European Geriatric Medicine (2021) 12:443–452

https://doi.org/10.1007/s41999-021-00479-3

REVIEW

Polypharmacy in older adults: a narrative review of definitions,

epidemiology and consequences
Farhad Pazan1 · Martin Wehling1

Received: 14 November 2020 / Accepted: 25 February 2021 / Published online: 10 March 2021
© The Author(s) 2021

Key summary points

Aim This narrative review aims to find and summarize recent publications on definitions, epidemiology and clinical con-
sequences of polypharmacy.
Findings Numerous definitions of polypharmacy and associated terms were found. The prevalence of polypharmacy greatly
varies, and numerous adverse clinical outcomes were associated with polypharmacy.
Message The clinically oriented definitions of polypharmacy are more useful and relevant. Regardless of the definition,
polypharmacy is highly prevalent in older adults. Approaches to increase the appropriateness of polypharmacy can improve
clinical outcomes in older adults.

Abstract
Background The number of older adults has been constantly growing around the globe. Consequently, multimorbidity and
related polypharmacy have become an increasing problem. In the absence of an accepted agreement on the definition of
polypharmacy, data on its prevalence in various studies are not easily comparable. Besides, the evidence on the potential
adverse clinical outcomes related to polypharmacy is limited though polypharmacy has been linked to numerous adverse
clinical outcomes. This narrative review aims to find and summarize recent publications on definitions, epidemiology and
clinical consequences of polypharmacy.
Methods The MEDLINE database was used to identify recent publications on the definition, prevalence and clinical conse-
quences of polypharmacy using their respective common terms and their variations. Systematic reviews and original studies
published between 2015 and 2020 were included.
Results One hundred and forty-three definitions of polypharmacy and associated terms were found. Most of them are
numerical definitions. Its prevalence ranges from 4% among community-dwelling older people to over 96.5% in hospitalized
patients. In addition, numerous adverse clinical outcomes were associated with polypharmacy.
Conclusion The term polypharmacy is imprecise, and its definition is yet subject to an ongoing debate. The clinically ori-
ented definitions of polypharmacy found in this review such as appropriate or necessary polypharmacy are more useful and
relevant. Regardless of the definition, polypharmacy is highly prevalent in older adults, particularly in nursing home residents
and hospitalized patients. Approaches to increase the appropriateness of polypharmacy can improve clinical outcomes in
older adults.

Keywords Polypharmacy · Multimorbidity · Older adults · Health outcomes · Inappropriate prescribing

Introduction

Multimorbidity is highly prevalent in our aging societies

* Martin Wehling and it often leads to the use of multiple medications in older
[email protected] patients [1–5]. There are also a number of other factors
1
which are associated with an increased use of drugs in older
Clinical Pharmacology Mannheim, Medical Faculty
Mannheim, Ruprecht-Karls-Heidelberg University,
people such as the wider and growing availability of effec-
Theodor‑Kutzer‑Ufer 1‑3, 68167 Mannheim, Germany tive medications in developed countries, and the numerous

13
Vol.:(0123456789)
444 European Geriatric Medicine (2021) 12:443–452

guidelines that encourage the use of multiple drugs for the and over: 80+ years. In addition, the period for our search
management of common diseases [4, 6, 7]. While there is was limited to the past 5 years because we aimed to find the
still no consensus on what should be regarded as polyphar- most recent literature which would add to the results from
macy, there are ongoing efforts to improve its definition the existing reviews also included in this narrative review.
and shift the focus from the mere number of drugs to their All papers found in MEDLINE were assessed for eligibility
appropriateness, effects and, ultimately, to related clinical by title and abstract screening. In addition to the database
outcomes in older patients [6, 8]. These efforts of course search, we performed manual searches from the reference
need to be backed by evidence. lists of some selected articles.
In this context, it is important to understand that the pro-
cess of biological aging is often accompanied by changes Definition of polypharmacy
in pharmacokinetics and pharmacodynamic in older peo-
ple [9]. Consequently, there should have been more efforts We found that there is no generally accepted definition for
towards clinical testing of medications specifically in older polypharmacy. This fact has also been acknowledged in a
adults. Regrettably, an exclusion of older adults from clini- recent report from the World Health Organization (WHO)
cal approval trials is often observed that has led to a lack of which stated that:
evidence regarding the safety and efficacy of many medica- “Polypharmacy is the concurrent use of multiple medica-
tions in this population [10, 11]. Thus, the appropriateness tions. Although there is no standard definition, polyphar-
of many drugs in an increasing number of older multimorbid macy is often defined as the routine use of five or more
patients remains still undetermined. Inevitably, this lack of medications. This includes over-the-counter, prescription
evidence has often led to inappropriate drug treatment and, and/or traditional and complementary medicines used by
consequently, to various adverse clinical outcomes. This a patient” [12].
large-scale issue was also recognized by the World Health Certainly, the concurrent intake of five or more drugs is
Organization and led to the initiation of the ongoing patient the most common definition of polypharmacy in the litera-
safety campaign called “The third Global Patient Safety ture [13]. Nevertheless, about 143 definitions of polyphar-
Challenge: Medication without harm” with “the goal of macy and associated terms exist according to a systematic
reducing severe, avoidable drug-related harm worldwide by review of definitions of polypharmacy [14] and other more
50%” until 2022 [12]. recent publications [7, 8, 11, 15, 16]. The vast majority of
Therefore, to differentiate between appropriate and inap- those (n = 112) are mere numerical definitions, meaning that
propriate drug treatment numerous listing tools/approaches only the number of drugs is used to check for the presence
(prescribing appropriateness assessment tools and criteria) of polypharmacy in an individual [14]. These numerical
have been proposed [4]. definitions are very heterogenous and partly include associ-
In this work, we searched for the existing definitions of ated terms such as minor [17–19], mild [20], moderate [21],
polypharmacy, its prevalence among older adults in various major [19, 22–25] and excessive polypharmacy [24, 26–29]
clinical settings and evidence for its clinical consequences to characterize the severity of polypharmacy [14]. Besides,
in geriatric patients. the threshold for these numerical definitions of polyphar-
macy ranges from 2 or more to 11 or more drugs [14] and
the cut-off/range for the associated terms partly varies as
Search strategy well. For instance, moderate polypharmacy is defined as the
daily intake of 4–5 medication in one source [14] and as the
This work represents a narrative review that provides an intake of 5–9 drugs in another [21].
overview of the recent publications on polypharmacy. It is In addition, there are 11 numerical definitions of polyp-
focused on definition, epidemiology, and outcomes of poly- harmacy which also classify the duration of drug treatment,
pharmacy in older adults. For this purpose, we used MED- about 4 numerical definitions which also include the health
LINE to search for related publications. The search terms care setting and 16 descriptive definitions of polypharmacy
were the following: (polypharmacy OR multiple medica- [8, 14–16, 30]. However, the few numerical definitions
tion* OR multiple medicine* OR multiple drug* OR Poly- including the health care setting can be regarded as only
pharmacy [Mesh] OR many medication* OR definition of numerical definitions which were used in a specific clinical
polypharmacy OR prevalence of polypharmacy or epidemi- setting. Categories [14] and examples for various defini-
ology of polypharmacy OR consequences of polypharmacy tions of polypharmacy and associated terms are depicted in
OR outcomes of polypharmacy). The following filters were Table 1. Examples for the more recent definitions of poly-
used to narrow our search: free full text, full text, review, pharmacy have been underlined in this table. Other new
systematic review, publication date: from November 2015 definitions of polypharmacy included unnecessary polyp-
to November 2020, Humans, English, Aged: 65+ years, 80 harmacy, and polypharmacy of unclear benefit [8].

13
 Table 1  Categories and examples for the existing definitions of polypharmacy and associated terms (examples of the more recent definitions of polypharmacy have been underlined)
Numerical only definitions Numerical definitions including a duration of Numerical definitions including a health care Descriptive definitions
European Geriatric Medicine (2021) 12:443–452

therapy setting

Five or more medications [27] Two or more medications for over 240 days [23] Five or more medications at hospital discharge Necessary polypharmacy: “Necessary polyphar-
[31] macy regimens should be considered additional
medications that can optimize functional status and
prevent disability in older adults. For older patients
receiving NP regimens, the benefits outweigh the
risks” [8]
Six or more medications [32] Over five medications for 90 or more days [33] Five to nine medications during hospital stay [34] Qualitative polypharmacy: “prescription of five
or more medications including at least one drug
considered potentially inappropriate for older
adults” [15]
Ten or more medications [14] Five to nine medications for 90 or more days [35] Ten or more medications during hospital stay (also Psychotropic polypharmacy: “the concurrent use of
called excessive polypharmacy) [34, 36] two or more psychotropic agents in one individual”
[16]
Seven or more medications [37] Five or more medications in the same quarter of a – Appropriate polypharmacy: “optimization of medi-
year [38] cations for patients with complex and/or multiple
conditions where medicine usage agrees with best
evidence” [14, 39]
Five to nine medications [40] Five or more medications in the same month [41] – “Use of medications which are not clinically indi-
cated” [14, 42]

13
445
446 European Geriatric Medicine (2021) 12:443–452

Epidemiology aged 85 years or older [43]. The overall prevalence of poly-

pharmacy was almost identical for women (32.1%; 95% CI
The prevalence of polypharmacy found in the literature 31.3–32.9) and men (32.2%; 95% CI 31.4–33.0) [43]. The
greatly varies. It ranges from around 4% to about 96.5% prevalence of polypharmacy across 17 European countries
depending on the age group, definition, healthcare setting and Israel according to Midão et al. [43] is depicted in Fig. 1.
and region [7, 11]. Based on the publications from the European Union
A cross-sectional analysis [43] from “wave 6” of the Sur- funded “Stimulating Innovation in the Management of
vey of Health, Ageing, and Retirement in Europe (SHARE) Polypharmacy and Adherence in the Elderly (SIMPATHY)
database including data from 34,232 participants (mean project” [44, 45] the prevalence of polypharmacy (defined as
age: 75.1 ± 7.2 years) revealed that the prevalence of poly- taking ten or more medications) using population-level data
pharmacy, defined as the concurrent use of five or more from “The Information Services Division (ISD), a division
medications, in older community-dwelling older adults of National Services Scotland” [46], ranged from approxi-
(aged 65 years or more) ranges from 26.3 to 39.9% across mately 6–36% in older adults. According to this data, depri-
17 European countries plus Israel [43]. It is noteworthy that vation had a profound impact on the prevalence of polyphar-
the medication use in this study was self-reported and a high macy in all age groups. For instance, in people aged 65–69
number of older adults with comorbidities might have been the prevalence of polypharmacy was about 24% in the most
excluded [43]. deprived vs. around 7% in the least deprived [45]. In people
The overall prevalence of polypharmacy across all aged 85 or older, the prevalence of polypharmacy for the two
countries was 32.1% (95% CI 31.5–32.7). Besides, the most deprived group of participants (around 33% in the most
lowest prevalence of polypharmacy was found in Switzer- deprived and about 30% in the second most deprived) was
land (26.3%; 95% CI 25.8–26.8), Croatia (27.3%; 95% CI unexpectedly lower as compared to the younger counterparts
26.8–27.9) and Slovenia (28.1%; 95% CI 27.6–28.6), while aged 80–84 (around 36% in the most deprived and about
Portugal (36.9%; 95% CI 36.3–37.5), Israel (37.5%; 95% 31% in the second most deprived) [44, 45].
CI 36.9–38.2) and the Czech Republic (39.9%; 95% CI The analysis of the population-level data from a pro-
39.3–40.5) had the highest prevalence of polypharmacy [43]. spective survey investigating the biological and genetic
The prevalence rates were also standardized by three age determinants of cardiovascular disease in the population of
groups: 65–74, 75–84 and 85 years or older. As anticipated, Lausanne (CoLaus study, Switzerland) revealed that 25.5%
in higher age groups the overall prevalence of polypharmacy of people aged from 65 to 81 years regularly used five or
was also greater. The overall prevalence for each age group more medications [47]. Another study based on claims data
was as follows: 25.3% (95% CI 24.6–26.0) for participants from the largest health insurance in Switzerland revealed
aged 65–74, 36.4% (95% CI 35.4–37.5) for participants that 41.2% of older adults concurrently received 5 or more
aged 75–44 and 46.5% (95% CI 44.6–48.4) for participants medications [48].

Fig. 1  Prevalence of poly-

pharmacy (defined as the
concurrent use of five or more
medications per day) in older
adults across 17 European
countries and Israel according
to “wave 6” of the Survey of
Health, Ageing, and Retire-
ment in Europe (SHARE) [43]
(Reprinted from Midão et al.
[43], Copyright (2018), with
permission from Elsevier.
https​://doi.org/10.1016/j.archg​
er.2018.06.018)

13
European Geriatric Medicine (2021) 12:443–452 447

In a repeated cross-sectional large-scale study (between drugs; long-term was defined as “for over 240 days in
338,025 and 539,752 older individuals) from Ireland a year” [23]) in only 4% of cases at the end of 1997.
using electronic data from pharmacy claims over 15 years Besides, at the end of the study 9% of the participants
(1997–2012), the prevalence of polypharmacy (five or more were prescribed 4–5 drugs (long-term) which was consid-
regular prescription medications) increased from 17.8% in ered as moderate polypharmacy and 28% of the patients
1997 to 60.4% in 2012. In addition, excessive polypharmacy 2–3 drugs (long-term) defined as minor polypharmacy in
(ten or more regular prescription medications) showed a this study [23]. Over the course of four years, about 20%
similar trend. It increased from 1.5 to 21.9% in people aged of older people developed polypharmacy in this popula-
65 years or older [49]. tion [23].
In a cross-sectional analysis of adult electronic primary According to another study based on the Korea Health
healthcare records from Scotland, the prevalence of polyp- Insurance Review and Assessment Service—National
harmacy (the use of 4–9 medications) was 28.6% in adults Patient Sample (HIRA-NPS) data from 2010 and 2011,
aged 60–69 years and 51.8% in those aged 80+ years [50]. polypharmacy (six or more medications) was found in
In this study, the term polypharmacy was not clearly stated. 86.4% of older people in Korea [24]. Major polypharmacy
Instead, the “higher” consumption of drugs was categorized (11 medications or more) and excessive polypharmacy (21
into either using 4–9 medications or ten or more. In addition, medications or more) was observed in 44.9% and 3% of
the prevalence of patients taking ten or more medications cases each [24].
was 7.4% in people aged 60–69 years and 18.6% in those A longitudinal observational study from Taiwan, which
aged 80+ years [50]. included frail Taiwanese older adults with long-term care
Another repeated cross-sectional analysis of community- needs also revealed a high prevalence of polypharmacy
dispensed prescribing data in the Tayside region of Scotland (prescribed with five or more drugs) in about 84% of cases
showed that the use of ten or more drugs in older adults [24]. However, persistent polypharmacy (polypharmacy
more than tripled from 4.9% in 1995 to 17.2% in 2010 [51]. for 181 days or more) was observed in only 31% of cases
In the same time period, the use of five or more medications [55]. Another study from Taiwan, which used Taiwan’s
was also increased by twofold from around 20% to around Longitudinal Health Insurance Database to assess drug use
40% among older adults aged 65–69 years old [51]. for older adults showed that polypharmacy (use of five or
In the U.S., population-level data from the National more drugs) was present in 28.2% of cases [56].
Health and Nutrition Examination Survey (NHANES) In the GLISTEN (Gruppo di Lavoro Italiano Sarcope-
revealed a significant increase in the prevalence of polyp- nia—Trattamento e Nutrizione; Italian working group on
harmacy (five or more prescription medications) from 24% sarcopenia—nutrition and treatment) study, the prevalence
(95% CI 21–26) to 39% (95% CI 35–44) among older adults of polypharmacy (using five or more medications daily) in
between 1999/2000 and 2011/2012 [52]. patients admitted to geriatric and internal medicine acute
In a large observational cohort study including 2057 older care ward of 12 Italian hospitals was 70.2%. In the same
emergency department patients from Ancona in Italy [53]. cohort, the prevalence of hyper-polypharmacy (using ten
The prevalence of polypharmacy (the concomitant use of or more medications daily) was 13% [57]. Another study
6–9 drugs in the last 3 months) was 30.3% and excessive in older hospitalized patients (aged 60 or over) from India
polypharmacy (the concomitant ten or more drugs in the showed that 45% of the patients presented with polyphar-
last 3 months) was observed in 17.8% of the patients [53]. macy (used 5–9 medications) and 45.5% with high-level
A large (N = 1,742,336) prospective, longitudinal, reg- polypharmacy (used ten or more drugs) [58]. Accord-
ister-based cohort study specifically analyzed the epidemi- ing to a cross-sectional study in older (aged 80 years or
ology of polypharmacy in older adults between 2010 and older) hospitalized patients from China, the prevalence of
2013 in Sweden [54]. This study showed that the preva- “hyper-polypharmacy” (11 or more drugs) was 96.5% [59].
lence of polypharmacy (5+ drugs) was 44% and 11.7% of In a cross-sectional analysis on nursing home residents
the individuals included had excessive polypharmacy (used participating in the “Services and Health for Elderly in
ten or more drugs) [54]. The incidence rate of newly devel- Long TERm care” (SHELTER) project the prevalence of
oped polypharmacy was 19.9 per 100 person-years, and the polypharmacy (5–9 drugs) and excessive polypharmacy
incidence rate of excessive polypharmacy was 8.0 per 100 (ten or more drugs) were 49.7% and 24.3% each [60].
person-years [54]. Other studies assessing medication use in nursing home
A longitudinal study from The Netherlands including residents reported a prevalence of polypharmacy (five or
data from elderly patients visiting general practices in more drugs) between 38.1 and 91.2% [61]. In addition, the
Groningen between 1994 and 1997 showed a prevalence prevalence of polypharmacy defined as the use of ten or
of “major polypharmacy” (long-term use of six or more more drugs ranged between 10.6 and 65% [61].

13
448 European Geriatric Medicine (2021) 12:443–452

Clinical consequences of polypharmacy issue, polypharmacy was not independently associated with
non-cancer mortality after adjustment for chronic condi-
Polypharmacy has been linked to numerous negative clinical tions and by using propensity score matching [62, 69, 70].
outcomes such as frailty, hospitalization and even mortality Conversely, in a recent nationwide longitudinal cohort study
[62, 63]. Here, we report the results from previous reviews from Korea, polypharmacy was associated with a higher
and recent studies on the association between polypharmacy risk of all-cause death, even after adjustment for comor-
and major clinical outcomes in older adults. bidities and propensity-score matching [71]. In addition, a
danish nationwide population-based cohort study showed
Polypharmacy and frailty an association between increasing number of medications
and mortality. In brief, for each extra medication the mortal-
While polypharmacy and frailty are highly prevalent in older ity increased by over 3% in the fully adjusted model which
adults the causal relationship between them is still unclear included diseases and Barthel Index [72].
[11]. According to a recent systematic review and meta-
analysis [64] which investigated the cross-sectional asso- Polypharmacy and hospitalization
ciation between polypharmacy/hyper-polypharmacy and the
presence of prefrailty/frailty as well as the risk of incident Several studies in community-dwelling older adults [73]
prefrailty/frailty in adults with polypharmacy, a strong and and nursing home residents [74] have shown an associa-
bidirectional association between both polypharmacy and tion between polypharmacy and hospitalization [11, 62].
hyper-polypharmacy and frailty was found [64]. This study The association has been shown for any hospitalization,
indicated that 75% of adults with polypharmacy are pre-frail/ unplanned hospitalization, and re-hospitalization in hospi-
frail. Besides, only a few longitudinal studies on the risk of tal-based samples [62]. For instance, a longitudinal health
incident prefrailty/frailty in adults with baseline polyphar- insurance database study from Taiwan investigating the
macy were found [64]. In a meta-analysis of three of those association between polypharmacy and all-cause and frac-
studies, a significantly higher odd of developing prefrailty ture-specific admission to hospital [56], showed an inde-
in robust persons was found in the presence of polyphar- pendent association between polypharmacy and all-cause
macy [64]. In a longitudinal study [65], taking seven or and fracture-specific hospitalization [56]. In another study
more medications was associated with a 2.5 increased risk of including nursing home residents from Australia, polyp-
developing frailty over 8 years [64]. This systematic review harmacy (defined as nine or more regular medications) was
concluded that more research is needed to assess the causal associated with time to the first hospitalization, number of
relationship between polypharmacy and frailty [64]. Another hospitalizations and hospital days per person-year [75]. Fur-
systematic review indicated that polypharmacy could play thermore, in a large observational cohort study in a geriatric
a major role in the development of frailty and stated that hospital, polypharmacy (having 6–9 drug prescriptions in
the causal relationship is uncertain and appears to be bidi- the last 3 months) and excessive polypharmacy (ten or more
rectional [66]. Besides, a cohort study of 772 Spanish older drug prescriptions in the last 3 months) were both associated
adults showed that polypharmacy is associated with death, with emergency department revisit and hospital admission
incident disability, hospitalization, and emergency depart- [53]. In the aforementioned nationwide longitudinal cohort
ment visits in frail and prefrail older adults, but not in robust study from Korea polypharmacy was associated with a
older people [67]. higher risk of hospitalization as well [71].

Polypharmacy and death Polypharmacy and falls

A systematic review and meta-analysis by Leelakanok Numerous studies have shown an association between polyp-
et al. revealed a significant association between mortality harmacy and falls [62, 76]. For example, a register-based study
and polypharmacy [68]. Besides, when polypharmacy was from Sweden confirmed the association between polyphar-
defined categorically, a dose–response relationship was macy and risk of falling [77]. However, an adjustment for fall-
seen across escalating thresholds (namely 1–4 medications, risk-inducing drugs (FRIDs) weakened this association [77].
five medications, 6–9 medications and ten or more medica- In a longitudinal study from England, the rate of falls was 21%
tions) [68]. It is important to mention that in patients with higher in older adults with polypharmacy (five or more drugs)
polypharmacy the risk of mortality may also be increased compared with people without polypharmacy [78]. Using a
by the presence of more chronic conditions. Hence, con- lower threshold of four or more drugs for polypharmacy, the
founding by indication has to be addressed when assessing rate of falls was 18% higher in people with polypharmacy
the relationship between polypharmacy and mortality [62]. compared with people without [78]. Hyper-polypharmacy
For instance, in a study by Schöttker et al. addressing this (ten or more drugs) was associated with a 50% higher rate of

13
European Geriatric Medicine (2021) 12:443–452 449

falls [78]. In another study, the number of fall-risk medications studies [7, 11, 14, 90]. The most common definition still is
was associated with fall-related hospital admissions though the concomitant use of five or more medications.
polypharmacy could not be identified as an independent risk While there was a wide range for the prevalence of poly-
factor [79]. Recently, a prospective cohort study also revealed pharmacy (4–96.5%), all studies showed an increase in the
that polypharmacy is associated with an increased falls risk in prevalence of polypharmacy in older adults over time and
UK care home residents [80]. most of them included community-dwelling older people.
In addition, in almost all studies the prevalence of polyp-
Polypharmacy and cognitive impairment harmacy directly correlated with age. Besides, the degree
of deprivation seems also to affect the prevalence of poly-
Polypharmacy and specially psychotropics and anticholinergic pharmacy in older adults [44, 45]. The highest prevalence
drugs have been associated with cognitive impairment in vari- of polypharmacy (over 96.5%) was observed in older hospi-
ous studies [62, 81]. talized patients in China [59]. Interestingly, the prevalence
For example, in a 12-year longitudinal register-based study of polypharmacy among community-dwelling older adults
from South Korea, polypharmacy was associated with the from Korea (86.4%) appeared to be higher than in any other
development of dementia [82]. Another longitudinal study in region of the world [24].
nursing home residents showed associations between polyp- The imprecise and heterogenous definition of polyphar-
harmacy and excessive polypharmacy and the decline in cog- macy complicates the analysis of its prevalence and impact
nitive function [83]. In Japan, a study in community-dwelling on relevant health outcomes. Generally, polypharmacy can
older adults also found an association between polypharmacy only serve as an indicator for adverse clinical outcomes, a
and cognitive impairment [81]. According to a cross-sectional causal relation with clinical outcomes has not been une-
study in patients with newly diagnosed Parkinson’s disease, quivocally proven as prospective interventional trials on its
those participants with polypharmacy had significantly lower clinical impact are largely missing. Association rather than
Mini-Mental State Examination scores as compared to other causation has been the most prevalent outcome from stud-
patients with no polypharmacy [84]. In another study investi- ies correlating clinical outcomes and polypharmacy. The
gating the associations between polypharmacy and cognitive descriptive definitions of polypharmacy found in this review
and physical capability, polypharmacy and longstanding poly- such as appropriate/necessary polypharmacy are potentially
pharmacy were associated with poorer cognitive capability; an more useful for clinical projections. The rational use of even
even stronger negative association was observed in individuals many drugs can be beneficial and sometimes would simply
with longstanding polypharmacy [85]. reflect the results from large randomized controlled trials,
thereby limiting the role of the mere number of drugs for
Polypharmacy and physical function clinical predictions. In other words, the quality of drug treat-
ment is crucial for a successful or beneficial therapy rather
Polypharmacy has been shown to be associated with a physi- than the number of drugs used in the same patient.
cal impairment in older adults [36, 85, 86]. A systematic Best approaches towards appropriate polypharmacy
review on the association between polypharmacy and physi- should address both over- as well as under-treatment in
cal function in older adults concluded that there is a strong older patients [4, 6]. To increase the appropriateness of
bidirectional relationship between polypharmacy and physi- drug therapy at least 73 listing tools/approaches have been
cal function [87]. However, a causal relationship based on developed. Most of them are designed to tackle the problem
the 18 observational studies included in that review could of over-treatment and polypharmacy and do not address the
not be proven [87]. The authors concluded that objective problem of undertreatment [4]. Hence, individualized and
measures of physical function and polypharmacy are nec- appropriate polypharmacy is not sufficiently addressed by
essary to prove this causal relationship in future studies most of these tools. In addition, only a minority of listing
[87]. Interestingly, in a multicenter study of European nurs- tools/approaches have been validated in randomized con-
ing homes polypharmacy was not associated with a faster trolled trials with relevant clinical outcomes, e.g. on physi-
decline in functional status [83]. Besides, disability has also cal function, hospitalization or death [4]. Of those, only
been associated with polypharmacy [67, 88]. the FORTA (Fit fOR The Aged) list [91] and the screening
tool of older people’s prescriptions (STOPP) and screening
tool to alert to right treatment (START) criteria [92] have
Conclusion shown a positive impact on relevant clinical endpoints in
randomized controlled trials [4].
The term polypharmacy is imprecise and its definition is yet The further improvement of drug treatment in older peo-
subject to an ongoing debate [89]. Our findings regarding the ple by interventional trials on such instruments is strongly
definition of polypharmacy are in line with several previous recommended; it should lead to a better understanding of

13
450 European Geriatric Medicine (2021) 12:443–452

polypharmacy and of the true clinical effects of medication 7. Guillot J, Maumus-Robert S, Bezin J (2020) Polypharmacy: a
optimization that are not necessarily linked to the mere num- general review of definitions, descriptions and determinants.
Therapies 75(5):407–416
ber of drugs. 8. Lee EA et al (2020) Refining the definition of polypharmacy and
its link to disability in older adults: conceptualizing necessary
polypharmacy, unnecessary polypharmacy, and polypharmacy of
Funding Open Access funding enabled and organized by Projekt unclear benefit. Perm J 24(18):212
DEAL. This research received no specific grant from any funding 9. Mangoni AA, Jackson SH (2004) Age-related changes in pharma-
agency in the public, commercial, or not-for-profit sectors. cokinetics and pharmacodynamics: basic principles and practical
applications. Br J Clin Pharmacol 57(1):6–14
10. Shenoy P, Harugeri A (2015) Elderly patients’ participation in
Declarations clinical trials. Perspect Clin Res 6(4):184–189
11. Khezrian M et al (2020) An overview of prevalence, determi-
Conflict of interest MW was employed by AstraZeneca R&D, nants and health outcomes of polypharmacy. Ther Adv Drug Saf
Mölndal, as director of discovery medicine (translational medicine) 11:1–10
from 2003 to 2006, while on sabbatical leave from his professor- 12. World Health Organization (2019) Medication Safety in Poly-
ship at the University of Heidelberg. Since returning to this position pharmacy (WHO/UHC/SDS/2019.11). https​://apps.who.int/iris/
in January 2007, he has received lecturing and consulting fees from bitst​ream/handl​e/10665​/32545​4/WHO-UHC-SDS-2019.11-eng.
Sanofi-Aventis, Bayer, Berlin-Chemie, Boehringer Ingelheim, Aspen, pdf?ua=1. Accessed 13 Aug 2019
Novartis, Takeda, Roche, Pfizer, Bristol-Myers, Daichii-Sankyo, Lilly, 13. Varghese D, Ishida C, Haseer Koya H (2020) Polypharmacy. Stat-
Otsuka, Novo-Nordisk, Shire and LEO Pharma. FP declares that he Pearls, Treasure Island
has no conflict of interest. 14. Masnoon N et al (2017) What is polypharmacy? A systematic
review of definitions. BMC Geriatr 17(1):230
Ethical approval Approval from a medical ethics committee was not 15. Gomes MS et al (2019) Polypharmacy in older patients at pri-
necessary for this narrative review. mary care units in Brazil. Int J Clin Pharm 41(2):516–524
16. McMahon M, Hatton C, Bowring DL (2020) Polypharmacy and
Informed consent For this narrative review formal consent is not psychotropic polypharmacy in adults with intellectual disability:
required. a cross-sectional total population study. J Intellect Disabil Res
64(12):834–851
17. Jorgensen TL et al (2012) Polypharmacy and drug use in elderly
Open Access This article is licensed under a Creative Commons Attri- Danish cancer patients during 1996 to 2006. J Geriatr Oncol
bution 4.0 International License, which permits use, sharing, adapta- 3(1):33–40
tion, distribution and reproduction in any medium or format, as long 18. Lu J et al (2014) Polypharmacy among nonagenarians/centenar-
as you give appropriate credit to the original author(s) and the source, ians in rural China. Intern Med J 44(12a):1193–1199
provide a link to the Creative Commons licence, and indicate if changes 19. Housley BC et al (2015) Comorbidity-polypharmacy score
were made. The images or other third party material in this article are predicts readmission in older trauma patients. J Surg Res
included in the article’s Creative Commons licence, unless indicated 199(1):237–243
otherwise in a credit line to the material. If material is not included in 20. Trevisan C et al (2019) Mild polypharmacy and MCI progres-
the article’s Creative Commons licence and your intended use is not sion in older adults: the mediation effect of drug-drug interac-
permitted by statutory regulation or exceeds the permitted use, you will tions. Aging Clin Exp Res 33(1):49–56
need to obtain permission directly from the copyright holder. To view a 21. Gallagher C et al (2020) Polypharmacy and health outcomes in
copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. atrial fibrillation: a systematic review and meta-analysis. Open
Heart 7(1):e001257
22. Ferreira AR et al (2015) P.5.f.007 Comorbidity and polyphar-
References macy in residential care and community-dwelling elderly in
Portugal. Eur Neuropsychopharmacol 25:S595
23. Veehof L et al (2000) The development of polypharmacy. A
1. United Nations, D.o.E.a.S.A., Population Division (2020) World longitudinal study. Fam Pract 17(3):261–267
Population Ageing 2019 (ST/ESA/SER.A/444). https​://www. 24. Kim HA et al (2014) Prevalence and predictors of polypharmacy
un.org/en/devel​opmen​t/desa/popul​ation​/publi​catio​ns/pdf/agein​ among Korean elderly. PLoS ONE 9(6):e98043
g/WorldP​ opula​ tionA
​ geing​ 2019-​ Report​ .pdf. Accessed 2 Nov 2020 25. Veehof LJG, Jong BMd, Haaijer-Ruskamp F (2000) Polyp-
2. Mc Namara KP et al (2017) Health professional perspectives on harmacy in the elderly -a literature review. Eur J Gen Pract
the management of multimorbidity and polypharmacy for older 6(3):98–106
patients in Australia. Age Ageing 46(2):291–299 26. Walckiers D, Van der Heyden J, Tafforeau J (2015) Factors associ-
3. Molokhia M, Majeed A (2017) Current and future perspectives ated with excessive polypharmacy in older people. Arch Public
on the management of polypharmacy. BMC Fam Pract 18(1):70 Health 73:50
4. Pazan F, Kather J, Wehling M (2019) A systematic review and 27. Diez-Manglano J et al (2015) Excessive polypharmacy and
novel classification of listing tools to improve medication in older survival in polypathological patients. Eur J Clin Pharmacol
people. Eur J Clin Pharmacol 75(5):619–625 71(6):733–739
5. Gutiérrez-Valencia M, Martínez-Velilla N, Vilches-Moraga A 28. Hovstadius B, Petersson G (2013) The impact of increasing poly-
(2019) Polypharmacy in older people: time to take action. Eur pharmacy on prescribed drug expenditure-a register-based study
Geriatr Med 10(1):1–3 in Sweden 2005–2009. Health Policy 109(2):166–174
6. Rankin A et al (2018) Interventions to improve the appropriate 29. Dorks M et al (2016) Polypharmacy and renal failure in nursing
use of polypharmacy for older people. Cochrane Database Syst home residents: results of the inappropriate medication in patients
Rev 9:CD008165 with renal insufficiency in nursing homes (IMREN) study. Drugs
Aging 33(1):45–51

13
European Geriatric Medicine (2021) 12:443–452 451

30. Kadam UT et al (2019) Conceptualizing multiple drug use in 50. Payne RA et al (2014) Prevalence of polypharmacy in a Scottish
patients with comorbidity and multimorbidity: proposal for stand- primary care population. Eur J Clin Pharmacol 70(5):575–581
ard definitions beyond the term polypharmacy. J Clin Epidemiol 51. Guthrie B et al (2015) The rising tide of polypharmacy and drug-
106:98–107 drug interactions: population database analysis 1995–2010. BMC
31. Nobili A et al (2011) Association between clusters of diseases Med 13:74
and polypharmacy in hospitalized elderly patients: results from 52. Kantor ED et al (2015) Trends in prescription drug use
the REPOSI study. Eur J Intern Med 22(6):597–602 among adults in the United States from 1999–2012. JAMA
32. Kuzuya M et al (2006) Underuse of medications for chronic dis- 314(17):1818–1831
eases in the oldest of community-dwelling older frail Japanese. J 53. Salvi F et al (2017) Is polypharmacy an independent risk factor
Am Geriatr Soc 54(4):598–605 for adverse outcomes after an emergency department visit? Intern
33. Narayan SW, Nishtala PS (2015) Associations of potentially Emerg Med 12(2):213–220
inappropriate medicine use with fall-related hospitalisations and 54. Morin L et al (2018) The epidemiology of polypharmacy in older
primary care visits in older new zealanders: a population-level adults: register-based prospective cohort study. Clin Epidemiol
study using the updated 2012 beers criteria. Drugs Real World 10:289–298
Outcomes 2(2):137–141 55. Chan DC, Hao YT, Wu SC (2009) Characteristics of outpatient
34. Vetrano DL et al (2014) Predictors of length of hospital stay prescriptions for frail Taiwanese elders with long-term care needs.
among older adults admitted to acute care wards: a multicentre Pharmacoepidemiol Drug Saf 18(4):327–334
observational study. Eur J Intern Med 25(1):56–62 56. Lu WH et al (2015) Effect of polypharmacy, potentially inap-
35. Nishtala PS, Salahudeen MS (2015) Temporal trends in polyp- propriate medications and anticholinergic burden on clinical out-
harmacy and hyperpolypharmacy in older new zealanders over a comes: a retrospective cohort study. CMAJ 187(4):E130–E137
9-year period: 2005–2013. Gerontology 61(3):195–202 57. Agosta L et al (2019) Polypharmacy and sarcopenia in hospital-
36. Sganga F et al (2014) Physical performance measures and polyp- ized older patients: results of the GLISTEN study. Aging Clin Exp
harmacy among hospitalized older adults: results from the CRIME Res 31(4):557–559
study. J Nutr Health Aging 18(6):616–621 58. Harugeri A et al (2010) Prescribing patterns and predictors of
37. Schuler J et al (2008) Polypharmacy and inappropriate prescrib- high-level polypharmacy in the elderly population: a prospective
ing in elderly internal-medicine patients in Austria. Wien Klin surveillance study from two teaching hospitals in India. Am J
Wochenschr 120(23–24):733–741 Geriatr Pharmacother 8(3):271–280
38. Kann IC, Lundqvist C, Luras H (2015) Polypharmacy among 59. Lai X et al (2018) Polypharmacy in the oldest old (≥80 years
the elderly in a list-patient system. Drugs Real World Outcomes of age) patients in China: a cross-sectional study. BMC Geriatr
2(3):193–198 18(1):64
39. Cadogan C et al (2015) Dispensing appropriate polypharmacy to 60. Onder G et al (2012) Polypharmacy in nursing home in
older people in primary care: a qualitative, theory-based study of Europe: results from the SHELTER study. J Gerontol Ser A
community pharmacists’ perceptions and experiences. IJPP Sup- 67A(6):698–704
plement 2:32–33 61. Jokanovic N et al (2015) Prevalence and factors associated with
40. Herr M et al (2015) Polypharmacy and frailty: prevalence, rela- polypharmacy in long-term care facilities: a systematic review. J
tionship, and impact on mortality in a French sample of 2350 old Am Med Dir Assoc 16(6):535 e1-612
people. Pharmacoepidemiol Drug Saf 24(6):637–646 62. Wastesson JW et al (2018) An update on the clinical consequences
41. Jiron M et al (2015) Prevalence and determinants of potentially of polypharmacy in older adults: a narrative review. Expert Opin
inappropriate medication prescribing among older US adults Drug Saf 17(12):1185–1196
according to STOPP criteria. Pharmacoepidemiol Drug Saf 63. Khezrian M et al (2019) The association between polypharmacy
24(S1):1–587 and late life deficits in cognitive, physical and emotional capabil-
42. Fulton MM, Allen ER (2005) Polypharmacy in the elderly: a lit- ity: a cohort study. Int J Clin Pharm 41(1):251–257
erature review. J Am Acad Nurse Pract 17(4):123–132 64. Palmer K et al (2019) Association of polypharmacy and hyper-
43. Midao L et al (2018) Polypharmacy prevalence among older polypharmacy with frailty states: a systematic review and meta-
adults based on the survey of health, ageing and retirement in analysis. Eur Geriatr Med 10(1):9–36
Europe. Arch Gerontol Geriatr 78:213–220 65. Veronese N et al (2017) Polypharmacy is associated with higher
44. Mair A, Wilson M, Dreischulte T (2020) Addressing the challenge frailty risk in older people: an 8-year longitudinal cohort study. J
of polypharmacy. Annu Rev Pharmacol Toxicol 60:661–681 Am Med Dir Assoc 18(7):624–628
45. Mair A et al (2017) Polypharmacy Management by 2030: a patient 66. Gutierrez-Valencia M et al (2018) The relationship between frailty
safety challenge. https​://ec.europ​a.eu/chafe​a/healt​h/newsr​oom/ and polypharmacy in older people: a systematic review. Br J Clin
news/docume​ nts/polyph​ armac​ y-handbo​ ok-second​ -editio​ n_en.pdf. Pharmacol 84(7):1432–1444
Accessed 8 Nov 2020 67. Bonaga B et al (2018) Frailty, polypharmacy, and health outcomes
46. Division T.I.S. (2017) National data from ISD Scotland. https​:// in older adults: the frailty and dependence in albacete study. J Am
www.isdsc​otlan​d.org. Accessed 8 Nov 2020 Med Dir Assoc 19(1):46–52
47. Castioni J et al (2017) Prevalence and determinants of polyphar- 68. Leelakanok N et al (2017) Association between polypharmacy and
macy in Switzerland: data from the CoLaus study. BMC Health death: a systematic review and meta-analysis. J Am Pharm Assoc
Serv Res 17(1):840 57(6):729-738.e10
48. Blozik E et al (2013) Polypharmacy and potentially inappropri- 69. Schottker B et al (2017) Polypharmacy and mortality: new insights
ate medication in the adult, community-dwelling population in from a large cohort of older adults by detection of effect modifi-
Switzerland. Drugs Aging 30(7):561–568 cation by multi-morbidity and comprehensive correction of con-
49. Moriarty F et al (2015) Trends and interaction of polypharmacy founding by indication. Eur J Clin Pharmacol 73(8):1041–1048
and potentially inappropriate prescribing in primary care over 70. Schottker B et al (2018) Updated analysis on polypharmacy
15 years in Ireland: a repeated cross-sectional study. BMJ Open and mortality from the ESTHER study. Eur J Clin Pharmacol
5(9):e008656 74(7):981–982

13
452 European Geriatric Medicine (2021) 12:443–452

71. Chang TI et al (2020) Polypharmacy, hospitalization, and mortal- longitudinal cohort database in South Korea. PLoS ONE
ity risk: a nationwide cohort study. Sci Rep 10(1):18964 12(1):e0169463
72. Brockhattingen KK et al (2020) Association between number 83. Vetrano DL et al (2018) Association of polypharmacy with 1-year
of medications and mortality in geriatric inpatients: a Dan- trajectories of cognitive and physical function in nursing home
ish nationwide register-based cohort study. Eur Geriatr Med residents: results from a multicenter european study. J Am Med
11(6):1063–1071 Dir Assoc 19(8):710–713
73. Beer C et al (2011) Quality use of medicines and health outcomes 84. Ishii N et al (2019) Polypharmacy associated with cognitive
among a cohort of community dwelling older men: an observa- decline in newly diagnosed Parkinson’s disease: a cross-sectional
tional study. Br J Clin Pharmacol 71(4):592–599 study. Dement Geriatr Cogn Disord Extra 9(3):338–343
74. Lalic S et al (2016) Polypharmacy and medication regimen com- 85. Rawle MJ et al (2018) Associations between polypharmacy and
plexity as factors associated with staff informant rated quality of cognitive and physical capability: a British birth cohort study. J
life in residents of aged care facilities: a cross-sectional study. Eur Am Geriatr Soc 66(5):916–923
J Clin Pharmacol 72(9):1117–1124 86. George C, Verghese J (2017) Polypharmacy and gait perfor-
75. Lalic S et al (2016) Polypharmacy and medication regimen com- mance in community-dwelling older adults. J Am Geriatr Soc
plexity as risk factors for hospitalization among residents of long- 65(9):2082–2087
term care facilities: a prospective cohort study. J Am Med Dir 87. Katsimpris A et al (2019) The association between polypharmacy
Assoc 17(11):1067 e1-1067 e6 and physical function in older adults: a systematic review. J Gen
76. Fried TR et al (2014) Health outcomes associated with polyphar- Intern Med 34(9):1865–1873
macy in community-dwelling older adults: a systematic review. J 88. Wang R et al (2015) Incidence and effects of polypharmacy on
Am Geriatr Soc 62(12):2261–2272 clinical outcome among patients aged 80+: a five-year follow-up
77. Laflamme L et al (2015) Type, number or both? A population- study. PLoS ONE 10(11):e0142123
based matched case-control study on the risk of fall injuries 89. Johnell K (2018) The controversies surrounding polyphar-
among older people and number of medications beyond fall- macy in old age - where are we? Expert Rev Clin Pharmacol
inducing drugs. PLoS ONE 10(3):e0123390 11(9):825–827
78. Dhalwani NN et al (2017) Association between polypharmacy 90. Taghy N et al (2020) Failure to reach a consensus in polypharmacy
and falls in older adults: a longitudinal study from England. BMJ definition: an obstacle to measuring risks and impacts-results of a
Open 7(10):e016358 literature review. Ther Clin Risk Manag 16:57–73
79. Ryan-Atwood TE et al (2017) Medication use and fall-related hos- 91. Wehling M et al (2016) VALFORTA: a randomised trial to vali-
pital admissions from long-term care facilities: a hospital-based date the FORTA (Fit fOR The Aged) classification. Age Ageing
case-control study. Drugs Aging 34(8):625–633 45(2):262–267
80. Izza MAD et al (2020) Polypharmacy, benzodiazepines, and anti- 92. O’Mahony D et al (2015) STOPP/START criteria for potentially
depressants, but not antipsychotics, are associated with increased inappropriate prescribing in older people: version 2. Age Ageing
falls risk in UK care home residents: a prospective multi-centre 44(2):213–218
study. Eur Geriatr Med 11(6):1043–1050
81. Niikawa H et al (2017) Association between polypharmacy and Publisher’s Note Springer Nature remains neutral with regard to
cognitive impairment in an elderly Japanese population residing jurisdictional claims in published maps and institutional affiliations.
in an urban community. Geriatr Gerontol Int 17(9):1286–1293
82. Park HY et al (2017) The association between polypharmacy
and dementia: a nested case-control study based on a 12-year

13