRM and JSF conducted the development of search strategies, selection
10. Lund BC, Carnahan RM, Egge JA, Chrischilles EA, Kaboli PJ
procedure, data extraction, data synthesis, and analysis. EVP contributed
(2010) Inappropriate prescribing predicts adverse drug events in
to the database organization and data extraction. SNH contributed to the
older adults. Ann Pharmacother 44(6):957–963. https://doi.org/10.
critical review and writing of this manuscript. All authors participated in
1345/aph.1M657A
the discussions, result interpretation, and approved the final version of
11. Cabre M, Elias L, Garcia M, Palomera E, Serra-Prat M (2017)
manuscript for submission.
Avoidable hospitalizations due to adverse drug reactions in an acute
geriatric unit. Analysis of 3,292 patients. Med Clin. https://doi.org/
Funding FRM was supported by the Coordination for the Improvement 10.1016/j.medcli.2017.06.075
of Higher Education Personnel—CAPES through a doctorate at 12. Price SD, Holman CD, Sanfilippo FM, Emery JD (2014)
University of Vale do Rio dos Sinos, Brazil. FRM was also supported Association between potentially inappropriate medications from
by CAPES through a sandwich doctorate fellowship at University of the Beers criteria and the risk of unplanned hospitalization in elder-
Sydney, Australia (number grant: 88881.134589/2016-01). This system- ly patients. Ann Pharmacother 48(1):6–16. https://doi.org/10.1177/
atic review was funded by the National Council for Scientific and 1060028013504904
Technological Development-CNPQ (number grant: 426720/2016-4). 13. Reich O, Rosemann T, Rapold R, Blozik E, Senn O (2014)
The funders were not involved in the design or conduct of the study, Potentially inappropriate medication use in older patients in Swiss
collection, analysis, or interpretation of the data or preparation or approv- managed care plans: prevalence, determinants and association with
al of the manuscript. hospitalization. PLoS One 9(8):e105425. https://doi.org/10.1371/
journal.pone.0105425
Compliance with ethical standards 14. Klarin I, Wimo A, Fastbom J (2005) The association of inappropri-
ate drug use with hospitalisation and mortality: a population-based
study of the very old. Drugs Aging 22(1):69–82
Conflict of interest The authors declare that they have no conflict of
15. Lau DT, Kasper JD, Potter DE, Lyles A, Bennett RG (2005)
interest.
Hospitalization and death associated with potentially inappropriate
medication prescriptions among elderly nursing home residents.
Arch Intern Med 165(1):68–74. https://doi.org/10.1001/archinte.
References 165.1.68
16. Muhlack DC, Hoppe LK, Weberpals J, Brenner H, Schottker B
(2017) The association of potentially inappropriate medication at
1. Dimitrow MS, Airaksinen MS, Kivela SL, Lyles A, Leikola SN older age with cardiovascular events and overall mortality: a sys-
(2011) Comparison of prescribing criteria to evaluate the appropri- tematic review and meta-analysis of cohort studies. J Am Med Dir
ateness of drug treatment in individuals aged 65 and older: a sys- Assoc 18(3):211–220. https://doi.org/10.1016/j.jamda.2016.11.025
tematic review. J Am Geriatr Soc 59(8):1521–1530. https://doi.org/ 17. American Geriatrics Society Beers Criteria Update Expert P (2012)
10.1111/j.1532-5415.2011.03497.x American Geriatrics Society updated beers criteria for potentially
2. Renom-Guiteras A, Meyer G, Thurmann PA (2015) The EU(7)- inappropriate medication use in older adults. J Am Geriatr Soc
PIM list: a list of potentially inappropriate medications for older 60(4):616–631. https://doi.org/10.1111/j.1532-5415.2012.03923.x
people consented by experts from seven European countries. Eur 18. American Geriatrics Society Beers Criteria Update Expert P (2015)
J Clin Pharmacol 71(7):861–875. https://doi.org/10.1007/s00228- American Geriatrics Society 2015 updated Beers criteria for poten-
015-1860-9 tially inappropriate medication use in older adults. J Am Geriatr Soc
3. Hedna K, Hakkarainen KM, Gyllensten H, Jonsson AK, Petzold M, 63(11):2227–2246. https://doi.org/10.1111/jgs.13702
Hagg S (2015) Potentially inappropriate prescribing and adverse 19. Bachyrycz A, Dodd MA, Priloutskaya G (2012) Development and
drug reactions in the elderly: a population-based study. Eur J Clin dissemination of a statewide system to minimize use of potentially
Pharmacol 71(12):1525–1533. https://doi.org/10.1007/s00228- inappropriate medications (PIMs). Med Care 50(11):993–996.
015-1950-8 https://doi.org/10.1097/MLR.0b013e31826ecfdc
4. Morin L, Laroche ML, Texier G, Johnell K (2016) Prevalence of 20. Basger BJ, Chen TF, Moles RJ (2012) Validation of prescribing
potentially inappropriate medication use in older adults living in appropriateness criteria for older Australians using the RAND/
nursing homes: a systematic review. J Am Med Dir Assoc 17(9): UCLA appropriateness method. BMJ Open 2(5):e001431. https://
862.e861–862.e869. https://doi.org/10.1016/j.jamda.2016.06.011 doi.org/10.1136/bmjopen-2012-001431
5. Nyborg G, Straand J, Brekke M (2012) Inappropriate prescribing 21. Beers MH (1997) Explicit criteria for determining potentially inap-
for the elderly—a modern epidemic? Eur J Clin Pharmacol 68(7): propriate medication use by the elderly. An update. Arch Intern
1085–1094. https://doi.org/10.1007/s00228-012-1223-8 Med 157(14):1531–1536
6. Opondo D, Eslami S, Visscher S, de Rooij SE, Verheij R, Korevaar 22. Beers MH, Ouslander JG, Rollingher I, Reuben DB, Brooks J, Beck
JC, Abu-Hanna A (2012) Inappropriateness of medication JC (1991) Explicit criteria for determining inappropriate medication
Eur J Clin Pharmacol

use in nursing home residents. UCLA Division of Geriatric 37. Lindblad CI, Hanlon JT, Gross CR, Sloane RJ, Pieper CF, Hajjar
Medicine. Arch Intern Med 151(9):1825–1832 ER, Ruby CM, Schmader KE (2006) Clinically important drug-
23. Castillo-Paramo A, Pardo-Lopo R, Gomez-Serranillos IR, Verdejo disease interactions and their prevalence in older adults. Clin Ther
A, Figueiras A, Claveria A (2013) Assessment of the appropriate- 28(8):1133–1143. https://doi.org/10.1016/j.clinthera.2006.08.006
ness of STOPP/START criteria in primary health care in Spain by 38. Maio V, Del Canale S, Abouzaid S, Investigators GAP (2010)
the RAND method. SEMERGEN 39(8):413–420. https://doi.org/ Using explicit criteria to evaluate the quality of prescribing in el-
10.1016/j.semerg.2013.01.017 derly Italian outpatients: a cohort study. J Clin Pharm Ther 35(2):
24. Chang CB, Yang SY, Lai HY, Wu RS, Liu HC, Hsu HY, Hwang SJ, 219–229. https://doi.org/10.1111/j.1365-2710.2009.01094.x
Chan DC (2012) Using published criteria to develop a list of poten- 39. Mann E, Bohmdorfer B, Fruhwald T, Roller-Wirnsberger RE,
tially inappropriate medications for elderly patients in Taiwan. Dovjak P, Duckelmann-Hofer C, Fischer P, Rabady S, Iglseder B
Pharmacoepidemiol Drug Saf 21(12):1269–1279. https://doi.org/ (2012) Potentially inappropriate medication in geriatric patients: the
10.1002/pds.3274 Austrian consensus panel list. Wien Klin Wochenschr 124(5–6):
25. Clyne B, Bradley MC, Hughes CM, Clear D, McDonnell R, 160–169. https://doi.org/10.1007/s00508-011-0061-5
Williams D, Fahey T, Smith SM, O-Ss t (2013) Addressing poten- 40. Mazhar F, Akram S, Malhi SM, Haider N (2017) A prevalence
tially inappropriate prescribing in older patients: development and study of potentially inappropriate medications use in hospitalized
pilot study of an intervention in primary care (the OPTI-SCRIPT Pakistani elderly. Aging Clin Exp Res 30:53–60. https://doi.org/10.
study). BMC Health Serv Res 13:307. https://doi.org/10.1186/ 1007/s40520-017-0742-7
1472-6963-13-307 41. McLeod PJ, Huang AR, Tamblyn RM, Gayton DC (1997) Defining
26. Fialova DT, Topinkova E, Ballokova A, Matejovska-Kubesova H inappropriate practices in prescribing for elderly people: a national
(2013) 2012 CZ expert consensus for potentially inappropriate consensus panel. CMAJ 156(3):385–391
medication use in old age: appropriate choice of drugs and drug 42. Nyborg G, Straand J, Klovning A, Brekke M (2015) The
dosing in geriatric patients (section I), drug-disease interactions in Norwegian general practice—nursing home criteria (NORGEP-
the old age (Section II). Klin Farmakol Farmacie 27(1):18–28 NH) for potentially inappropriate medication use: a web-based
27. Fick DM, Cooper JW, Wade WE, Waller JL, Maclean JR, Beers Delphi study. Scand J Prim Health Care 33(2):134–141. https://
MH (2003) Updating the Beers criteria for potentially inappropriate doi.org/10.3109/02813432.2015.1041833
medication use in older adults: results of a US consensus panel of 43. O’Mahony D, O’Sullivan D, Byrne S, O’Connor MN, Ryan C,
experts. Arch Intern Med 163(22):2716–2724. https://doi.org/10. Gallagher P (2015) STOPP/START criteria for potentially inappro-
1001/archinte.163.22.2716 priate prescribing in older people: version 2. Age Ageing 44(2):
28. Galan Retamal C, Garrido Fernandez R, Fernandez Espinola S, 213–218. https://doi.org/10.1093/ageing/afu145
Ruiz Serrato A, Garcia Ordonez MA, Padilla Marin V (2014) 44. Passi A, Margozzini P, Valenzuela E, Hoyl T, Marin PP, Carrasco
Prevalence of potentially inappropriate medication in hospitalized M, Olea R, Gac H (2016) Inappropriate medication use among
elderly patients by using explicit criteria. Farm Hosp 38(4):305– Chilean older people. Rev Med Chil 144(4):417–425. https://doi.
316. https://doi.org/10.7399/fh.2014.38.4.1148 org/10.4067/S0034-98872016000400001
29. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D (2008) 45. Pugh MJ, Hanlon JT, Zeber JE, Bierman A, Cornell J, Berlowitz
STOPP (Screening Tool of Older Person's Prescriptions) and DR (2006) Assessing potentially inappropriate prescribing in the
START (Screening Tool to Alert doctors to Right Treatment). elderly Veterans Affairs population using the HEDIS 2006 quality
Consensus validation. Int J Clin Pharmacol Ther 46(2):72–83 measure. J Manag Care Pharm 12(7):537–545. https://doi.org/10.
30. Holt S, Schmiedl S, Thurmann PA (2010) Potentially inappropriate 18553/jmcp.2006.12.7.537
medications in the elderly: the PRISCUS list. Dtsch Arztebl Int 46. Rancourt C, Moisan J, Baillargeon L, Verreault R, Laurin D,
107(31–32):543–551. https://doi.org/10.3238/arztebl.2010.0543 Gregoire JP (2004) Potentially inappropriate prescriptions for older
31. Imai HB, MH FDM (2008) Development of Japanese Beers criteria patients in long-term care. BMC Geriatr 4:9. https://doi.org/10.
for inappropriate medication use in elderly patients. Jpn Med Assoc 1186/1471-2318-4-9
J 137:1348–1358 47. Rognstad S, Brekke M, Fetveit A, Spigset O, Wyller TB, Straand J
32. Khodyakov D, Ochoa A, Olivieri-Mui BL, Bouwmeester C, (2009) The Norwegian General Practice (NORGEP) criteria for
Zarowitz BJ, Patel M, Ching D, Briesacher B (2017) Screening assessing potentially inappropriate prescriptions to elderly patients.
Tool of Older Person's Prescriptions/Screening Tools to Alert A modified Delphi study. Scand J Prim Health Care 27(3):153–159.
Doctors to Right Treatment medication criteria modified for U.S. https://doi.org/10.1080/02813430902992215
nursing home setting. J Am Geriatr Soc 65(3):586–591. https://doi. 48. Stuck AE, Beers MH, Steiner A, Aronow HU, Rubenstein LZ,
org/10.1111/jgs.14689 Beck JC (1994) Inappropriate medication use in community-
33. Kim DS, Heo SI, Lee SH (2010) Development of a list of poten- residing older persons. Arch Intern Med 154(19):2195–2200
tially inappropriate drugs for the korean elderly using the Delphi 49. Tommelein E, Mehuys E, Petrovic M, Somers A, Van Damme C,
method. Healthc Inf Res 16(4):231–252. https://doi.org/10.4258/ Pattyn E, Mattelin K, Boussery K (2016) Potentially inappropriate
hir.2010.16.4.231 prescribing in nursing home residents detected with the community
34. Kim S-O, Jang S, Kim C-M, Kim Y-R, Sohn HS (2015) Consensus pharmacist specific GheOP(3)S-tool. Int J Clin Pharm 38(5):1063–
validated list of potentially inappropriate medication for the elderly 1068. https://doi.org/10.1007/s11096-016-0366-6
and their prevalence in South Korea. Int J Gerontol 9(3):136–141. 50. Winit-Watjana W, Sakulrat P, Kespichayawattana J (2008) Criteria
https://doi.org/10.1016/j.ijge.2015.05.013 for high-risk medication use in Thai older patients. Arch Gerontol
35. Kuhn-Thiel AM, Weiss C, Wehling M, Faep m (2014) Consensus Geriatr 47(1):35–51. https://doi.org/10.1016/j.archger.2007.06.006
validation of the FORTA (Fit fOR The Aged) List: a clinical tool for 51. Zhan C, Sangl J, Bierman AS, Miller MR, Friedman B, Wickizer
increasing the appropriateness of pharmacotherapy in the elderly. SW, Meyer GS (2001) Potentially inappropriate medication use in
Drugs Aging 31(2):131–140. https://doi.org/10.1007/s40266-013- the community-dwelling elderly: findings from the 1996 Medical
0146-0 Expenditure Panel Survey. JAMA 286(22):2823–2829
36. Laroche ML, Charmes JP, Merle L (2007) Potentially inappropriate 52. Cherubini A, Oristrell J, Pla X, Ruggiero C, Ferretti R, Diestre G,
medications in the elderly: a French consensus panel list. Eur J Clin Clarfield AM, Crome P, Hertogh C, Lesauskaite V, Prada GI,
Pharmacol 63(8):725–731. https://doi.org/10.1007/s00228-007- Szczerbinska K, Topinkova E, Sinclair-Cohen J, Edbrooke D,
0324-2 Mills GH (2011) The persistent exclusion of older patients from
 Eur J Clin Pharmacol

ongoing clinical trials regarding heart failure. Arch Intern Med 64. Vozoris NT, Fischer HD, Wang X, Anderson GM, Bell CM,
171(6):550–556. https://doi.org/10.1001/archinternmed.2011.31 Gershon AS, Stephenson AL, Gill SS, Rochon PA (2013)
53. Crome P, Lally F, Cherubini A, Oristrell J, Beswick AD, Clarfield Benzodiazepine use among older adults with chronic obstructive
AM, Hertogh C, Lesauskaite V, Prada GI, Szczerbinska K, pulmonary disease: a population-based cohort study. Drugs Aging
Topinkova E, Sinclair-Cohen J, Edbrooke D, Mills G (2011) 30(3):183–192. https://doi.org/10.1007/s40266-013-0056-1
Exclusion of older people from clinical trials: professional views 65. Leipzig RM, Cumming RG, Tinetti ME (1999) Drugs and falls in
from nine European countries participating in the PREDICT study. older people: a systematic review and meta-analysis: II. Cardiac and
Drugs Aging 28(8):667–677. https://doi.org/10.2165/11591990- analgesic drugs. J Am Geriatr Soc 47(1):40–50
000000000-00000 66. Finkle WD, Der JS, Greenland S, Adams JL, Ridgeway G,
54. Marriott J, Stehlik P (2012) A critical analysis of the methods used Blaschke T, Wang Z, Dell RM, VanRiper KB (2011) Risk of frac-
to develop explicit clinical criteria for use in older people. Age tures requiring hospitalization after an initial prescription for
Ageing 41(4):441–450. https://doi.org/10.1093/ageing/afs064 zolpidem, alprazolam, lorazepam, or diazepam in older adults. J
55. Chang CB, Chan DC (2010) Comparison of published explicit Am Geriatr Soc 59(10):1883–1890. https://doi.org/10.1111/j.
criteria for potentially inappropriate medications in older adults. 1532-5415.2011.03591.x
Drugs Aging 27(12):947–957. https://doi.org/10.2165/11584850- 67. Tannenbaum C, Martin P, Tamblyn R, Benedetti A, Ahmed S
000000000-00000 (2014) Reduction of inappropriate benzodiazepine prescriptions
56. Kaufmann CP, Tremp R, Hersberger KE, Lampert ML (2014) among older adults through direct patient education: the
Inappropriate prescribing: a systematic overview of published as- EMPOWER cluster randomized trial. JAMA Intern Med 174(6):
sessment tools. Eur J Clin Pharmacol 70(1):1–11. https://doi.org/ 890–898. https://doi.org/10.1001/jamainternmed.2014.949
10.1007/s00228-013-1575-8 68. Cheng JS, Huang WF, Lin KM, Shih YT (2008) Characteristics
57. Lucchetti G, Lucchetti AL (2017) Inappropriate prescribing in older associated with benzodiazepine usage in elderly outpatients in
persons: a systematic review of medications available in different Taiwan. Int J Geriatr Psychiatry 23(6):618–624. https://doi.org/10.
criteria. Arch Gerontol Geriatr 68:55–61. https://doi.org/10.1016/j. 1002/gps.1950
archger.2016.09.003 69. Johnson CF, Frei C, Downes N, McTaggart SA, Akram G (2016)
58. Duran CE, Azermai M, Vander Stichele RH (2013) Systematic Benzodiazepine and z-hypnotic prescribing for older people in pri-
review of anticholinergic risk scales in older adults. Eur J Clin mary care: a cross-sectional population-based study. Br J Gen Pract
Pharmacol 69(7):1485–1496. https://doi.org/10.1007/s00228-013- 66(647):e410–e415. https://doi.org/10.3399/bjgp16X685213
1499-3 70. Ray WA, Stein CM, Byrd V, Shorr R, Pichert JW, Gideon P, Arnold
59. Dalkey NC (1969) The Delphi method: an experimental study of K, Brandt KD, Pincus T, Griffin MR (2001) Educational program
group opinion. Santa Monica, CA: RAND Corporation. Available for physicians to reduce use of non-steroidal anti-inflammatory
at: http://www.rand.org/pubs/research_memoranda/RM5888 drugs among community-dwelling elderly persons: a randomized
60. Boulkedid R, Abdoul H, Loustau M, Sibony O, Alberti C (2011) controlled trial. Med Care 39(5):425–435
Using and reporting the Delphi method for selecting healthcare 71. Marcum ZA, Hanlon JT (2010) Recognizing the risks of chronic
quality indicators: a systematic review. PLoS One 6(6):e20476. nonsteroidal anti-inflammatory drug use in older adults. Ann
https://doi.org/10.1371/journal.pone.0020476 Longterm Care 18(9):24–27
61. Islam MM, Iqbal U, Walther B, Atique S, Dubey NK, Nguyen PA, 72. Hines LE, Murphy JE (2011) Potentially harmful drug-drug inter-
Poly TN, Masud JH, Li YJ, Shabbir SA (2016) Benzodiazepine use actions in the elderly: a review. Am J Geriatr Pharmacother 9(6):
and risk of dementia in the elderly population: a systematic review 364–377. https://doi.org/10.1016/j.amjopharm.2011.10.004
and meta-analysis. Neuroepidemiology 47(3–4):181–191. https:// 73. Marengoni A, Pasina L, Concoreggi C, Martini G, Brognoli F,
doi.org/10.1159/000454881 Nobili A, Onder G, Bettoni D (2014) Understanding adverse drug
62. Zhong G, Wang Y, Zhang Y, Zhao Y (2015) Association between reactions in older adults through drug-drug interactions. Eur J Intern
benzodiazepine use and dementia: a meta-analysis. PLoS One Med 25(9):843–846. https://doi.org/10.1016/j.ejim.2014.10.001
10(5):e0127836. https://doi.org/10.1371/journal.pone.0127836 74. Budnitz DS, Lovegrove MC, Shehab N, Richards CL (2011)
63. Clegg A, Young JB (2011) Which medications to avoid in people at Emergency hospitalizations for adverse drug events in older
risk of delirium: a systematic review. Age Ageing 40(1):23–29. Americans. N Engl J Med 365(21):2002–2012. https://doi.org/10.
https://doi.org/10.1093/ageing/afq140 1056/NEJMsa1103053