1Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

What is the epidemiology of medication errors, error-related adverse events and risk factors for errors in adults managed in community care contexts? A systematic review of the international literature

Ghadah Asaad Assiri,1,2,3 Nada Atef Shebl,4 Mansour Adam Mahmoud,5 Nouf Aloudah,2 Elizabeth Grant,6 Hisham Aljadhey,7 Aziz Sheikh8

To cite: Assiri GA, Shebl NA, Mahmoud MA, et al. What is the epidemiology of medication errors, error-related adverse events and risk factors for errors in adults managed in community care contexts? A systematic review of the international literature. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

► Prepublication history and additional material for this paper are available online. To view these files, please visit the journal online (http:// dx. doi. org/ 10. 1136/ bmjopen- 2017- 019101).

Received 11 August 2017Revised 13 February 2018Accepted 14 February 2018

For numbered affiliations see end of article.

Correspondence toGhadah Asaad Assiri; s1373565@ sms. ed. ac. uk

Research

AbstrACtObjective To investigate the epidemiology of medication errors and error-related adverse events in adults in primary care, ambulatory care and patients’ homes.Design Systematic review.Data source Six international databases were searched for publications between 1 January 2006 and 31 December 2015.Data extraction and analysis Two researchers independently extracted data from eligible studies and assessed the quality of these using established instruments. Synthesis of data was informed by an appreciation of the medicines’ management process and the conceptual framework from the International Classification for Patient Safety.results 60 studies met the inclusion criteria, of which 53 studies focused on medication errors, 3 on error-related adverse events and 4 on risk factors only. The prevalence of prescribing errors was reported in 46 studies: prevalence estimates ranged widely from 2% to 94%. Inappropriate prescribing was the most common type of error reported. Only one study reported the prevalence of monitoring errors, finding that incomplete therapeutic/safety laboratory-test monitoring occurred in 73% of patients. The incidence of preventable adverse drug events (ADEs) was estimated as 15/1000 person-years, the prevalence of drug–drug interaction-related adverse drug reactions as 7% and the prevalence of preventable ADE as 0.4%. A number of patient, healthcare professional and medication-related risk factors were identified, including the number of medications used by the patient, increased patient age, the number of comorbidities, use of anticoagulants, cases where more than one physician was involved in patients’ care and care being provided by family physicians/general practitioners.Conclusion A very wide variation in the medication error and error-related adverse events rates is reported in the studies, this reflecting heterogeneity in the populations studied, study designs employed and outcomes evaluated. This review has identified important limitations and discrepancies in the methodologies used and gaps in the

literature on the epidemiology and outcomes of medication errors in community settings.

IntrODuCtIOn Patient safety is a public concern in health-care systems across the world.1 Medication errors and error-related adverse drug events (ADEs) are common and are responsible for considerable patient harm.1 More specifically,

strengths and limitations of this study

► This is the first systematic review on the epidemi-ology of medication errors and medication-asso-ciated harm in community settings. The use of the International Classification for Patient Safety con-ceptual framework helped with framing and organ-ising the findings from this systematic review.

► A rigorous and transparent process has been em-ployed, which included no language restrictions in undertaking searches, independent screening of titles, abstracts and full-text papers, independent data extraction, and critical appraisal of included studies by two reviewers.

► Outcomes have been reported in a variety of ways using different tools and methodology, which made it difficult to undertake any quantitative pooled sum-mary of the results.

► Despite the comprehensiveness of the searches, we found no data regarding errors during medication dispensing and administration. This might be due to the lack of ‘dispensing error’ and ‘administration er-ror’ terms in our search strategy, although ‘medica-tion therapy management’ was included as a more overarching search term.

► There is at present no agreed, consistently applied set of confounders that should be taken into account when trying to make causal inferences.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

on June 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2017-019101 on 5 M

ay 2018. Dow

nloaded from

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

2 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

ADEs can lead to morbidity, hospitalisation, increased healthcare costs and, in some cases, death.1 It has been estimated that 5%–6% of all hospitalisations are drug-re-lated,2 3 with one estimate suggesting that ADEs causing hospital admission in the UK occur in around 10% of inpatients; approximately half of these ADEs are believed to be preventable.4 The cost of medication errors world-wide has been estimated as US$42 billion/year.5

Since the release of To Err is Human: Building a Safer Health System by the Institute of Medicine (now the National Academy of Medicine),6 which focused on acute care settings, most patient safety research has been conducted in hospital settings.7 8 Given that interna-tional and national policy drivers are for patients to be increasingly managed in primary, ambulatory and home settings in order to realise the goals of more accessible, patient-centred and efficient healthcare,9 there is an increased sense of urgency to further focus attention on community care contexts, particularly in relation to medication safety. With an ageing population, particu-larly in economically developed countries, as well as the use of polypharmacy, there is a need to empower patients, particularly those with chronic diseases, to self-care safely.

The aim of this systematic review was to investigate the epidemiology of medication errors, error-related adverse events and risk factors for errors in adults managed in community care contexts (ie, primary care, ambulatory and home settings). Box 1 provides definitions of the key terms employed in this review.

MethODsProtocol and reportingThe study protocol was developed following the Preferred Reporting Items for Systematic Reviews and Meta-Anal-yses (PRISMA) guidelines, and was registered in PROS-PERO.10 11 The detailed systematic review protocol has also been published.12

eligibility criteria/study selectionStudies conducted in adults (≥18 years) who were looked after in the community and living in their own or family homes without home healthcare or nursing home were eligible for inclusion in this review. The studied patients could have been self-managing, receiving care in primary care or ambulatory care settings, or any combination of the above. Studies were included if they were popu-lation-based, cross-sectional or cohort studies, which were suitable to estimate the incidence and prevalence of medication errors or ADEs. These study designs and case–control studies were considered eligible to study risk factors for the development of error-related ADEs. Studies with prescribed and/or over-the-counter (OTC) medications as the exposure of interest were eligible.

Paediatric studies (<18 years) and studies on patients receiving care in hospital at home settings (ie, contin-uous medical and/or nursing care provided to patients in their own homes), in nursing homes, as hospitalised inpa-tients or in emergency departments (ED) were excluded. Randomised controlled trials were excluded since these could not be used to reliably assess the incidence and/or prevalence of the outcomes of interest. Existing reviews were also excluded since the focus was on the primary literature. Incompletely reported studies, for example, in the form of abstracts, were not eligible for inclusion. Studies on illegal substance abuse, herbal products and those focusing on particular medications were also excluded.

No restriction on the language of publication was employed.

Data sources and search strategySearch terms were developed based on the system-atic review protocol.12 The search terms and detailed search strategies are presented in online supplemen-tary appendix 1. In summary, these involved identi-fying search terms (and their synonyms) in relation to medication safety, community care settings and study design, and combining these concepts with the Boolean operator AND to identify studies that inter-sected all three search concepts of interest. Exam-ples of the search terms used included the following: for the outcome: medication safety, medication error, preventable adverse drug event and patient error; for the setting: ambulatory care, outpatient, self-care, primary healthcare and general practice; and for the study design: cohort study, cross sectional study and observational study. Six biomedical databases were searched, including the Cumulative Index to Nursing

box 1 Key definitions

► Adverse drug event (ADE): Bates et al84 define ADE as ‘an injury resulting from medical intervention related to a drug’.84 Some ADEs are caused by underlying medication errors and therefore they are preventable.

► Medication error: The National Coordinating Council for Medication Error Reporting and Prevention defines a medication error as ‘any preventable event that may cause or lead to inappropriate medi-cation use or patient harm, while the medication is in the control of the health care professional, patient, or consumer. Such events may be related to professional practice, health care products, pro-cedures, and systems, including prescribing; order communication; product labelling, packaging, and nomenclature; compounding; dispensing; distribution; administration; education; monitoring; and use’.85 Medication errors can result from any step of the med-ication-use process: selection and procurement, storage, order-ing and transcribing, preparing and dispensing, administration, or monitoring.1

► Non-prescription drugs: Medicines that can be sold legally without a drug prescription.

► Over-the-counter (OTC) drug: The Food and Drug Administration defines OTC drugs as ‘drugs that have been found to be safe and appropriate for use without the supervision of a health care profes-sional such as a physician, and they can be purchased by consum-ers without a prescription’.86

► Prescription drug: Drugs that cannot be sold legally without a prescription.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

3Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

and Allied Health Literature, EMBASE, Eastern Medi-terranean Regional Office of the WHO, MEDLINE, PsycINFO and Web of Science, between 1 January 2006 and 31 December 2015. Google Scholar was searched for additional studies. An international panel of experts was also contacted to identify unpublished work and research in progress (online supplementary appendix 1). The reference list of all included studies was further reviewed for additional possible eligible studies.

The databases were searched by GAA. The title and abstracts were then independently screened for eligible studies according to the above detailed selection criteria by GAA and a second reviewer, NAS. The corre-sponding authors of the eligible articles were contacted if additional information was needed. Disagreements were resolved by discussion between the reviewers or by arbitration by a third reviewer, AS, if a decision could not be reached. Full-text articles were retrieved from selected studies and reviewed according to the selection criteria. Each copy of the selected studies was retrieved and the reason for excluding other studies was clearly noted.

Data extraction and risk of bias assessmentData were independently extracted and recorded onto a customised data extraction sheet by two reviewers (GAA and NAS, or GAA and MAM). Discrepancies were resolved by discussion or by arbitration by an additional reviewer (AS), if necessary.

Key information, such as study design, study type (retro-spective, prospective), population of interest, exposure of interest, outcomes of interest and main findings, was extracted.

The risk of bias assessment was independently carried out on each study by two reviewers (GAA and NAS, or GAA and NA) using the Critical Appraisal Skills Programme (CASP) quality assessment tool for cohort and case–control studies,13 and cross-sectional studies were assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for descriptive studies.14 Any disagreements were resolved by consensus or by arbitra-tion by a third reviewer (AS) if a decision could not be reached. Each study was given an overall grading as being at high, medium or low risk of bias.

Data synthesisData were summarised in detailed data tables, which included information on the incidence, prevalence, rela-tive risk and ORs, together with 95% CIs, for each study (where available). A descriptive and narrative synthesis of the extracted data was undertaken.

The following is the definition of incidence rate used in this review: ‘the number of patients with one or more [medica-tion error or preventable ADE] (numerator) divided by the total number of patients at risk per time unit (denominator)’.15 The following is the definition of prevalence rate used in the data extraction: ‘the number of patients experiencing one or more [medication error or preventable ADE] (numerator) divided

by the total number of patients in the study population (denom-inator)’.16 The prevalence rate per population was either reported and extracted directly from the included study or calculated from data provided in the study.

We worked with the definitions of medication errors and error-related ADEs employed in individual studies. These errors may have occurred anywhere in the medicines’ management process.1 Medication errors were described according to (1) the stage in the medicines’ management process when the error occurred, that is, prescribing, dispensing, administration and monitoring1; and (2) the type of error that occurred in each stage according to the conceptual framework for the International Classification for Patient Safety (ICPS) definitions (box 2).17

Risk factors were categorised as patient, healthcare professional and medication-related risk factors.

box 2 Classification of definitions used in this systematic review

► Administration error: ‘Any discrepancy between how the medication is given to the patient and the administration directions from the physician or hospital guidelines’.1

► Prescribing error: ‘Medication error occurring during the prescrip-tion of a medicine that is about writing the drug order or taking the therapeutic decision, appreciated by any non-intentional deviation from standard reference such as: the actual scientific knowledge, the appropriate practices usually recognized, the summary of the characteristics of the medicine product, or the mentions accord-ing to the regulations. A prescribing error notably can concern: the choice of the drug (according to the indications, the contraindica-tions, the known allergies and patient characteristics, interactions whatever nature it is with the existing therapeutics, and the other factors), dose, concentration, drug regimen, pharmaceutical form, route of administration, duration of treatment, and instructions of use; but also the failure to prescribe a drug needed to treat an al-ready diagnosed pathology, or to prevent the adverse effects of oth-er drugs’.17

► Inappropriate prescribing:  ‘The use of medicines that introduce a significant risk of an adverse drug-related event where there is ev-idence for an equally or more effective but lower-risk alternative therapy available for treating the same condition. Inappropriate pre-scribing also includes the use of medicines at a higher frequency and for longer than clinically indicated, the use of multiple medicines that have recognized drug–drug interactions and drug–disease in-teractions, and importantly, the under-use of beneficial medicines that are clinically indicated but not prescribed for ageist or irrational reasons’.87

► Monitoring error: ‘Failure to review a prescribed regimen for appro-priateness and detection of problems, or failure to use appropriate clinical or laboratory data for adequate assessment of patient re-sponse to prescribed theory’.17

► Dispensing error:  ‘Deviation from the prescriber’s order, made by staff in the pharmacy when distributing medications to nursing units or to patients in an ambulatory pharmacy setting’.17

► Other discrepancies:‘Any differences between the medication de-scribed by the patient and caregivers with the drugs listed by their general practitioners (GP) or between the medications listed in the discharge letter for the primary care physician with those in the patient discharge medication list’.31 32

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

4 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Changes from the original protocolThe following changes were made from the plans described in the research protocol12: (1) due to the large quantity of studies found during the initial search and because of medications and practice changes over the years, only studies published in the last 10 years were included: 1 January 2006–31 December 2015; (2) only studies with the incidence or prevalence rate per number of patients were included; and (3) meta-analysis was not

possible due to the heterogeneity of outcomes, methods and definitions.

resultsA total of 13 033 potentially eligible studies were iden-tified after removing duplicates, of which 59 studies met the inclusion criteria. One additional study was identified through hand-searching. Therefore, a total

Figure 1 PRISMA flow diagram (from Moher et al88). CINAHL, Cumulative Index to Nursing and Allied Health Literature; EMRO, Eastern Mediterranean Regional Office; RCT, randomised controlled trial. *Articles may be duplicated between the excluded groups.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

5Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

of 60 studies were included in the systematic review (figure 1).

One study was available only in German and one in Spanish. Those two papers were retrieved and translated into English by native speakers.18 19

The key characteristics of all included studies are presented in table 1. The quality assessments of these studies are summarised in tables 2A and 2B.

Nine studies were conducted in Asia, 4 in Australia, 32 in Europe, 8 in North America, 5 in South America and 2 were conducted across continents (one study covering two Australian countries, three European countries, one North American country and one South American country,20 and one study across two Australian countries, four European countries, one North American country and one South American country).21 Nineteen studies were conducted in primary healthcare or general prac-tice contexts, 15 studies in home or community settings, 16 studies in ambulatory care or outpatient settings, 5 studies in community pharmacies and 2 studies in post-discharge settings, while 3 studies used secondary data analysis.

Eleven studies enrolled adults in all age groups (>18 years), three studies reported the mean age only,22–24 one enrolled those 55 years or older,25 five enrolled those aged 60 years or older,26–30 and the majority of studies (n=40 studies, 67%) enrolled patients 65 years or older. If the study included adult and paediatric data, only rele-vant adult data were extracted.

The quality of the cross-sectional or descriptive studies using the JBI Critical Appraisal Checklist was high for nine studies, moderate for ten studies and low for one study. The quality of the cohort studies using the CASP quality assessment tool was high for 37 studies and moderate for 3 studies.

Different methods of medication errors and error-re-lated adverse events identification were used in the studies, including data review (electronic/paper-based medical record review, lab review, prescription review), database analysis, patient survey (face-to-face or tele-phone interview and survey or questionnaire), patient self-report and home visits.

MeDICAtIOn errOrsIncidence and/or prevalenceWe found no study reporting data on the incidence of medication errors. Estimates of community setting medication error prevalence were available from 53 studies.18–21 23 24 26 27 29–73

Self-reported medication errorsThe period prevalence of self-reported medication errors was measured in four cross-sectional studies by Adams et al, Lu and Roughead, Sears et al21 and Mira et al.20 21 72 73 In the first three studies, the period prevalence was reported as 2%, 6% and 6%, respectively,20 21 72 while in Mira et al’s study 75% of elderly patients with multiple comorbidities

and polypharmacy (five or more drugs) reported having made at least one mistake with their medication (including errors related to dose, similar appearance of medications and lack of understanding of the physician’s instructions).73 In this study, in 5% of cases, errors due to drug confusion had very severe consequences, requiring a visit to the emergency services or hospital admission.73 That wide differences in prevalence were seen between the first three studies and the last may be due to popula-tion factors. Mira et al’s study population comprised older polymedicated patients with multiple comorbidities. This elderly group had a greater risk of error, while the first three studies had populations including any patient over 18 years.

MeDICAtIOn errOr ACCOrDIng tO MeDICInes’ MAnAgeMent PrOCessPrescribing errorsThe point or period prevalence of prescribing errors was reported in 46 studies. In these studies, prescribing errors included errors in drug indications, drug–disease inter-actions, drug–drug interactions (DDI) and dosing error, as well as inappropriate prescribing, which was the most common error reported.

IndicationKoper et al23 found that, on average, 2.7 medications per patient were not indicated, with a total of 94% of patients having medications prescribed by the general practi-tioner (GP), but not mentioned in the indication of the UpToDate.23

Drug–disease interactions or contraindicationsDrug–disease interactions were measured in one study by Mand et al33 with a prevalence of 10%.33

Drug–drug interactionsThe prevalence of DDIs was measured in 11 studies and ranged from 2% to 58%.23 24 26 27 30 34–39 This could in part have been due to the fact that different DDI screening tools were used, namely DDI compendia and ePocrates RX, Thomson Micromedex program, Pharmavista data-base, BotPlus program of the General Council of Pharma-cists’ Official Colleges, British National Formulary 2010, Italian computerised interaction database, DrugDigest, Drugs, Micromedex and Medscape.

Inappropriate prescribingA. The prevalence of potentially inappropriate medica-

tion (PIM) was measured in 37 studies in the elder-ly age group only (≥65 years) and ranged from 5% to 94%.18 19 23 26 29 37 40–70 This extremely wide range of inappropriate prescribing prevalence estimates is likely to be, at least in part, due to the different detec-tion tools used, namely Beers 2003, the 2006 Health Plan Employer Data and Information Set, improved prescribing in the elderly tool, Medication Appro-priate Index, PRISCUS and Screening Tool of Older

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

6 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Tab

le 1

S

yste

mat

ic r

evie

w d

ata

extr

actio

n ta

ble

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

Sel

f-re

por

ted

med

icat

ion

erro

rs

1.A

dam

et

al, 2

00972

Aus

tral

iaC

ross

-sec

tiona

lA

naly

sis

of d

ata

from

352

2 ad

ults

par

ticip

atin

g in

st

age

2 of

the

Nor

th W

est

Ad

elai

de

Hea

lth S

tud

y ag

ed ≥

18 y

ears

Unc

lear

Sel

f-re

por

ted

ad

vers

e ev

ent

(med

icat

ion,

dia

gnos

is a

nd o

ther

s).

Usi

ng s

urve

y.

Of t

he t

otal

352

2 su

rvey

par

ticip

ants

, 148

(4.2

%) r

epor

ted

an

adve

rse

even

t ca

usin

g ha

rm in

the

pre

viou

s 12

mon

ths,

giv

ing

an a

nnua

l inc

iden

ce o

f 4.2

%

(95%

CI 3

.4%

to

5.0%

).M

edic

atio

n er

ror:

The

mai

n ty

pes

of a

dve

rse

even

ts p

erce

ived

as

caus

ing

harm

wer

e m

edic

atio

n er

ror

(rep

orte

d b

y 46

% o

f the

148

par

ticip

ants

rep

ortin

g ad

vers

e ev

ents

).

Med

icat

ion

erro

r p

reva

lenc

e:

68/3

522=

1.9%

Sub

ject

ive

dat

a ra

ther

than

ob

ject

ive

2.Lu

and

Rou

ghea

d,

2011

20A

ustr

alia

, Can

ada,

New

Z

eala

nd, U

K, U

SA

, G

erm

any

and

The

N

ethe

rland

s

Cro

ss-s

ectio

nal

(sec

ond

ary

anal

ysis

)11

910

ad

ult

resp

ond

ents

ag

ed ≥

18 y

ears

.D

ata

from

the

200

7 C

omm

onw

ealth

Fun

d

Inte

rnat

iona

lH

ealth

Pol

icy

Sur

vey.

Pre

scrib

ed d

rug

Sel

f-re

por

ted

med

icat

ion

erro

r an

d c

omp

are

fact

ors

asso

ciat

ed

with

med

icat

ion

erro

rs a

cros

s th

e se

ven

coun

trie

s.U

sing

sur

vey.

Sel

f-re

por

ted

med

icat

ion

erro

rs p

reva

lenc

e:75

2 re

spon

den

ts h

ad m

edic

atio

n er

ror

(Aus

tral

ia=

7.4%

; Can

ada=

5.7%

; New

Z

eala

nd=

5.9%

; UK

=5.

2%; U

SA

=7%

; Ger

man

y=5.

2%; T

he N

ethe

rland

s=8%

).R

isk

fact

ors

acro

ss c

ount

ries

incl

uded

see

ing

mul

tiple

sp

ecia

lists

, mul

tiple

ch

roni

c co

nditi

ons,

hos

pita

lisat

ion

and

mul

tiple

em

erge

ncy

room

vis

its.

Med

icat

ion

erro

r p

reva

lenc

e: 7

52/1

1 91

0=6.

3%

Pre

vale

nce

for

med

icat

ion

erro

r al

one

from

tab

le 1

, whi

le t

heris

k fa

ctor

s fo

r b

oth

med

ical

and

med

icat

ion

erro

r.

3.S

ears

et

al, 2

01221

Aus

tral

ia, C

anad

a,

Fran

ce, G

erm

any,

the

N

ethe

rland

s, N

ew

Zea

land

, UK

and

US

A

Des

crip

tive

(sec

ond

ary/

retr

osp

ectiv

e an

alys

is)

9944

ad

ults

age

d ≥

18 y

ears

fr

om t

heco

mm

unity

set

ting

Taki

ng m

edic

atio

n re

gula

rlyP

atie

nt-r

elat

ed r

isk

fact

ors

asso

ciat

ed w

ith s

elf-

rep

orte

d

med

icat

ion

erro

rs.

Usi

ng t

elep

hone

sur

vey.

Med

icat

ion

erro

r p

reva

lenc

e:57

0 re

spon

den

ts w

ith m

edic

atio

n er

rors

occ

urrin

g in

the

com

mun

ity s

ettin

g.

Ap

pro

xim

atel

y 4

out

of e

very

5 s

elf-

rep

orte

d m

edic

atio

n er

rors

occ

urre

d in

the

co

mm

unity

set

ting.

Med

icat

ion

erro

r p

reva

lenc

e:

570/

9944

=5.

7%

Ris

k fa

ctor

s fo

r b

oth

hosp

ital a

nd c

omm

unity

sett

ing

4.M

ira e

t al

, 201

373A

lican

te, S

pai

nC

ross

-sec

tiona

l38

2 el

der

ly a

ged

≥65

yea

rs

from

prim

ary

care

.P

atie

nts

on p

olyp

harm

acy

(five

or

mor

e d

rugs

) an

d w

ith c

omor

bid

ity:

card

iova

scul

ar (5

1.6%

); d

iab

etes

(34.

3%).

Pre

scrib

ed a

nd s

elf-

med

icat

ions

Freq

uenc

y of

mis

take

s in

co

mm

unic

atio

n b

etw

een

the

phy

sici

an a

nd t

he p

atie

nt a

nd t

heir

med

icat

ion

erro

r in

the

last

yea

r.U

sing

sem

istr

uctu

red

inte

rvie

ws.

Med

icat

ion

erro

r p

reva

lenc

e:75

.1%

of t

he p

atie

nts

rep

orte

d h

avin

g m

ade

at le

ast

one

mis

take

with

the

m

edic

atio

n in

the

last

yea

r.R

isk

fact

ors:

Mul

tiple

com

orb

iditi

es (p

=0.

006)

, fre

que

nt c

hang

es in

pre

scrip

tion

(p=

0.02

), no

t co

nsid

erin

g th

e p

resc

riptio

ns o

f oth

er p

hysi

cian

s (p

=0.

01),

inco

nsis

tenc

y in

the

m

essa

ges

(p=

0.01

), b

eing

tre

ated

by

vario

us d

iffer

ent

phy

sici

ans

at t

he s

ame

time

(p=

0.03

), a

feel

ing

of n

ot b

eing

list

ened

to

(p<

0.00

1) o

r lo

ss o

f tru

st in

the

p

hysi

cian

(p<

0.00

1).

*The

err

or d

ue t

o d

rug

conf

usio

n ha

d v

ery

seve

re c

onse

que

nces

, req

uirin

g a

visi

t to

the

em

erge

ncy

serv

ice

or h

osp

ital a

dm

issi

on.

Med

icat

ion

erro

r p

reva

lenc

e:

287/

382=

75%

Con

seq

uenc

e*

Ris

k fa

ctor

s

5.S

oren

sen

et a

l, 20

0676

4 st

ates

of A

ustr

alia

Cro

ss-s

ectio

nal,

pro

spec

tive

204

gene

ral p

ract

ice

pat

ient

s liv

ing

in t

heir

own

hom

e ag

ed 3

7–99

yea

rs

Pre

scrib

ed d

rugs

Pre

vale

nce

and

inte

rrel

atio

nshi

ps

of m

edic

atio

n-re

late

d r

isk

fact

ors

for

poo

r p

atie

nt h

ealth

out

com

es

iden

tifiab

le t

hrou

gh ‘i

n-ho

me’

vis

it ob

serv

atio

ns.

Ris

k fa

ctor

s:P

reva

lenc

e of

nom

inal

med

icat

ion-

rela

ted

ris

k fa

ctor

s an

d h

ealth

out

com

es

amon

g th

e sa

mp

le o

f 204

pat

ient

s.1.

Mul

tiple

med

icat

ion

stor

age

loca

tions

use

d=

17 (8

.3%

).2.

Exp

ired

med

icat

ion

pre

sent

=40

(19.

6%).

3. D

isco

ntin

ued

med

icat

ion

rep

eats

ret

aine

d=

43 (2

1%).

4. H

oard

ing

of m

edic

atio

ns=

43 (2

1%).

5. T

hera

peu

tic d

uplic

atio

n p

rese

nt=

50 (2

4.5%

).A

dm

inis

trat

ion

erro

r:6.

No

med

icat

ion

adm

inis

trat

ion

rout

ine=

56 (2

7.5%

).7.

Poo

r ad

here

nce=

107

(52.

5%).

8. C

onfu

sed

by

gene

ric a

nd t

rad

e na

mes

=11

4 (5

5.9%

).

6.Vu

ong

and

Mar

riott

, 20

0625

Mel

bou

rne,

Aus

tral

iaD

escr

iptiv

e14

2 d

isch

arge

d a

dul

ts

aged

≥55

yea

rs w

ho w

ere

retu

rnin

g to

ind

epen

den

t ca

re a

t ho

me.

Pat

ient

s at

ris

k of

m

edic

atio

n m

isad

vent

ure.

Dis

char

ge

pre

scrib

ed d

rugs

Unn

eces

sary

med

icin

e st

ored

at

hom

e as

a r

isk

fact

or.

Usi

ng h

ome

visi

t w

ithin

5 d

ays

of

dis

char

ge.

Unn

eces

sary

med

icin

e st

ored

at

hom

e p

reva

lenc

e: 8

5/14

2=60

%.

85 (6

0%) o

f 142

pat

ient

s w

ho r

ecei

ved

a h

ome

visi

t al

low

ed r

emov

al o

f m

edic

ines

tha

t ha

d e

xpire

d o

r no

long

er r

equi

red

.P

resc

ribin

g er

ror:

dru

g d

uplic

atio

n p

reva

lenc

e:32

(27%

) pat

ient

s al

low

ed r

emov

al o

f 82

dup

licat

e p

acks

of t

he s

ame

item

tha

t w

as n

o lo

nger

req

uire

d.

A t

otal

of 3

90 m

edic

ines

wer

e re

mov

ed w

ith a

mea

n of

4.6

med

icin

es p

er

pat

ient

(ran

ge 1

–21)

.

Unn

eces

sary

med

icin

e st

ored

at

hom

e p

reva

lenc

e: 8

5/14

2=60

%

No

info

rmat

ion

on h

owm

any

pat

ient

s ha

dun

nece

ssar

y m

edic

ine.

Info

rmat

ion

avai

lab

le is

on t

he p

atie

nt a

llow

ed t

ore

mov

e un

nece

ssar

ym

edic

ine.

7.P

it et

al,

2008

74N

ew S

outh

Wal

es,

Aus

tral

ia.

Cro

ss-s

ectio

nal s

tud

y84

9 el

der

ly a

ged

≥65

yea

rs

from

gen

eral

pra

ctic

eS

elf-

med

icat

ions

Pre

vale

nce

of s

elf-

rep

orte

d

risk

fact

ors

for

med

icat

ion

mis

adve

ntur

es.

Tool

use

d: M

edic

atio

n R

isk

Ass

essm

ent

Form

(pat

ient

sur

vey)

Ris

k fa

ctor

s:1.

Usi

ng a

t le

ast

one

med

icat

ion

for

mor

e th

an 6

mon

ths

(95%

).2.

Mor

e th

an o

ne d

octo

r in

volv

ed in

the

ir ca

re (5

9%).

3. H

ad t

hree

or

mor

e he

alth

con

diti

ons

(57%

).4.

Use

d fi

ve o

r m

ore

med

icin

es (5

4%).

5. A

DR

s, in

the

last

mon

th 3

9% o

f par

ticip

ants

exp

erie

nced

diffi

culti

es s

leep

ing,

fe

lt d

row

sy o

r d

izzy

(34%

), ha

d a

ski

n ra

sh (2

8%),

leak

ed u

rine

(27%

), ha

d

stom

ach

pro

ble

ms

(22%

) or

had

bee

n co

nstip

ated

(22%

).

*AD

R a

s a

risk

fact

or fo

rm

edic

atio

nm

isad

vent

ure

may

not

be

rela

ted

to

the

use

ofm

edic

atio

n in

all

case

s.

8.M

oshe

r et

al,

2012

75Io

wa,

US

AC

ohor

t p

rosp

ectiv

e31

0 el

der

ly a

ged

≥65

yea

rs

who

wer

e co

gniti

vely

in

tact

from

a V

eter

ans

Ad

min

istr

atio

n p

rimar

y ca

re c

linic

Taki

ng fi

ve o

r m

ore

non-

top

ical

m

edic

atio

ns

Ass

ocia

tion

of h

ealth

lite

racy

with

m

edic

atio

n kn

owle

dge

, ad

here

nce

and

AD

Es.

Usi

ng in

terv

iew

and

cha

rt r

evie

w.

Tota

l: 31

0 p

atie

nts

Pre

vale

nce

of A

DE

s:A

DE

s oc

curr

ed in

51

pat

ient

s (1

6.5%

) of t

he p

atie

nts

with

in t

he fi

rst

3 m

onth

s of

the

stu

dy,

whi

ch in

crea

sed

to

119

pat

ient

s (3

8.4%

) ove

r th

e fu

ll 12

-mon

th

follo

w-u

p p

erio

d.

Ris

k fa

ctor

:A

ssoc

iatio

n of

hea

lth li

tera

cy w

ith A

DE

s:Th

e in

cid

ence

of A

DE

s at

3 a

nd 1

2 m

onth

s ap

pea

red

hig

her

amon

g p

atie

nts

with

low

hea

lth li

tera

cy, b

ut t

his

was

not

sta

tistic

ally

sig

nific

ant.

Low

hea

lth li

tera

cy

incr

ease

d t

he r

isk

of

AD

Es.

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

7Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

Med

icin

es’ m

anag

emen

t p

roce

ss:

9.K

oper

et

al, 2

01323

Aus

tria

Des

crip

tive

169

pat

ient

s fr

om g

ener

al

pra

ctic

e ta

king

five

or

mor

e m

edic

ines

.M

ean

age:

76.

4±8.

5 S

D

year

s.O

f the

169

pat

ient

s,

158

wer

e el

der

ly a

ged

≥6

5 ye

ars.

Pre

scrib

ed a

nd

OTC

dru

gM

edic

atio

n er

rors

incl

udin

g no

n-ev

iden

ce-b

ased

med

icat

ions

, d

osin

g er

rors

and

pot

entia

lly

dan

gero

us in

tera

ctio

ns in

all

pat

ient

s.P

oten

tial i

nter

actio

ns w

ere

iden

tified

usi

ng t

he L

exi-

Inte

ract

d

atab

ase.

PIM

s in

sub

grou

p o

f eld

erly

p

atie

nts

acco

rdin

g to

the

P

RIS

CU

S li

st.

Usi

ng c

ase

rep

ort

form

fille

d b

y th

e G

Ps.

Pre

scrib

ing

erro

r p

reva

lenc

e:In

dic

atio

n:15

8 of

the

169

pat

ient

s (9

3.5%

) had

at

leas

t on

e no

n-ev

iden

ce-b

ased

m

edic

atio

n.D

osin

g er

ror:

74 o

f the

169

pat

ient

s (4

3.8%

) had

at

leas

t on

e d

osin

g er

ror.

DD

I pre

vale

nce:

Cat

egor

y D

inte

ract

ions

: 99

pat

ient

s (5

8%) h

ad a

t le

ast

one

cate

gory

D

inte

ract

ion.

Cat

egor

y X

inte

ract

ions

: 4 p

atie

nts

(2.4

%) h

ad a

t le

ast

one

cate

gory

X

inte

ract

ion.

PIM

pre

vale

nce:

59 o

f sen

iors

(37.

3%) h

ad a

t le

ast

one

med

icat

ion

that

was

inap

pro

pria

te.

Med

icat

ion

erro

r p

reva

lenc

e:1.

Non

-evi

den

ce-

bas

ed m

edic

atio

ns:

158/

169=

93.5

%.

2. D

osin

g er

ror:

74

/169

=43

.8%

.3.

Cat

egor

y D

dru

gin

tera

ctio

n:

99/1

68=

58%

; cat

egor

y X

dru

g in

tera

ctio

n:

4/16

8=2.

4%.

4. P

IMs:

59/

158=

37.3

%.

A m

edic

atio

n w

ascl

assi

fied

as

non-

evid

ence

-bas

ed if

the

ind

icat

ion

for

use

ind

icat

ed b

y th

e G

Pw

as n

ot m

entio

ned

inan

y p

eer-

revi

ewed

chap

ter

of U

pTo

Dat

e.

10.

Man

d e

t al

, 201

433G

erm

any

Des

crip

tive

retr

osp

ectiv

e24

619

eld

erly

age

d

≥65

year

s fr

om fa

mily

p

ract

ice

with

at

leas

t on

e d

iagn

osis

nam

ed in

the

B

eers

list

Pre

scrib

ed d

rug

PD

DI f

req

uenc

y an

d w

heth

er t

here

ar

e ge

nder

-rel

ated

or

age-

rela

ted

d

iffer

ence

s.A

naly

sis

from

ele

ctro

nic

pat

ient

re

cord

s.

Pre

scrib

ing

erro

r:C

ontr

aind

icat

ion

or d

rug–

dis

ease

inte

ract

ion

pre

vale

nce:

10.4

% o

f eld

erly

wer

e ex

pos

ed t

o at

leas

t on

e P

DD

I.R

isk

fact

ors:

1. P

atie

nts

over

75

year

s (O

R 1

.10,

CI 1

.05

to 1

.15)

.2.

Num

ber

of d

rugs

pre

scrib

ed (>

4 d

rugs

: OR

1.9

1, C

I 1.8

3 to

2.0

0).

3. B

lood

clo

ttin

g d

isor

der

s/re

ceiv

ing

antic

oagu

lant

the

rap

y (O

R 2

.38,

CI 2

.15

to

2.64

) sho

wed

the

str

onge

st a

ssoc

iatio

n w

ith P

DD

I.

PD

DI p

reva

lenc

e:

2560

/24

619=

10.4

%

11.

Gag

ne e

t al

, 200

836R

egio

ne E

mili

a-R

omag

na, I

taly

Coh

ort

retr

osp

ectiv

e4

222

165

regi

onal

Em

ilia-

Rom

agna

res

iden

ts.

Out

pat

ient

age

d fr

om 0

to

≥85

year

s.

Pre

scrib

ed d

rug

Clin

ical

ly im

por

tant

pot

entia

l DD

I.R

isk

fact

ors.

Out

pat

ient

pre

scrip

tion

dat

a fr

om

the

Reg

iona

l Em

ilia-

Rom

agna

.D

DI s

cree

ning

too

l: a

list

of

clin

ical

ly im

por

tant

pot

entia

l DD

Is

incl

uded

12

dru

g p

airs

tha

t co

uld

b

e ca

ptu

red

usi

ng t

he r

egio

nal

Em

ilia-

Rom

agna

dat

abas

e.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: e

xpos

ed t

o p

oten

tial D

DI a

dul

ts (1

9 to

≥85

yea

rs)=

7893

.U

nexp

osed

ad

ult=

7013

.To

tal=

14 9

06.

DD

I pre

vale

nce:

789

3/14

90

6=53

%R

isk

fact

ors

for

all

age

grou

ps

incl

udin

g p

aed

iatr

ics.

All

age

grou

ps

incl

uded

so

resu

lts s

houl

d b

e co

nsid

ered

cau

tious

ly.

12.

Dal

lenb

ach

et a

l, 20

0724

Gen

eva,

Sw

itzer

land

Des

crip

tive,

re

tros

pec

tive

file

revi

ew59

1 ou

tpat

ient

s, m

ean

age

39 y

ears

Pre

scrip

tion

dru

g an

d d

rug

curr

ently

ta

king

Clin

ical

ly s

igni

fican

t A

DI.

Pre

scrip

tion

revi

ew.

DD

I scr

eeni

ng t

ool:

DD

I co

mp

end

ia a

nd (e

Poc

rate

s R

X)

with

clin

ical

dec

isio

n su

pp

ort.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: In

135

of t

he c

onsu

ltatio

ns, a

pot

entia

lly c

linic

ally

sig

nific

ant

AD

I was

iden

tified

.

DD

I pre

vale

nce:

13

5/59

1=23

%

13.

Ob

reli

Net

o et

al,

2011

26B

razi

lC

ross

-sec

tiona

l26

27 e

lder

ly a

ged

60–

88 y

ears

from

the

prim

ary

heal

thca

re

Pre

scrib

ed d

rug

Pot

entia

l ris

ks in

dru

g p

resc

riptio

ns: D

DI a

nd P

IM.

Usi

ng p

resc

riptio

n re

view

.D

DI s

cree

ning

too

l: (D

rugD

iges

t,

Med

scap

e an

d M

icro

med

ex).

PIM

usi

ng B

eers

crit

eria

200

3.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: U

sing

Dru

gDig

est

show

ed t

hat

4.7%

and

28.

4% o

f the

el

der

ly p

rese

nted

at

leas

t on

e p

oten

tial D

DI c

lass

ified

as

maj

or a

nd m

oder

ate,

re

spec

tivel

y.U

sing

Med

scap

e sh

owed

tha

t 3.

4% a

nd 1

9.3%

of t

he e

lder

ly p

rese

nted

at

leas

t on

e p

oten

tial D

DI c

lass

ified

as

maj

or a

nd m

oder

ate,

res

pec

tivel

y.U

sing

Mic

rom

edex

sho

wed

tha

t 3.

1% a

nd 2

9.1%

of t

he e

lder

ly p

rese

nted

at

leas

t on

e p

oten

tial D

DI c

lass

ified

as

maj

or a

nd m

oder

ate,

res

pec

tivel

y.P

resc

ribin

g er

ror:

PIM

pre

vale

nce:

26.

9% o

f the

pat

ient

s ha

d p

resc

riptio

ns w

ith a

t le

ast

one

PIM

.

DD

I pre

vale

nce:

3.

1%–2

9.1%

PIM

pre

vale

nce:

26.

9%

14.

Sec

oli e

t al

, 201

030S

ao P

aulo

, Bra

zil

Cro

ss-s

ectio

nal

2143

com

mun

ity-d

wel

ling

eld

erly

age

d ≥

60 y

ears

.D

ata

wer

e ob

tain

ed fr

om

the

SA

BE

(Hea

lth, W

ell-

Bei

ng a

nd A

gein

g) s

urve

y.

≥2 p

resc

ribed

d

rug

use

Pot

entia

l DD

Is a

nd id

entif

y as

soci

ated

fact

ors.

Usi

ng h

ome

inte

rvie

w.

DD

I scr

eeni

ng t

ool:

Mic

rom

edex

H

ealth

care

Ser

ies.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: 5

68/2

143=

26.5

%.

Ris

k fa

ctor

s:Th

e us

e of

six

or

mor

e m

edic

atio

ns (O

R 3

.37,

95%

CI 2

.08t

o 5.

48) o

r ha

ving

hy

per

tens

ion

(OR

2.5

6, 9

5% C

I 1.7

3 to

3.7

9), d

iab

etes

(OR

1.7

3, 9

5% C

I 1.2

2to

2.44

) or

hear

t p

rob

lem

s (O

R 3

.36,

95%

CI 2

.11

to 5

.34)

sig

nific

antly

incr

ease

d

the

risk

of p

oten

tial D

DI.

DD

I pre

vale

nce:

56

8/21

43=

26.5

%

15.

Ob

reli

Net

o et

al,

2012

275

citie

s of

Bra

zil

Cro

ss-s

ectio

nal

12 3

43 e

lder

ly a

ged

≥6

0 ye

ars

from

the

prim

ary

pub

lic h

ealth

sys

tem

Pre

scrip

tion

for

two

or m

ore

dru

gs

(pre

scrib

ed b

oth

with

in a

nd a

cros

s p

resc

riptio

ns)

Pot

entia

l DD

Is (p

rese

nce

of a

m

inim

um o

f 5 d

ays

over

lap

in

sup

ply

of a

n in

tera

ctin

g d

rug

pai

r)

and

pre

dic

tor

of D

DI.

Usi

ng m

edic

al p

resc

riptio

ns a

nd

pat

ient

s’ m

edic

al r

ecor

ds

revi

ew.

DD

I scr

eeni

ng t

ool:

DD

I che

cker

p

rogr

amm

es (D

rugD

iges

t, D

rugs

, M

icro

med

ex a

nd M

edsc

ape)

.

12 3

43 p

atie

nts

(585

5 ex

pos

ed; 6

488

unex

pos

ed).

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: 4

7.4%

Ris

k fa

ctor

s:Fe

mal

e se

x (O

R=

2.49

(95%

CI 2

.29

to 2

.75)

), d

iagn

osis

of ≥

3 d

isea

ses

(OR

=6.

43

(95%

CI 3

.25

to 1

2.44

)) an

d d

iagn

osis

of h

yper

tens

ion

(OR

=1.

68 (9

5% C

I 1.2

3 to

2.4

1)) w

ere

asso

ciat

ed w

ith p

oten

tial D

DIs

.A

ge w

as a

ssoc

iate

d w

ith a

n in

crea

sing

ris

k of

DD

Is. N

umb

er o

f pre

scrib

ers,

nu

mb

er o

f dru

gs c

onsu

med

, ATC

cod

es a

nd d

rugs

tha

t ac

t on

CY

P45

0 p

rese

nted

pos

itive

ass

ocia

tions

with

pot

entia

l DD

Is in

uni

varia

te a

nd

mul

tivar

iate

ana

lyse

s of

dru

g th

erap

y ch

arac

teris

tics.

DD

I pre

vale

nce:

585

5/12

34

3=47

.4%

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

8 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

16.

Ind

erm

itte

et a

l, 20

0734

Sw

itzer

land

Des

crip

tive

434

pas

ser-

by

cust

omer

s ag

ed ≥

18 y

ears

from

co

mm

unity

pha

rmac

ies

Pre

scrip

tion-

only

m

edic

ines

and

OTC

d

rug

Pot

entia

l dru

g in

tera

ctio

ns.

1. O

bse

rvat

ion

of c

usto

mer

co

ntac

ts a

nd in

terv

iew

s w

ith

pas

ser-

by

cust

omer

s p

urch

asin

g se

lect

ed O

TC d

rugs

.2.

Tel

epho

ne in

terv

iew

s w

ith

regu

lar

cust

omer

s tr

eate

d w

ith

sele

cted

pre

scrip

tion-

only

m

edic

ines

iden

tified

in c

omm

unity

p

harm

acie

s’ d

atab

ases

.D

DI s

cree

ning

too

l: P

harm

avis

ta

dat

abas

e.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: O

bse

rvat

ion

of p

asse

r-b

y cu

stom

ers.

Of 1

183

pas

ser-

by

cust

omer

s ob

serv

ed, 1

64 p

urch

ased

at

leas

t on

e of

the

se

lect

ed O

TC d

rugs

.10

2 (6

2.2%

) of t

hose

sub

ject

s w

ere

inte

rvie

wed

. 43

(42.

2%) m

entio

ned

tak

ing

pre

scrib

ed d

rugs

and

3 o

f the

m w

ere

exp

osed

to

pot

entia

l dru

g in

tera

ctio

ns o

f m

oder

ate

seve

rity.

Tele

pho

ne in

terv

iew

with

reg

ular

cus

tom

ers.

Out

of 5

92 r

egul

ar c

usto

mer

s se

lect

ed fr

om t

he c

omm

unity

pha

rmac

y d

atab

ase,

434

(73.

3%) c

ould

be

inte

rvie

wed

.P

reva

lenc

e of

DD

I in

regu

lar

cust

omer

s:69

(15.

9%) o

f the

m w

ere

exp

osed

to

a p

oten

tial d

rug

inte

ract

ion

bet

wee

n p

urch

ased

OTC

dru

g fo

r se

lf-m

edic

atio

n an

d t

heir

pre

scrip

tion-

only

med

icin

es.

Furt

herm

ore,

116

(26.

7%) r

egul

ar c

usto

mer

s w

ere

exp

osed

to

pot

entia

l dru

g in

tera

ctio

ns w

ithin

the

ir p

resc

ribed

dru

gs a

nd in

28

(6.5

%) m

ultip

le (>

2) p

oten

tial

dru

g in

tera

ctio

ns w

ere

foun

d.

DD

I pre

vale

nce:

3/

102=

3%,

69/4

34=

16%

, 11

6/43

4=26

.7%

17.

Mah

moo

d e

t al

, 20

0735

US

AC

ross

-sec

tiona

l, re

tros

pec

tive

2 79

5 34

5 p

atie

nts

who

fil

led

pre

scrip

tions

for

med

icat

ions

invo

lved

p

oten

tial D

DI f

rom

128

Ve

tera

ns A

ffairs

med

ical

ce

ntre

s.A

mb

ulat

ory

care

clin

ic.

Pre

scrib

ed d

rug

Clin

ical

ly im

por

tant

DD

I.D

atab

ase

anal

ysis

of p

harm

acy

reco

rds.

DD

I scr

eeni

ng t

ool:

a lis

t of

25

pot

entia

l DD

I.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e:Th

e ov

eral

l rat

e of

pot

entia

l DD

Is w

as 2

1.54

per

100

0 ve

tera

ns e

xpos

ed t

o th

e ob

ject

or

pre

cip

itant

med

icat

ions

of i

nter

est.

DD

I pre

vale

nce:

2.1

5%A

ge n

ot m

entio

ned

18.

Lap

i et

al, 2

00937

Dic

oman

o, It

aly

Coh

ort,

a t

wo-

wav

e,

pop

ulat

ion-

bas

ed s

urve

y56

8 co

mm

unity

-dw

ellin

g el

der

ly a

ged

≥65

yea

rsP

resc

riptio

n an

d

non-

pre

scrip

tion

dru

gs u

sed

at

leas

t 1

wee

k b

efor

e en

rolm

ent

Sub

optim

al p

resc

ribin

g:In

app

rop

riate

med

icat

ion=

1991

B

eers

crit

eria

(13

item

s ou

t of

th

e or

igin

al 3

9 (3

3.3%

) Bee

rs li

st

med

icat

ions

wer

e co

nsid

ered

).D

DI s

cree

ning

too

l: M

icro

med

ex

Dru

g-R

eax

syst

em.

Usi

ng p

opul

atio

n-b

ased

sur

vey.

Pre

scrib

ing

erro

r:P

oten

tial D

DI p

reva

lenc

e w

as s

igni

fican

tly h

ighe

r in

199

9 co

mp

ared

with

199

5 (3

0.5%

vs

20.1

%; p

<0.

001)

.In

app

rop

riate

pre

scrip

tions

wer

e si

gnifi

cant

ly h

ighe

r in

199

5 co

mp

ared

with

19

99 (9

.1%

vs

5.1%

; p=

0.00

4).

Pot

entia

l DD

I p

reva

lenc

e: 3

0.5%

, p

<0.

001

Inap

pro

pria

te m

edic

atio

n p

reva

lenc

e: 5

.1%

, p

=0.

004

1995

1999

P v

alue

s

Inap

pro

pria

te

med

icat

ion

47 (9

.1%

)26

(5.1

%)

0.00

4

DD

I97

(20.

1%)

147

(30.

5%)

<0.

001

Maj

or D

DI

20 (4

.7%

)24

(5.6

%)

0.58

5

Ris

k fa

ctor

s:P

olyp

harm

acy

alw

ays

pre

dic

ted

a s

ubst

antia

l inc

reas

e in

the

ris

k of

the

PIM

an

d D

DI.

19.

Nob

ili e

t al

, 200

938Le

cco,

Ital

yC

ross

-sec

tiona

l, re

tros

pec

tive

58 8

00 c

omm

unity

-dw

ellin

g el

der

ly a

ged

≥65

yea

rs

regi

ster

ed u

nder

the

loca

l he

alth

auth

ority

of L

ecco

Rec

eivi

ng a

t le

ast

two

coad

min

iste

red

p

resc

riptio

ns

DD

Is a

nd a

ssoc

iate

d r

isk

fact

ors

(age

, sex

and

num

ber

of

pre

scrip

tions

).D

DI s

cree

ning

too

l: Ita

lian

com

put

eris

ed in

tera

ctio

n d

atab

ase.

Ana

lyse

d a

ll p

resc

riptio

ns d

isp

ense

d fr

om 1

Ja

nuar

y 20

03 t

o 31

Dec

emb

er

2003

.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: 9

427

eld

erly

peo

ple

(16%

) wer

e ex

pos

ed t

o d

rug

com

bin

atio

ns

with

the

pot

entia

l for

13

932

seve

re D

DIs

.M

ean

num

ber

of D

DI p

er p

atie

nt w

as 0

.2 (r

ange

0–9

).R

isk

fact

ors:

Age

and

num

ber

of c

hron

ic d

rugs

wer

e as

soci

ated

with

an

incr

easi

ng r

isk

of

DD

Is.

The

adju

sted

OR

incr

ease

d fr

om 1

.07

(95%

CI 1

.3 t

o 1.

11) i

n p

atie

nts

aged

70

–74

year

s to

1.5

2 (9

5% C

I 1.4

6 to

1.6

0) in

tho

se a

ged

85

or o

lder

.E

lder

ly t

akin

g m

ore

than

five

chr

onic

dru

gs h

ad a

sta

tistic

ally

sig

nific

ant

high

er

risk

of p

oten

tially

sev

ere

DD

Is (O

R=

5.59

, 95%

CI 5

.39

to 5

.80)

tha

n th

ose

rece

ivin

g le

ss t

han

3 (re

fere

nce

cate

gory

) or

3–5

chro

nic

dru

gs (O

R=

2.71

, 95

% C

I 2.6

3 to

2.8

0).

Pot

entia

lly s

ever

e D

DI

pre

vale

nce:

942

7/58

80

0=16

%

Onl

y th

e in

tera

ctio

ns

iden

tified

as

seve

re w

ere

cons

ider

ed in

the

se

anal

yses

.

20.

Gut

hrie

et

al, 2

01539

Sco

tland

, UK

Cro

ss-s

ectio

nal

311

881

resi

den

ts a

ged

≥2

0 ye

ars

from

the

co

mm

unity

-dis

pen

sed

p

resc

ribin

g d

ata

(gen

eral

p

ract

ice)

.Li

ving

in o

wn

hom

e:

308

660.

Pre

scrib

ed d

rugs

Pot

entia

lly s

erio

us D

DI.

Pat

ient

cha

ract

eris

tics

asso

ciat

ed

with

the

pre

senc

e of

pot

entia

lly

serio

us D

DI.

DD

I scr

eeni

ng t

ool:

anal

ysis

of

com

mun

ity-d

isp

ense

d p

resc

ribin

g d

ata

usin

g B

ritis

h N

atio

nal

Form

ular

y 20

10.

Pre

scrib

ing

erro

r:D

DI p

reva

lenc

e: 4

0 68

9 ad

ults

(13%

) had

pot

entia

lly s

erio

us D

DI i

n 20

10 (f

or

bot

h re

sid

ents

livi

ng in

ow

n ho

me

and

car

e ho

me)

.N

umb

er o

f pat

ient

with

pot

entia

lly s

erou

s D

DI f

or r

esid

ence

livi

ng in

the

ir ow

n ho

me

in 2

010=

13 6

15.

DD

I pre

vale

nce:

13

615/

308

660=

4.4%

Res

iden

t liv

ing

in b

oth

care

hom

e or

ow

n ho

me.

Ris

k fa

ctor

s fo

r ow

nho

me

and

car

e ho

me.

21.

Mai

o et

al,

2006

40E

mili

a-R

omag

na, I

taly

Coh

ort

retr

osp

ectiv

e84

9 42

5 el

der

ly o

utp

atie

nts

aged

≥65

yea

rs fr

om t

he

Em

ilia-

Rom

agna

out

pat

ient

p

resc

riptio

n cl

aim

s d

atab

ase

Pre

scrib

ed d

rugs

PIM

usi

ng t

he 2

002

Bee

rs c

riter

ia

and

fact

ors

asso

ciat

ed w

ith P

IM.

Pre

scrip

tion

revi

ew.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: A

tot

al o

f 152

641

(18%

) eld

erly

had

one

or

mor

e oc

curr

ence

s of

PIM

pre

scrib

ing.

Ris

k fa

ctor

s:1.

Old

er a

ge (≥

85 y

ears

) (O

R 1

.18,

95%

CI 1

.16

to 1

.2, p

<0.

05).

2. ≥

10 d

rugs

pre

scrib

ed (O

R 7

.33,

95%

CI 7

.15

to 7

.51,

p<

0.05

).3.

≥4

chro

nic

cond

ition

s (O

R 1

.76,

95%

CI 1

.72

to 1

.81,

p<

0.05

).

PIM

pr e

vale

nce:

152

64

1/84

9 42

5=18

%

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

9Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

22.

de

Oliv

eira

Mar

tins

et

al, 2

00641

Lisb

on, P

ortu

gal

Cro

ss-s

ectio

nal

213

eld

erly

age

d ≥

65 y

ears

fr

om 1

2 co

mm

unity

p

harm

acie

s

Pre

scrip

tion

and

ho

me

med

icat

ions

IDU

by

1997

Bee

rs a

nd 2

003

Bee

rs e

xplic

it cr

iteria

.U

sing

sur

vey.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: U

sing

the

199

7 B

eers

exp

licit

crite

ria, 7

5 oc

curr

ence

s of

in

app

rop

riate

med

icin

es w

ere

det

ecte

d in

59

pat

ient

s (2

7.7%

). U

sing

the

200

3 B

eers

exp

licit

crite

ria in

app

rop

riate

med

icat

ion

was

det

ecte

d in

82

pat

ient

s (3

8.5%

).R

isk

fact

ors:

The

occu

rren

ce o

f ina

pp

rop

riate

med

icin

es w

as s

igni

fican

tly a

ssoc

iate

d w

ith t

he

cons

ump

tion

of a

hig

h nu

mb

er o

f dru

gs.

IDU

pre

vale

nce:

59

/213

=27

.7%

usi

ng

1997

Bee

rsID

U p

reva

lenc

e:

82/2

13=

38.5

% u

sing

20

03 B

eers

23.

Pug

h et

al,

2006

42A

ustin

, Tex

as, U

SA

Cro

ss-s

ectio

nal,

retr

osp

ectiv

e1

096

361

outp

atie

nt

eld

erly

age

d ≥

65 y

ears

us

ing

natio

nal d

ata

from

th

e Ve

tera

ns H

ealth

A

dm

inis

trat

ion

Pre

scrib

ed d

rug

only

Pot

entia

lly IP

incl

uded

in t

he 2

006

HE

DIS

crit

eria

and

to

det

erm

ine

if p

atie

nt r

isk

fact

ors

are

sim

ilar

to

thos

e fo

und

usi

ng B

eers

crit

eria

.U

sing

dat

abas

e.

Pre

scrib

ing

erro

r:IP

pre

vale

nce:

Ove

rall,

19.

6% o

f old

er v

eter

ans

wer

e ex

pos

ed t

o H

ED

IS 2

006

dru

gs.

Ris

k fa

ctor

s:1.

Pat

ient

s re

ceiv

ing

≥10

med

icat

ions

wer

e at

gre

ates

t ris

k of

exp

osur

e in

men

(O

R 8

.2, 9

5% C

I 8 t

o 8.

4) a

nd w

omen

(OR

9.6

, 95%

CI 8

.2 t

o 11

.2).

2. P

atie

nts

with

mor

e ou

tpat

ient

clin

ic v

isits

(≥10

) wer

e at

gre

ater

ris

k re

gard

less

of

gen

der

(OR

1.4

, 95%

CI 1

.3 t

o 1.

6).

3. D

iagn

osis

with

oth

er m

enta

l illn

ess

(eg,

dep

ress

ion,

anx

iety

) alo

ne o

r in

co

mb

inat

ion

with

ser

ious

men

tal i

llnes

s w

as a

ssoc

iate

d w

ith h

ighe

r ris

k of

p

oten

tially

IP fo

r w

omen

(OR

1.3

, 95%

CI 1

.1 t

o 1.

5).

Pot

entia

lly IP

p

reva

lenc

e: 2

14 8

87/1

09

6 36

1=19

.6%

24.

Saa

b e

t al

, 200

643Le

ban

onD

escr

iptiv

e27

7 el

der

ly a

ged

≥65

yea

rs

from

10

com

mun

ity

pha

rmac

ies

Pre

scrip

tion

and

/or

OTC

med

icat

ions

IDU

(Bee

rs c

riter

ia, m

issi

ng

dos

es, i

nap

pro

pria

te fr

eque

ncy

of a

dm

inis

trat

ion,

poo

r m

emor

y,

dru

g–d

isea

se in

tera

ctio

n, D

DI,

inap

pro

pria

te d

ose,

dup

licat

ed

ther

apy,

dis

cont

inua

tion

of

ther

apy,

ad

vers

e ef

fect

and

in

app

rop

riate

ind

icat

ion)

.Fa

ctor

s th

at p

red

ict

pot

entia

lly

inap

pro

pria

te d

rug

inta

ke.

Rev

iew

pat

ient

pro

file

usin

g co

mm

unity

pha

rmac

y d

ata

and

in

per

son

inte

rvie

ws.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: T

he p

reva

lenc

e of

eld

erly

out

pat

ient

with

at

leas

t on

e in

app

rop

riate

med

icat

ion:

165

/277

(59.

6%) (

incl

ude

five

pat

ient

s w

ith A

DR

).In

app

rop

riate

med

icat

ion

use

was

mos

t fr

eque

ntly

iden

tified

in t

erm

s of

B

eers

crit

eria

(22.

4%),

mis

sing

dos

es (1

8.8%

) and

inco

rrec

t fr

eque

ncy

of

adm

inis

trat

ion

(13%

).D

rug–

dis

ease

inte

ract

ion

in 2

8 p

atie

nts

(10.

1%),

DD

I 14

(5.1

%),

dup

licat

e th

erap

y 12

(4.3

%).

Ris

k fa

ctor

s:Fe

mal

e se

x (6

5.7%

vs

53.3

% fo

r m

ale,

p=

0.03

).Th

ere

wer

e al

so s

igni

fican

t as

soci

atio

ns b

etw

een

the

likel

ihoo

d o

f use

of a

n in

app

rop

riate

dru

g an

d (1

) inc

reas

ed n

umb

er o

f med

ical

illn

esse

s (p

<0.

0000

2);

(2) c

onsu

mp

tion

of a

n O

TC d

rug

and

/or

pre

scrip

tion

dru

g (p

=0.

048

and

p

=0.

0035

, res

pec

tivel

y); a

nd (3

) con

sum

ptio

n of

bot

h O

TC a

nd p

resc

riptio

n d

rugs

(p<

0.00

02).

IDU

pre

vale

nce:

62

/277

=22

.4%

usi

ng

Bee

rs c

riter

ia

Just

ext

ract

ed t

he

IDU

by

Bee

rs c

riter

ia

bec

ause

the

IDU

in

clud

es 5

cas

es o

f AD

R

and

som

e p

atie

nts

had

m

ore

than

one

IDU

.R

isk

fact

ors

for

all

typ

es o

fID

U.

25.

Zuc

kerm

an e

t al

, 20

0644

US

AC

ohor

t re

tros

pec

tive

487

383

com

mun

ity-

dw

elle

r el

der

ly a

ged

≥6

5 ye

ars.

Dat

a fr

om M

arke

tSca

n M

edic

are

Sup

ple

men

tal

and

Coo

rdin

atio

n of

B

enefi

ts d

atab

ase.

Pre

scrib

ed d

rug

Inap

pro

pria

te m

edic

atio

n us

e us

ing

Bee

rs c

riter

iaP

resc

ribin

g er

ror:

PIM

pre

vale

nce:

204

083

eld

erly

use

d in

app

rop

riate

med

icat

ion.

Use

of i

nap

pro

pria

te d

rugs

was

ass

ocia

ted

with

a 3

1% in

crea

se in

ris

k of

nu

rsin

g ho

me

adm

issi

on, c

omp

ared

with

no

use

of in

app

rop

riate

dru

gs

(ad

just

ed r

elat

ive

risk

1.31

, 99%

CI 1

.26

to 1

.36)

.

Inap

pro

pria

te m

edic

atio

n us

e p

reva

lenc

e: 2

04

083/

487

383=

41.9

%

26.

Bre

gnhø

j et

al, 2

00745

Cop

enha

gen,

Den

mar

kC

ross

-sec

tiona

l21

2 el

der

ly a

ged

≥65

yea

rs

with

pol

ypha

rmac

y (≥

5 d

rugs

) pat

ient

s fr

om

prim

ary

care

Sub

sid

ised

and

no

n-su

bsi

dis

ed

med

icat

ions

p

resc

ribed

IP m

easu

red

by

the

MA

I: 10

crit

eria

are

ind

icat

ion,

ef

fect

iven

ess,

dos

age,

dire

ctio

ns

pra

ctic

ality

, dire

ctio

ns c

orre

ctne

ss,

DD

I, d

rug–

dis

ease

inte

ract

ion,

d

uplic

atio

n, d

urat

ion

and

ex

pen

se).

Pat

ient

s ex

pos

ed t

o p

olyp

harm

acy

wer

e id

entifi

ed v

ia t

he d

atab

ase

reco

rdin

g th

e d

rug

sub

sid

y sy

stem

of D

anis

h p

harm

acie

s an

d

que

stio

nnai

re.

Pre

scrib

ing

erro

r:IP

pre

vale

nce:

The

maj

ority

of t

he p

atie

nts,

nam

ely

94.3

%, h

ad o

ne o

r m

ore

inap

pro

pria

te r

atin

gs a

mon

g th

eir

med

icat

ions

.

IP p

reva

lenc

e:

200/

212=

94.3

%

27.

John

ell a

nd F

astb

om,

2008

46S

wed

enC

ross

-sec

tiona

l73

1 10

5 p

eop

le a

ged

≥7

5 ye

ars

from

the

S

wed

ish

Pre

scrib

ed D

rug

Reg

iste

r (s

econ

dar

y d

ata

anal

ysis

)

Pre

scrib

ed d

rug

only

and

mul

tidos

e d

rug

dis

pen

sing

Whe

ther

the

use

of m

ultid

ose

dru

g d

isp

ensi

ng is

ass

ocia

ted

w

ith p

oten

tial I

DU

(ie,

an

ticho

liner

gic

dru

gs, l

ong-

actin

g b

enzo

dia

zep

ines

, con

curr

ent

use

of ≥

3 p

sych

otro

pic

dru

gs a

nd

com

bin

atio

ns o

f dru

gs t

hat

may

le

ad t

o p

oten

tially

ser

ious

DD

Is).

Info

rmat

ion

from

the

Sw

edis

h P

resc

ribed

Dru

g R

egis

ter.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: P

reva

lenc

e of

pot

entia

l ID

U in

mul

tidos

e d

isp

ensi

ng u

sers

: 40

.3%

(wom

en: 4

1%, m

en: 3

8.5%

).P

reva

lenc

e of

pot

entia

l ID

U In

pre

scrip

tion

user

s: 1

3.6%

(wom

en: 1

5%, m

en:

11.5

%).

The

mul

tidos

e us

ers

had

hig

her

pre

vale

nce

of a

ll in

dic

ator

s of

pot

entia

l in

app

rop

riate

dru

g th

an p

resc

riptio

n us

ers.

1. T

he y

oung

er e

lder

ly (a

ged

75–

79 y

ears

) who

use

d m

ultid

ose

dru

g d

isp

ensi

ng

had

the

hig

hest

freq

uenc

y of

all

ind

icat

ors

of p

oten

tial I

DU

.2.

Mos

t in

dic

ator

s of

IDU

wer

e m

ore

com

mon

in w

omen

tha

n m

en.

3. M

ultid

ose

dru

g d

isp

ensi

ng a

mon

g th

ose

aged

75–

79 y

ears

old

was

eve

n m

ore

stro

ngly

ass

ocia

ted

with

any

IDU

, ant

icho

liner

gic

dru

gs, t

hree

or

mor

e p

sych

otro

pic

dru

gs in

bot

h m

en a

nd w

omen

, and

long

-act

ing

ben

zod

iaze

pin

es

amon

g m

en.

PIM

pre

vale

nce:

Mul

tidos

e d

isp

ensi

ng

user

s: 2

92 7

37/7

31

105=

40%

Pre

scrip

tion

user

s: 9

9 43

0.3/

731

105=

13.6

%

Mul

tidos

e d

rug

dis

pen

sing

mea

ns t

hat

pat

ient

s ge

t th

eir

dru

gs

mac

hine

-dis

pen

sed

into

on

e un

it fo

r ea

ch d

ose

occa

sion

and

pac

ked

in

dis

pos

able

bag

s.

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

10 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

28.

Ber

dot

et

al, 2

00947

Dijo

n, B

ord

eaux

, M

ontp

ellie

r, Fr

ance

Coh

ort

pro

spec

tive

6343

com

mun

ity-d

wel

ling

eld

erly

age

d ≥

65 y

ears

Pre

scrib

ed d

rug

PIM

usi

ng 1

997

and

200

3 B

eers

cr

iteria

, Fic

k an

d L

aroc

he.

Face

-to-

face

inte

rvie

w u

sing

st

and

ard

ised

que

stio

nnai

re.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: O

ne-t

hird

(31.

6%) o

f the

stu

dy

par

ticip

ants

rep

orte

d u

sing

at

leas

t on

e in

app

rop

riate

med

icat

ion

at s

tud

y en

try.

PIM

pre

vale

nce:

20

04/6

343=

31.6

%,

p<

0.00

1

29.

Hai

der

et

al, 2

00948

Sw

eden

Cro

ss-s

ectio

nal,

regi

ster

-b

ased

stu

dy

626

258

old

er p

eop

le

aged

75–

89 y

ears

from

the

S

wed

ish

Pre

scrib

ed D

rug

Reg

iste

r (s

econ

dar

y d

ata

anal

ysis

)

Pre

scrib

ed d

rug

only

If lo

w e

duc

atio

n as

soci

ated

w

ith p

oten

tial I

DU

(ie,

an

ticho

liner

gic

dru

gs, l

ong-

actin

g b

enzo

dia

zep

ines

, con

curr

ent

use

of ≥

3 p

sych

otro

pic

dru

gs a

nd

clin

ical

ly r

elev

ant

pot

entia

l DD

I).In

form

atio

n fr

om t

he S

wed

ish

Pre

scrib

ed D

rug

Reg

iste

r.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: T

he p

rop

ortio

n of

par

ticip

ants

rep

ortin

g us

e of

at

leas

t on

e p

oten

tial I

DU

was

34.

6%.

Ris

k fa

ctor

s:S

ubje

cts

with

low

ed

ucat

ion

had

a h

ighe

r p

rob

abili

ty o

f pot

entia

l ID

U (O

R 1

.09,

95

% C

I 1.0

7 to

1.1

7).

Old

er a

ge, b

eing

a w

oman

and

hig

her

CC

I wer

e as

soci

ated

with

the

hig

hest

fr

eque

ncie

s of

pot

entia

l ID

U.

IDU

pre

vale

nce:

216

68

5/62

6 25

8=34

.6%

30.

Lai e

t al

, 200

949Ta

iwan

Des

crip

tive

2 13

3 86

4 p

atie

nts

aged

≥6

5 ye

ars

bet

wee

n 20

01

and

200

4 fr

om a

mb

ulat

ory

care

Nat

iona

l Hea

lth

Insu

ranc

e cl

aim

dat

abas

e

Pre

scrib

ed d

rug

PIM

pre

scrib

ing

usin

g up

dat

ed

2003

Bee

rs c

riter

ia a

nd t

he

char

acte

ristic

s of

and

ris

k fa

ctor

s fo

r su

ch p

resc

ribin

g

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: A

mea

n of

63.

8% o

f the

old

er p

opul

atio

n re

ceiv

ed a

PIM

at

leas

t on

ce a

yea

r in

200

1–20

04.

Det

ails

:20

01: 1

974

869

pat

ient

s of

who

m 1

297

425

had

inap

pro

pria

te p

resc

riptio

n (6

5.7)

.20

02: 2

026

737

pat

ient

s of

who

m 1

312

147

had

inap

pro

pria

te p

resc

riptio

n (6

4.7)

.20

03: 2

077

677

pat

ient

s of

who

m 1

295

227

had

inap

pro

pria

te p

resc

riptio

n (6

2.3)

.20

04: 2

133

864

pat

ient

s of

who

m 1

333

792

had

IP (6

2.5)

.R

isk

fact

ors:

The

only

pat

ient

cha

ract

eris

tic a

ssoc

iate

d w

ith a

n in

crea

sed

like

lihoo

d o

f the

p

resc

ribin

g of

PIM

was

fem

ale

sex

(mal

e se

x: O

R 0

.982

(95%

CI 0

.980

to

0.98

3)) (

p<

0.00

1) a

nd w

hen

≥4 d

rugs

wer

e p

resc

ribed

(p<

0.00

1).

The

follo

win

g ar

e p

hysi

cian

cha

ract

eris

tics

asso

ciat

ed w

ith a

gre

ater

like

lihoo

d

of t

he p

resc

ribin

g of

PIM

:1.

Mal

e se

x (O

R 1

.206

, 95%

CI 1

.202

to

1.21

0, p

<0.

001)

.2.

Old

er a

ge (4

3–50

yea

rs: O

R 1

.021

, 95%

CI 1

.018

to

1.02

5, p

<0.

001;

≥5

1 ye

ars:

OR

1.2

38, 9

5% C

I 1.2

35 t

o 1.

242,

p<

0.00

1).

3. F

amily

med

icin

e/ge

nera

l pra

ctic

e (O

R 1

.267

, 95%

CI 1

.265

to

1.26

9, p

<0.

001)

.

PIM

pr e

vale

nce:

2001

: 65.

7%20

02: 6

4.7%

2003

: 62.

3%20

04: 1

333

792

/2 1

33

864=

62.5

%

31.

Rya

n et

al,

2009

50Ire

land

Coh

ort

pro

spec

tive

500

pat

ient

s ag

ed

≥65

year

s fr

om p

rimar

y ca

re

Pre

scrib

ed d

rug

IP u

sing

200

3 B

eers

crit

eria

an

d IP

ET.

Scr

eeni

ng p

atie

nts’

med

ical

re

cord

s (e

lect

roni

c an

d p

aper

).

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 6

5 p

atie

nts

(13%

) and

52

pat

ient

s (1

0.4%

) had

at

leas

t on

e m

edic

ine

pre

scrib

ed in

app

rop

riate

ly u

sing

200

3 B

eers

and

IPE

T cr

iteria

, re

spec

tivel

y.

IP p

reva

lenc

e:B

eers

200

3:

65/5

00=

13%

IPE

T: 5

2/50

0=10

.4%

32.

Rya

n et

al,

2009

51C

ork,

Sou

ther

n Ire

land

Des

crip

tive

case

rec

ord

re

view

1329

eld

erly

age

d

≥65

year

s fr

om p

rimar

y ca

re

Pre

scrib

ed d

rugs

A—

1. P

IM u

sing

200

3 B

eers

and

S

TOP

P c

riter

ia.

2. P

PO

usi

ng S

TAR

T cr

iteria

.B

—R

elat

ions

hip

bet

wee

n ag

e an

d n

umb

er o

f pre

scrip

tion

dru

gs

and

IP.

Cas

e re

cord

thr

ough

pap

er a

nd

elec

tron

ic r

ecor

d r

evie

w.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: IP

rat

e id

entifi

ed b

y B

eers

crit

eria

in 1

8.3%

(243

) of p

atie

nts.

IP r

ate

iden

tified

by

STO

PP

was

21.

4% (2

84).

PP

O w

as id

entifi

ed in

22.

7% (3

02) o

f pat

ient

s us

ing

the

STA

RT

tool

.R

isk

fact

ors:

A s

igni

fican

t co

rrel

atio

n w

as fo

und

bet

wee

n th

e oc

curr

ence

of P

IM a

nd t

he

follo

win

g:1.

The

num

ber

of m

edic

ines

pre

scrib

ed w

hen

calc

ulat

ed u

sing

Bee

rs c

riter

ia

(rs=

0.27

0, p

<0.

01) a

nd S

TOP

P (r

s=0.

356,

p<

0.01

) usi

ng S

pea

rman

’s ρ

cor

rela

tion

test

.2.

Age

usi

ng B

eers

crit

eria

(rs=

0.06

8, p

<0.

01) a

nd S

TOP

P (r

s=0.

071,

p<

0.01

).3.

Incr

easi

ng C

CI s

core

iden

tified

by

STO

PP

(rs=

0.21

0, p

<0.

01).

PIM

pre

vale

nce:

Bee

rs: 2

43/1

329=

18.3

%S

TOP

P:

284/

1329

=21

.4%

PP

O p

reva

lenc

e:S

TAR

T: 3

02/1

329=

22.7

%

Sp

earm

an’s

ρ c

orre

latio

nte

st

33.

Aka

zaw

a et

al,

2010

52To

kyo,

Jap

anC

ohor

t re

tros

pec

tive

6628

eld

erly

pat

ient

s ag

ed

≥65

year

s fr

om h

ealth

in

sura

nce

clai

m d

ata

(sec

ond

ary

dat

a an

alys

is)

Pre

scrib

ed d

rugs

PIM

usi

ng m

odifi

ed B

eers

crit

eria

in

Jap

an.

Dru

g ut

ilisa

tion

revi

ew u

sing

m

edic

al a

nd p

harm

acy

clai

m fr

om

dat

abas

e of

Jap

an M

edic

al D

ata

Cen

ter.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 4

3.6%

(288

9/66

28) w

ere

pre

scrib

ed a

t le

ast

one

PIM

.R

isk

fact

ors:

Fact

ors

pos

itive

ly a

ssoc

iate

d w

ith P

IM p

resc

riptio

ns a

t a

sign

ifica

nce

leve

l of

5%

incl

uded

the

follo

win

g: h

osp

ital a

dm

issi

on (O

R=

3.35

, 95%

CI 2

.43

to 4

.62)

; pol

ypha

rmac

y (O

R=

5.69

, 95%

CI 5

to

6.48

); p

resc

riptio

ns fr

om a

ho

spita

l (O

R=

1.19

), ge

nera

l med

icin

e p

ract

ition

er (O

R=

1.46

) or

psy

chia

tris

t/ne

urol

ogis

t (O

R=

2.33

); an

d c

omor

bid

con

diti

ons

incl

udin

g p

eptic

ulc

er d

isea

se

with

out

ble

edin

g (O

R=

4.18

, 95%

CI 3

.52

to 4

.97)

, dep

ress

ion

(OR

=3.

69),

card

iac

arrh

ythm

ias

(OR

=1.

93),

othe

r ne

urol

ogic

al d

isor

der

s (P

arki

nson

’s

dis

ease

, mul

tiple

scl

eros

is a

nd e

pile

psy

; OR

=1.

88) a

nd c

onge

stiv

e he

art

failu

re

(OR

=1.

46).

PIM

use

rs h

ad s

igni

fican

tly h

ighe

r ho

spita

lisat

ion

risk

(1.6

8-fo

ld),

mor

e ou

tpat

ient

vis

it d

ays

(1.1

8-fo

ld) a

nd h

ighe

r m

edic

al c

osts

(33%

incr

ease

) tha

n d

id n

on-u

sers

.

PIM

pre

vale

nce:

28

89/6

628=

43.6

%*C

onse

que

nce

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

11Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

34.

Zav

eri e

t al

, 201

053A

hmed

abad

city

, Ind

iaD

escr

iptiv

e p

rosp

ectiv

e40

7 ge

riatr

ic p

atie

nts

aged

≥6

5 ye

ars

from

med

icin

e ou

tpat

ient

dep

artm

ent

Pre

scrib

ed d

rug

PIM

usi

ng 2

003

Bee

rs c

riter

ia.

Usi

ng p

rosp

ectiv

e p

rofo

rma

dat

a co

llect

ion.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: O

ut o

f 407

pat

ient

s, 9

6 p

atie

nts

(23.

6%) r

ecei

ved

at

leas

t on

e d

rug

that

was

pot

entia

lly in

app

rop

riate

.R

isk

fact

ors:

Ther

e w

as h

ighl

y si

gnifi

cant

ass

ocia

tion

bet

wee

n th

e nu

mb

er o

f dru

gs

pre

scrib

ed a

nd fr

eque

ncy

of u

se o

f PIM

s (p

<0.

0002

).

PIM

pre

vale

nce:

96

/407

=23

.6%

35.

Bar

nett

et

al, 2

01154

Tays

ide,

Sco

tland

, UK

Coh

ort

65 7

42 e

lder

ly a

ged

66

–99

year

s liv

ing

in h

ome

Pre

scrib

ed d

rug

PIM

usi

ng 2

003

Bee

rs c

riter

ia a

nd

the

asso

ciat

ion

bet

wee

n ex

pos

ure

to P

IM a

nd m

orta

lity.

Usi

ng d

isp

ensi

ng a

nd p

resc

ribin

g d

atab

ase

and

med

ical

rec

ord

.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: P

IM fo

und

in 2

0 30

4 (3

0.9%

) pat

ient

s liv

ing

at h

ome.

Ris

k fa

ctor

s:A

fter

ad

just

men

t fo

r ag

e, s

ex a

nd p

olyp

harm

acy,

1. P

atie

nt a

t in

crea

sed

ris

k of

rec

eivi

ng a

t le

ast

one

PIM

if t

hey

wer

e yo

unge

r, fe

mal

e an

d h

ad h

ighe

r p

olyp

harm

acy.

2. R

ecei

ving

at

leas

t on

e P

IM w

as n

ot a

ssoc

iate

d w

ith in

crea

sed

ris

k of

m

orta

lity

(ad

just

ed O

R 0

.98,

95%

CI 0

.92

to 1

.05)

.

PIM

pre

vale

nce:

20

304/

65 7

42=

30.9

%R

isk

fact

ors

for

bot

h ca

reho

me

and

hom

e

36.

Cha

ng e

t al

, 201

155Ta

ipei

, Tai

wan

Coh

ort

193

outp

atie

nt e

lder

ly

pat

ient

s ag

ed ≥

65 y

ears

w

ith p

olyp

harm

acy

(≥8

chro

nic

med

icat

ions

) fr

om M

edic

atio

n S

afet

y R

evie

w C

linic

in T

aiw

anes

e E

lder

s (M

SR

C-T

aiw

an)

stud

y

Pre

scrib

ed d

rugs

an

d d

ieta

ry

sup

ple

men

t ex

clud

ing

herb

als

PIM

usi

ng s

ix d

iffer

ent

crite

ria

and

dru

g-re

late

d p

rob

lem

: the

20

03 v

ersi

on o

f the

Bee

rs c

riter

ia

(from

the

US

A),

the

Ran

cour

t (fr

om C

anad

a), t

he L

aroc

he (f

rom

Fr

ance

), S

TOP

P (f

rom

Irel

and

), th

e W

init-

Wat

jana

(fro

m T

haila

nd)

and

the

NO

RG

EP

crit

eria

(fro

m

Nor

way

).A

naly

se b

asel

ine

dat

a fr

om t

he

MS

RC

-Tai

wan

stu

dy.

Sec

ond

ary

dat

a an

alys

is.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: T

he p

rop

ortio

n of

pat

ient

s w

ho h

ad a

t le

ast

one

PIM

var

ied

fr

om 2

4% (t

he N

OR

GE

P c

riter

ia) t

o 73

% (t

he W

init-

Wat

jana

crit

eria

).A

pp

roxi

mat

ely

31%

(the

STO

PP

crit

eria

) to

42%

(the

NO

RG

EP

crit

eria

) of P

IMs

iden

tified

wer

e co

nsid

ered

as

dru

g-re

late

d p

rob

lem

s b

y th

e m

edic

atio

n re

view

te

am e

xper

ts.

Ris

k fa

ctor

s:In

the

biv

aria

te a

naly

sis,

the

com

mon

cha

ract

eris

tics

asso

ciat

ed w

ith h

avin

g at

le

ast

one

PIM

in a

ll cr

iteria

wer

e a

high

num

ber

of c

hron

ic c

ond

ition

s an

d a

hig

h nu

mb

er o

f chr

onic

med

icat

ions

.

PIM

pre

vale

nce:

24

%–7

3%

37.

Leik

ola

et a

l, 20

1156

Finl

and

Cro

ss-s

ectio

nal

841

509

non-

inst

itutio

nalis

ed e

lder

ly

pat

ient

s ag

ed ≥

65 y

ears

fr

om F

inla

nd’s

Soc

ial

Insu

ranc

e In

stitu

tion

pre

scrip

tion

regi

ster

of

all r

eim

bur

sed

dru

gs fo

r ou

tpat

ient

s

Pre

scrib

ed a

nd O

TC

med

icat

ions

tha

t ar

e re

imb

urse

d

PIM

usi

ng 2

003

Bee

rs c

riter

iaP

resc

ribin

g er

ror:

PIM

pre

vale

nce:

14.

7% (n

=12

3 54

5) h

ad r

ecei

ved

PIM

s ac

cord

ing

to t

he B

eers

20

03 c

riter

ia.

PIM

pre

vale

nce:

123

54

5/84

1 50

9=14

.7%

38.

Lin

et a

l, 20

1157

Taiw

anC

ross

-sec

tiona

l, re

tros

pec

tive

anal

ysis

327

eld

erly

pat

ient

s ag

ed

≥65

year

s fr

om o

utp

atie

nt

clin

ic o

f a c

omm

unity

he

alth

cen

tre

Pre

scrib

ed d

rugs

PIM

usi

ng 2

003

Bee

rs c

riter

ia a

nd

risk

fact

ors

of P

IM u

se.

Usi

ng d

ata

revi

ew.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: T

he p

reva

lenc

e of

pat

ient

s ha

ving

at

leas

t on

e P

IM w

as 2

7.5%

(9

0/32

7).

Ris

k fa

ctor

s:In

dep

end

ent

risk

fact

ors

for

PIM

s ar

e ol

der

age

(OR

=1.

05, 9

5% C

I 1.0

0 to

1.0

9,

p=

0.04

6), h

ighe

r nu

mb

er o

f pre

scrib

ed m

edic

atio

ns (O

R=

1.06

, 95%

CI 1

.39

to

1.98

, p<

0.00

1) a

nd d

iagn

osis

of a

cute

dis

ease

s (O

R=

8.98

, 95%

CI 4

.71

to 1

7.1,

p

<0.

001)

.

PIM

pre

vale

nce:

90

/327

=27

.5%

39.

Woe

lfel e

t al

, 201

170C

alifo

rnia

, US

AC

ross

-sec

tiona

l29

5 el

der

ly a

ged

≥6

5 ye

ars

from

am

bul

ator

y p

opul

atio

n of

Med

icar

e b

enefi

ciar

ies

Pre

scrib

ed d

rug

PIM

usi

ng 2

003

Bee

rs c

riter

ia.

Usi

ng m

edic

atio

n re

view

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 5

4 (1

8.3%

ben

efici

arie

s w

ere

taki

ng a

t le

ast

one

PIM

.R

isk

fact

ors:

The

num

ber

of m

edic

atio

ns w

as s

igni

fican

tly g

reat

er in

the

PIM

tha

n th

e no

n-P

IM g

roup

(p<

0.00

1).

PIM

pre

vale

nce:

54

/295

=18

.3%

40.

Zha

ng e

t al

, 201

158U

SA

Coh

ort

retr

osp

ectiv

e35

70 e

lder

ly c

omm

unity

-b

ased

res

pon

den

ts a

ged

≥6

5 fr

om 2

007

ME

PS

, a

natio

nally

rep

rese

ntat

ive

surv

ey o

f the

US

co

mm

unity

-dw

ellin

g p

opul

atio

n

Pre

scrib

ed d

rug

PIM

usi

ng Z

han

crite

ria a

nd r

isk

fact

ors

for

PIM

use

.In

form

atio

n fr

om M

EP

S d

atab

ase.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: P

IM p

reva

lenc

e in

200

7: 1

3.84

% (C

I 12.

52 t

o 15

.17)

.P

IM p

reva

lenc

e in

199

6: 2

1.3%

(CI 1

9.5

to 2

3.1)

.R

isk

fact

ors:

Old

er w

omen

, peo

ple

tak

ing

≥25

pre

scrip

tions

, peo

ple

with

mid

dle

fam

ily

inco

me,

peo

ple

livi

ng in

the

Sou

th c

ensu

s re

gion

, and

peo

ple

who

sai

d t

hey

wer

e in

fair

or p

oor

heal

th w

ere

mor

e lik

ely

to h

ave

rece

ived

an

inap

pro

pria

te

med

icat

ion

dur

ing

the

year

.

PIM

pre

vale

nce:

13

.84%

–21.

3%

41.

Haa

sum

et

al, 2

01259

Sw

eden

Cro

ss-s

ectio

nal,

retr

osp

ectiv

e1

260

843

hom

e-d

wel

ling

eld

erly

age

d ≥

65 y

ears

fr

om t

he S

wed

ish

Pre

scrib

ed D

rug

Reg

iste

r

Pre

scrib

ed d

rug

only

Pot

entia

lly ID

U (u

se o

f an

ticho

liner

gic

dru

gs, l

ong-

actin

g b

enzo

dia

zep

ines

, con

curr

ent

use

of ≥

3 p

sych

otro

pic

s an

d p

oten

tially

se

rious

DD

Is).

Info

rmat

ion

from

the

Sw

edis

h P

resc

ribed

Dru

g R

egis

ter.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 1

1.6%

of t

he h

ome-

dw

ellin

g el

der

ly w

ere

exp

osed

to

pot

entia

lly ID

U.

Pot

entia

lly ID

U

pre

vale

nce:

145

749

/1

260

843=

11.

6%

Info

rmat

ion

on b

oth

inst

itutio

nalis

ed a

ndho

me-

dw

ellin

g.E

xtra

cted

hom

e-d

wel

ling

info

rmat

ion

only

.

42.

Can

del

a M

arro

quí

et

al, 2

01219

Các

eres

, Sp

ain

Des

crip

tive

471

pat

ient

s ag

ed

≥65

year

s fr

om h

ealth

ce

ntre

s

Con

sum

ed

med

icat

ions

Pot

entia

lly IP

usi

ng S

TOP

P/S

TAR

T cr

iteria

.U

sing

pat

ient

inte

rvie

w a

nd

med

ical

cha

rt r

evie

w.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 2

49 p

atie

nts

(52.

8%, 9

5% C

I 48.

3 to

57.

3) h

ad p

oten

tially

IP

acco

rdin

g to

STO

PP

/STA

RT

crite

ria.

STO

PP

: 162

pat

ient

s (3

4.3%

, 95%

CI 3

0.2%

to

38.8

%).

STA

RT:

114

pat

ient

s (2

4.2%

, 95%

CI 2

0.5%

to

28.2

%).

Pot

entia

lly IP

pre

vale

nce:

24

9/47

1=52

.8%

(95%

CI

48.3

to

57.3

)

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

12 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

43.

Nyb

org

et a

l, 20

1260

Nor

way

Cro

ss-s

ectio

nal,

retr

osp

ectiv

e44

5 90

0 ho

me-

dw

ellin

g el

der

ly a

ged

≥70

yea

rs

from

the

Nor

weg

ian

Pre

scrip

tion

Dat

abas

e

Pre

scrib

ed d

rug

Pre

vale

nce

of a

nd p

red

icto

rs fo

r P

IM u

se b

y th

e N

OR

GE

P c

riter

ia.

Sur

vey

und

erta

ken

bas

ed o

n d

ata

from

the

Nor

weg

ian

Pre

scrip

tion

Dat

abas

e.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 3

4.8%

of t

he s

tud

y p

opul

atio

n w

as e

xpos

ed t

o at

leas

t on

e P

IM.

Ris

k fa

ctor

s:Th

e od

ds

of r

ecei

ving

pot

entia

lly h

arm

ful p

resc

riptio

ns in

crea

sed

with

the

nu

mb

er o

f doc

tors

invo

lved

in p

resc

ribin

g (O

R 3

.52,

99%

CI 3

.44

to 3

.60

for

thos

e w

ith ≥

5 co

mp

ared

with

tho

se w

ith 1

or

two

pre

scrib

ers)

.W

omen

wer

e at

hig

her

risk

for

PIM

s (O

R 1

.6, 9

9% C

I 1.5

8 to

1.6

4).

PIM

pr e

vale

nce:

155

34

1/44

5 90

0= 3

4.8%

(9

9% C

I34

.7 t

o 35

)

44.

Yase

in e

t al

, 201

261Jo

rdan

Cro

ss-s

ectio

nal

400

eld

erly

age

d ≥

65 y

ears

fr

om fa

mily

pra

ctic

e cl

inic

Pre

scrib

ed d

rug

Pol

ypha

rmac

y (≥

5 d

rugs

) and

IP

usin

g 20

03 B

eers

crit

eria

.U

sing

pat

ient

file

and

pat

ient

in

terv

iew

.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: In

app

rop

riate

med

icat

ions

as

det

erm

ined

by

Bee

rs c

riter

ia

ind

epen

den

t of

dia

gnos

is a

ccou

nted

for

118

(29.

5%) p

atie

nts.

IP p

reva

lenc

e:

118/

400=

29.5

%

45.

Blo

zik

et a

l, 20

1362

Hel

sana

, Sw

itzer

land

Coh

ort

2008

: 1 0

59 4

9520

09: 1

047

939

2010

: 929

791

Com

mun

ity-d

wel

ling

adul

ts

aged

>18

yea

rs fr

om c

laim

d

ata

of H

elsa

na

Pre

scrib

ed d

rug

sub

mitt

ed fo

r re

imb

urse

men

t

Pre

vale

nce

of p

olyp

harm

acy

and

P

IM u

sing

200

3 B

eers

crit

eria

or

the

PR

ISC

US

list

.U

sing

ana

lysi

s of

dat

a b

ased

on

clai

m d

ata

from

Sw

itzer

land

hea

lth

insu

ranc

e.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: A

ccor

din

g to

200

3 B

eers

crit

eria

, 10.

3% o

f the

com

mun

ity-

dw

ellin

g p

opul

atio

n ag

ed >

65 y

ears

rec

eive

d a

t le

ast

one

med

icat

ion

whi

ch is

P

IM, a

nd a

ccor

din

g to

the

PR

ISC

US

list

1, 1

6.0%

of p

erso

ns h

ad a

PIM

.W

hen

usin

g b

oth

Bee

rs a

nd P

RIS

CU

S c

riter

ia, 2

1.1%

of t

he p

opul

atio

n re

ceiv

ed

at le

ast

one

PIM

.O

f tho

se p

erso

ns o

lder

tha

n 65

yea

rs a

skin

g fo

r re

imb

urse

men

t of

med

icat

ions

, 12

.9%

rec

eive

d a

t le

ast

one

PIM

acc

ord

ing

to 2

003

Bee

rs, 2

0.2%

acc

ord

ing

to

PR

ISC

US

and

26.

6% o

f eith

er d

efini

tion.

Ris

k fa

ctor

s:W

omen

wer

e m

ore

likel

y to

rec

eive

a P

IM: 2

5.5%

of w

omen

as

com

par

ed w

ith

15.4

% o

f men

whe

n b

oth

Bee

rs a

nd P

RIS

CU

S d

efini

tions

wer

e us

ed.

PIM

pre

vale

nce:

21.

1%Th

ere

are

huge

d

iscr

epan

cies

in

estim

atin

g th

e p

reva

lenc

e of

PIM

d

epen

din

g on

the

d

efini

tion

used

.

46.

Cah

ir et

al,

2013

63Ire

land

Coh

ort

retr

osp

ectiv

e93

1 co

mm

unity

-dw

ellin

g el

der

ly a

ged

≥70

yea

rs

from

15

gene

ral p

ract

ices

Pre

scrib

ed d

rug

and

OTC

The

asso

ciat

ion

bet

wee

n p

oten

tially

IP u

sing

STO

PP

and

he

alth

-rel

ated

out

com

es (A

DE

s,

HR

QO

L, a

nd h

osp

ital a

ccid

ent

and

ED

).U

sing

pat

ient

sel

f-re

por

t an

d

med

ical

rec

ord

.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: P

reva

lenc

e of

pot

entia

lly IP

was

40.

5% (n

=37

7).

AD

E p

reva

lenc

e: In

tot

al, 6

74 o

f 859

par

ticip

ants

(78%

) wer

e cl

assi

fied

as

havi

ng a

t le

ast

one

AD

E d

urin

g th

e st

udy

per

iod

.R

isk

fact

ors:

Pat

ient

s w

ith ≥

2 p

oten

tially

IP in

dic

ator

s w

ere:

1. T

wic

e as

like

ly t

o ha

ve a

n A

DE

(ad

just

ed O

R 2

.21,

95%

CI 1

.02

to 4

.83,

p

<0.

05).

2. S

igni

fican

tly lo

wer

mea

n H

RQ

OL

utili

ty (a

dju

sted

coe

ffici

ent

−0.

09, S

E 0

.02,

p

<0.

001)

.3.

A t

wof

old

incr

ease

d r

isk

in t

he e

xpec

ted

rat

e of

ED

vis

its (a

dju

sted

inci

den

ce

rate

rat

io 1

.85,

95%

CI 1

.32

to 2

.58,

p<

0.00

1).

Pot

entia

lly IP

pre

vale

nce:

37

7/93

1=40

.5%

AD

E p

reva

lenc

e:

674/

859=

78%

*Con

seq

uenc

e.Ty

pe

of A

DE

was

not

men

tione

d.

47.

Wen

g et

al,

2013

64Ta

iwan

Cro

ss-s

ectio

nal,

retr

osp

ectiv

e78

0 ol

der

pat

ient

s ag

ed

≥65

year

s fr

om t

he

outp

atie

nt g

eria

tric

clin

ic

Long

-ter

m

pre

scrib

ed d

rugs

(≥

28 d

ays)

for

chro

nic

dis

ease

s,

not

OTC

Imp

act

of n

umb

er o

f dru

gs

pre

scrib

ed o

n th

e ris

k of

PIM

usi

ng

STO

PP

crit

eria

.P

atie

nt m

edic

al c

hart

rev

iew

.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 3

02 p

atie

nts

(39%

) had

at

leas

t on

e P

IM.

Ris

k fa

ctor

s:M

ultiv

aria

te a

naly

sis

reve

aled

tha

t P

IM r

isk

was

ass

ocia

ted

with

the

num

ber

of

med

icat

ions

pre

scrib

ed (p

<0.

001)

and

the

pre

senc

e of

car

dio

vasc

ular

(p<

0.00

1)

or g

astr

oint

estin

al d

isea

se (p

=0.

003)

.P

atie

nts

pre

scrib

ed ≥

5 d

rugs

(ad

just

ed O

R=

5.4)

had

sig

nific

antly

hig

her

PIM

ris

k th

an t

hose

pre

scrib

ed ≤

2 d

rugs

.

PIM

pre

vale

nce:

30

2/78

0=39

%

48.

Zim

mer

man

n et

al,

2013

18G

erm

anC

ohor

t lo

ngitu

din

al

anal

ysis

Follo

w-u

p 3

: n=

1942

Bas

elin

e: n

=32

1418

55 e

lder

ly a

ged

≥7

5 ye

ars

from

prim

ary

care

.D

ata

from

the

pro

spec

tive,

m

ultic

entr

e, o

bse

rvat

iona

l st

udy

‘Ger

man

Stu

dy

on

Age

ing,

Cog

nitio

n an

d

Dem

entia

in P

rimar

y C

are

Pat

ient

s (A

geC

oDe)

’.

Pre

scrib

ed d

rug

PIM

usi

ng B

eers

, PR

ISC

US

list

.B

y ch

ecki

ng m

edic

atio

ns d

urin

g vi

sits

to

pat

ient

s’ h

omes

.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: A

t b

asel

ine,

PIM

pre

vale

nce

is 2

9% (8

48) (

acco

rdin

g to

th

e P

RIS

CU

S li

st, w

hich

dec

reas

ed t

o 25

.0%

(464

) 4.5

yea

rs la

ter

(χ2 : 7

.87,

p

=0.

004)

.Th

e B

eers

list

yie

lded

a p

reva

lenc

e of

21%

(620

) at

bas

elin

e, d

ecre

asin

g af

ter

4.5

year

s to

17.

1% (3

17) (

χ2 : 10.

77, p

=0.

000)

.R

isk

fact

ors:

By

PR

ISC

US

list

:Th

e ris

k fo

r P

IM in

crea

se w

ith:

1. In

crea

sing

age

of t

he p

atie

nts

(OR

: 1.0

6, C

I 1.0

0 to

1.1

3, p

=0.

037)

.2.

The

pre

senc

e of

dep

ress

ion

(OR

: 2.4

2, C

I 1.6

5 to

3.5

7, p

=0.

000)

.3.

Incr

easi

ng n

umb

er o

f pre

scrip

tion

dru

gs (O

R: 1

.99,

CI 1

.80

to 2

.18,

p=

0.00

0).

By

cont

rast

, the

ris

k of

tak

ing

PIM

dec

reas

e b

y us

ing

PR

ISC

US

list

with

the

nu

mb

er o

f pre

sent

illn

ess

(OR

: 0.8

8, C

I 0.8

0 to

0.9

7, p

=0.

012)

.A

s th

e gr

owin

g nu

mb

er o

f ing

este

d p

resc

riptio

n d

rugs

incr

ease

d t

he r

isk

for

the

inge

stio

n of

PIM

from

the

Bee

rs li

st (O

R: 1

.66,

CI 1

.50

to 1

.84;

p=

0.00

0).

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e:

17%

–29%

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

13Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

49.

Bal

don

i et

al, 2

01429

Rib

eira

o P

reto

, Bra

zil

Cro

ss-s

ectio

nal

1000

eld

erly

age

d

≥60

year

s fr

om o

utp

atie

nt

pha

rmac

y

Pre

scrib

ed d

rug,

se

lf-m

edic

atio

n (3

09

user

s) a

nd O

TC (8

02

user

s)

Pre

vale

nce

and

fact

ors

asso

ciat

ed

with

PIM

usi

ng 2

003

and

201

2 B

eers

crit

eria

.U

sing

str

uctu

red

inte

rvie

w

que

stio

nnai

re.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: A

ccor

din

g to

Bee

rs c

riter

ia 2

003,

480

(48.

0%) p

artic

ipan

ts

used

at

leas

t on

e P

IM, t

he m

ean

bei

ng 1

.38

(SD

=0.

65) P

IMs/

per

son,

ran

ging

fr

om 1

to

5.A

ccor

din

g to

Bee

rs c

riter

ia 2

012,

592

(59.

2%) p

artic

ipan

ts u

sed

at

leas

t on

e P

IM, t

he m

ean

bei

ng 1

.56

(SD

=0.

81) P

IMs/

per

son,

ran

ging

from

1 t

o 6.

AD

E: D

urin

g th

e in

terv

iew

45.

5% o

f par

ticip

ants

rep

orte

d c

omp

lain

ts r

elat

ed t

o A

DE

s; 9

4.5%

of t

hese

wer

e ca

used

by

pre

scrib

ed m

edic

atio

n.R

isk

fact

ors:

Fact

ors

that

are

ass

ocia

ted

with

PIM

s us

e w

ere

fem

ale

gend

er, s

elf-

med

icat

ion,

us

e of

OTC

med

icat

ions

, com

pla

ints

rel

ated

to

AD

Es,

psy

chot

rop

ic m

edic

atio

n an

d m

ore

than

five

med

icat

ions

.*T

en m

edic

atio

ns w

ith t

he h

ighe

st p

reva

lenc

e of

sel

f-re

por

ted

AD

Es

com

pla

ints

ar

e cl

onid

ine,

am

itrip

tylin

e, m

etfo

rmin

, fluo

xetin

e, d

exch

lorp

heni

ram

ine,

d

iclo

fena

c, c

apto

pril

, ace

tyls

alic

ylic

aci

d, s

imva

stat

in a

nd h

ydro

chlo

roth

iazi

de.

A

mon

g th

em, fi

ve w

ere

cons

ider

ed P

IMs

acco

rdin

g to

Bee

rs c

riter

ia, o

f whi

ch

clon

idin

e, a

mitr

ipty

line

and

dex

chlo

rphe

nira

min

e ar

e lis

ted

in b

oth

crite

ria, w

hile

flu

oxet

ine

is li

sted

onl

y in

Bee

rs c

riter

ia 2

003

and

dic

lofe

nac

is li

sted

onl

y in

B

eers

crit

eria

201

2.

PIM

pre

vale

nce

by

Bee

rs

crite

ria 2

003:

480

/100

0=

48.0

%P

IM p

reva

lenc

e b

y B

eers

cr

iteria

201

2: 5

92/1

000=

59

.2%

*Err

or-r

elat

ed a

dve

rse

even

t

50.

Cas

tillo

-Pár

amo

et a

l, 20

1465

Sp

ain

Cro

ss-s

ectio

nal

272

elec

tron

ic r

ecor

ds

of e

lder

ly a

ged

≥65

yea

rs

from

prim

ary

heal

thca

re

Pre

scrib

ed d

rugs

PIM

usi

ng S

TOP

P/S

TAR

T cr

iteria

an

d v

ersi

on a

dap

ted

to

Sp

anis

h p

rimar

y he

alth

care

and

fact

ors

may

mod

ulat

e P

IM o

nset

.U

sing

ele

ctro

nic

heal

th r

ecor

d a

nd

pap

er c

linic

al r

ecor

d.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: T

he p

reva

lenc

e of

PIM

(mis

pre

scrib

ing

and

ove

rpre

scrib

ing)

us

ing

the

STO

PP

orig

inal

crit

eria

was

37.

5% (9

5% C

I 31.

7 to

43.

2), a

nd 5

0.7%

(9

5% C

I 44.

7 to

56.

6) u

sing

the

STO

PP

Sp

anis

h A

P20

12 v

ersi

on.

The

pre

vale

nce

of u

nder

pre

scrib

ing

was

45.

9% (9

5% C

I 40.

0 to

51.

8) w

ith

the

STA

RT

orig

inal

crit

eria

, and

43.

0% (9

5% C

I 37.

1 to

48.

9) w

ith t

he S

TAR

T A

P20

12 v

ersi

on.

Ris

k fa

ctor

s:A

sig

nific

ant

corr

elat

ion

was

foun

d b

etw

een

the

num

ber

of S

TOP

P P

IM a

nd a

ge

or n

umb

er o

f pre

scrip

tions

, and

bet

wee

n th

e nu

mb

er o

f STA

RT

PIM

with

age

, C

CI a

nd n

umb

er o

f pre

scrip

tions

.

PIM

pre

vale

nce:

102

/272

(S

TOP

P)=

37.5

%

(95%

CI 3

1.7

to 4

3.2)

, 13

8/27

2 (S

TOP

P

AP

2012

)=50

.7%

(95%

CI

44.7

to

56.6

), 12

5/27

2 (S

TAR

T)=

45.9

%,

117/

272

(STA

RT

AP

2012

)=43

%

51.

Vezm

ar K

ovač

ević

et

al, 2

01466

Ser

bia

Bel

grad

eC

ross

-sec

tiona

l, p

rosp

ectiv

e50

9 el

der

ly a

ged

≥65

yea

rs

from

five

com

mun

ity

pha

rmac

ies

Pre

scrib

ed d

rug

PIM

and

PP

O u

sing

STO

PP

/S

TAR

T cr

iteria

.U

sing

pat

ient

inte

rvie

w a

nd

med

ical

, bio

med

ical

rec

ord

.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: T

here

wer

e 16

4 P

IMs

iden

tified

in 1

39 p

atie

nts

(27.

3%) b

y S

TOP

P a

nd 4

39 P

PO

s id

entifi

ed in

257

pat

ient

s (5

0.5%

) by

STA

RT.

Ris

k fa

ctor

s:P

atie

nts

with

mor

e th

an fo

ur p

resc

riptio

ns h

ad a

hig

her

risk

for

PIM

(OR

2.8

5,

95%

CI 1

.97

to 4

.14,

p<

0.00

1) a

nd ≥

9 m

edic

atio

ns (O

R 7

.43,

95%

CI 3

.20

to

17.2

3, p

<0.

001)

.P

atie

nts

old

er t

han

74 y

ears

wer

e m

ore

likel

y to

hav

e a

PP

O (7

5–84

yea

rs: O

R

1.47

, 95%

CI 1

.01

to 2

.13,

p=

0.04

1; a

nd ≥

85 y

ears

: OR

1.7

9, 9

5% C

I 1.1

9 to

2.

83, p

=0.

009)

.

PIM

pre

vale

nce:

13

9/50

9=27

.3%

PP

O p

reva

lenc

e:

257/

509=

50.5

%

52.

Am

os e

t al

, 201

567E

mili

a-R

omag

na, I

taly

Coh

ort

retr

osp

ectiv

e86

5 35

4 el

der

ly a

ged

≥6

5 ye

ars

com

mun

ity-

dw

ellin

g fr

om

adm

inis

trat

ive

care

dat

a

Pre

scrib

ed d

rug

only

PIM

usi

ng u

pd

ated

Mai

o cr

iteria

an

d p

atie

nt c

hara

cter

istic

s re

late

d

to IP

.U

sing

reg

iona

l Em

ilia-

Rom

agna

ad

min

istr

ativ

e he

alth

care

d

atab

ase.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: A

tot

al o

f 240

310

(28%

) old

er a

dul

ts w

ere

exp

osed

to

at le

ast

one

PIM

.R

isk

fact

ors:

The

old

est

grou

p (≥

85) f

ollo

wed

by

pat

ient

s ag

ed 7

5–84

had

53%

and

25%

gr

eate

r od

ds

of r

ecei

ving

PIM

tha

n p

atie

nts

65–7

5 ye

ars

old

, res

pec

tivel

y (O

R=

1.53

, 95%

CI 1

.50

to 1

.55;

OR

=1.

25, 9

5% C

I 1.2

3 to

1.2

6, r

esp

ectiv

ely)

.Th

ese

odd

s of

exp

osur

e to

any

PIM

wer

e sl

ight

ly lo

wer

am

ong

men

tha

n w

omen

(OR

=0.

98, 9

5% C

I 0.9

7 to

1.0

0).

An

incr

ease

in t

he n

umb

er o

f med

icat

ions

pre

scrib

ed t

o th

e p

atie

nt

corr

esp

ond

ed w

ith h

ighe

r od

ds

of P

IM e

xpos

ure.

Old

er G

Ps

(≥56

), m

ale

GP

s an

d s

olo

pra

ctic

e G

Ps

wer

e m

ore

likel

y to

pre

scrib

e P

IMs

to t

heir

old

er p

atie

nts.

PIM

pre

vale

nce:

240

31

0/86

5 35

4=28

%

53.

Hed

na e

t al

, 201

568S

wed

enC

ohor

t re

tros

pec

tive

542

eld

erly

age

d ≥

65 y

ears

fr

om t

he S

wed

ish

Tota

l P

opul

atio

n R

egis

ter

(prim

ary

care

)

Pre

scrib

ed d

rug

Pre

vale

nce

of p

oten

tially

IPs

usin

g S

TOP

P c

riter

ia a

nd t

o in

vest

igat

e th

e as

soci

atio

n b

etw

een

pot

entia

lly IP

s an

d o

ccur

renc

e of

AD

Rs.

Usi

ng t

he S

wed

ish

Pre

scrib

edD

rug

Reg

iste

r, m

edic

al r

ecor

ds

and

hea

lth a

dm

inis

trat

ive

dat

a.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

e: 2

26 p

atie

nts

usin

g p

rimar

y he

alth

care

had

pot

entia

lly IP

.R

isk

fact

ors:

Per

sons

pre

scrib

ed p

oten

tially

IP h

ad m

ore

than

tw

ofol

d in

crea

sed

od

ds

to

exp

erie

nce

AD

Rs

(OR

2.4

7, 9

5 %

CI (

1.65

to

3.69

), p

<0.

001)

, com

par

ed w

ith

that

in p

erso

ns w

ithou

t p

oten

tially

IP.

Pot

entia

lly IP

pre

vale

nce:

22

6/54

2=42

%*E

rror

-rel

ated

ad

vers

eev

ent.

The

asso

ciat

ion

bet

wee

nP

IPs

and

occ

urre

nce

ofA

DR

s w

as fo

r p

rimar

yca

re, o

utp

atie

nt o

rin

pat

ient

and

ho

spita

lised

pat

ient

s.

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

14 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

54.

Mor

iart

y et

al,

2015

69Ire

land

Coh

ort

pro

spec

tive

2051

eld

erly

age

d ≥

65

year

s fr

om T

he Ir

ish

Long

itud

inal

Stu

dy

on

agei

ng.

Com

mun

ity-d

wel

ling

eld

erly

.

Pre

scrib

ed d

rug

only

PIM

and

PP

O u

sing

STO

PP,

Bee

rs

crite

ria, A

CO

VE

ind

icat

ors

and

S

TAR

T.U

sing

face

-to-

face

inte

rvie

w, t

hen

follo

w-u

p a

fter

1 a

nd 2

yea

rs.

Pre

scrib

ing

erro

r:P

IM p

reva

lenc

eP

IM: 3

6.7%

–64.

8%

Bas

elin

eN

(%) (

95%

C

I)

Follo

w-u

pN

(%) (

95%

CI)

Any

PIM

usi

ng S

TOP

P,

Bee

rs, A

CO

VE

1259

(61.

4%)

(CI 5

9.3

to 6

3.5)

1330

(64.

8%) (

CI 6

2.8

to 6

6.9)

Any

PP

O u

sing

STA

RT,

A

CO

VE

1094

(53.

3%)

(CI 5

1.2

to 5

5.5)

1161

(56.

6%) (

CI 5

4.5

to 5

8.8)

Bot

h P

IM a

nd P

PO

753

(36.

7%)

843

(41.

1%)

Ris

k fa

ctor

s:Fe

mal

e se

x, a

ge a

nd h

ighe

r nu

mb

er o

f med

icin

es w

ere

sign

ifica

ntly

ass

ocia

ted

w

ith c

hang

e in

PIM

pre

vale

nce.

Age

and

hig

her

num

ber

s of

med

icin

es a

nd c

hron

ic c

ond

ition

s w

ere

foun

d t

o b

e si

gnifi

cant

ly a

ssoc

iate

d w

ith c

hang

e in

PP

O p

reva

lenc

e.

55.

Ram

ia a

nd Z

eenn

y,

2014

71Le

ban

onC

ross

-sec

tiona

l28

4 ou

tpat

ient

s ag

ed

≥18

year

s vi

sitin

g co

mm

unity

pha

rmac

y

Pat

ient

s on

≥1

of t

he

chro

nic

med

icat

ions

m

entio

ned

in t

he

stud

y

The

com

ple

tion

of t

hera

peu

tic/

safe

ty m

onito

ring

test

s.P

atie

nts

wer

e su

bje

cted

to

a q

uest

ionn

aire

ass

essi

ng

the

app

rop

riate

ness

of t

heir

lab

orat

ory-

test

mon

itorin

g.

Mon

itorin

g er

ror:

1. 1

85 o

f the

pat

ient

s (6

5%) w

ere

foun

d t

o co

mp

lete

som

e, b

ut n

ot a

ll, o

f the

re

com

men

ded

the

rap

eutic

/saf

ety

mon

itorin

g te

sts.

2. 7

6 of

the

pat

ient

s (2

7%) c

omp

lete

d a

ll re

com

men

ded

the

rap

eutic

/saf

ety

mon

itorin

g.3.

23

of t

he p

atie

nts

(8%

) did

not

com

ple

te a

ny o

f the

rec

omm

end

ed m

onito

ring

test

s.

Inco

mp

lete

the

rap

eutic

/sa

fety

lab

orat

ory-

test

m

onito

ring

pre

vale

nce:

20

8/28

4=73

%

Oth

er: d

iscr

epan

cies

56.

Tuln

er e

t al

, 200

931A

mst

erd

am, T

he

Net

herla

nds

Des

crip

tive

pro

spec

tive

120

eld

erly

age

d >

65 y

ears

fr

om D

utch

ger

iatr

ic

outp

atie

nt

Usi

ng m

ore

than

on

e p

resc

ribed

or

OTC

med

icat

ions

1. F

req

uenc

y an

d r

elev

ancy

of

dis

crep

anci

es in

dru

g us

e.2.

Fre

que

ncy

of M

DA

PE

s.3.

Con

trib

utin

g fa

ctor

s su

ch a

s in

crea

sing

age

, cog

nitiv

e st

atus

an

d d

epre

ssiv

e sy

mp

tom

s, t

he

num

ber

of m

edic

atio

ns u

sed

, and

th

e nu

mb

er o

f phy

sici

ans

visi

ted

b

y th

e p

atie

nt.

By

com

par

ing

the

med

icat

ion

des

crib

ed b

y th

e p

atie

nt a

nd

care

give

rs w

ith t

he d

rugs

list

ed

by

thei

r G

P.

Oth

er: d

iscr

epan

cies

pre

vale

nce:

At

leas

t on

e d

iscr

epan

cy (d

elet

ion,

ad

diti

on o

r d

iffer

ence

in d

osag

e) b

etw

een

the

med

icat

ion

lists

from

the

pat

ient

, the

GP

or

the

pha

rmac

y w

as p

rese

nt in

10

4 p

atie

nts

(86.

7%) i

nvol

ving

386

dru

gs.

MD

AP

ES

:M

DA

PE

S w

ere

iden

tified

in 2

9 p

atie

nts

(24.

2%).

7 p

atie

nts

had

und

ertr

eatm

ent

due

to

del

etio

ns.

9 p

atie

nts

had

AD

R d

ue t

o ad

diti

ons.

13 p

atie

nt h

ad D

DI.

Ris

k fa

ctor

s:P

atie

nts

with

≥1

dis

crep

ancy

rep

orte

d u

sing

a h

ighe

r m

ean

num

ber

of d

rugs

(5.9

vs

4.0

; p<

0.05

) and

had

mor

e p

resc

ribin

g p

hysi

cian

s in

ad

diti

on t

o th

eir

GP

(1.1

vs

0.4

3; p

<0.

05).

Bot

h th

e p

rese

nce

of d

iscr

epan

cies

(Pea

rson

’s r

’, 0.

293;

p=

0.05

) and

MD

AP

Es

(Pea

rson

’s r

’, 0.

230;

p=

0.01

2) w

ere

sign

ifica

ntly

cor

rela

ted

with

the

num

ber

of

med

icat

ions

rep

orte

d b

y th

e p

atie

nt.

*The

hig

hest

rat

es o

f dis

crep

anci

es w

ere

seen

for

acet

amin

ophe

n (8

6.7%

), la

xativ

es (8

2.9%

) and

form

ulat

ions

for

der

mat

olog

ical

or

opht

halm

olog

ical

d

isea

ses

(81.

3%).

Dis

crep

anci

es

pre

vale

nce:

10

4/12

0=86

.7%

*Err

or-r

elat

ed a

dve

rse

even

t: M

DA

PE

s:

29/1

20=

24.2

%

*Err

or-r

elat

ed a

dve

rse

even

t

Tab

le 1

C

ontin

ued

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

15Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Key

cha

ract

eris

tics

of

incl

uded

stu

die

s

Aut

hor,

year

Co

untr

y/ci

tyS

tud

y d

esig

n/ty

pe

Po

pul

atio

n o

f in

tere

stE

xpo

sure

of

inte

rest

Out

com

e o

f in

tere

stM

ain

find

ing

Co

nclu

sio

n, n

/N (%

)A

dd

itio

nal n

ote

s

57.

Cor

nu e

t al

, 201

232B

russ

els,

Bel

gium

Coh

ort

retr

osp

ectiv

e18

9 el

der

ly a

ged

≥65

yea

rs

dis

char

ged

from

acu

te

geria

tric

dep

artm

ent

of a

B

elgi

an u

nive

rsity

hos

pita

l

Pre

scrib

ed d

rug

Inci

den

ce a

nd t

ype

of

dis

crep

anci

es b

etw

een

the

dis

char

ge le

tter

for

the

prim

ary

care

phy

sici

an a

nd t

he p

atie

nt

dis

char

ge m

edic

atio

n an

d

iden

tify

pos

sib

le p

atie

nt-r

elat

ed

det

erm

inan

ts fo

r ex

per

ienc

ing

dis

crep

anci

es.

Dis

crep

anci

es w

ere

cate

goris

ed

as o

mitt

ed d

rug,

uni

nten

ded

co

ntin

uatio

n (d

isco

ntin

ued

hom

e m

edic

atio

n d

ocum

ente

d a

s ho

me

med

icat

ion)

, dis

crep

ant

dos

e,

mis

sing

dos

e, a

nd d

iscr

epan

t b

rand

, om

issi

on o

f a b

rand

na

me,

dis

crep

ant

freq

uenc

y,

mis

sing

freq

uenc

y or

an

inco

rrec

t p

harm

aceu

tical

form

.B

y co

mp

arin

g th

e m

edic

atio

ns

liste

d in

the

dis

char

ge le

tter

for

the

prim

ary

care

phy

sici

an w

ith

thos

e in

the

pat

ient

dis

char

ge

med

icat

ion

list.

Oth

er: d

iscr

epan

cies

pre

vale

nce:

Alm

ost

half

of t

hese

pat

ient

s (n

=90

, 47.

6%) (

95%

CI 4

0.5

to 5

4.7)

had

one

or

mor

e d

iscr

epan

cies

in m

edic

atio

n in

form

atio

n at

dis

char

ge.

*Tw

o d

iscr

epan

cies

(1.2

%) w

ere

cate

goris

ed a

s ha

ving

the

pot

entia

l to

caus

e se

vere

pat

ient

har

m. T

hese

dis

crep

anci

es c

onsi

sted

of a

wro

ng d

ose

(dou

ble

d

the

pre

scrib

ed d

ose)

of d

igox

in in

the

pat

ient

dis

char

ge m

edic

atio

n lis

t an

d t

he

listin

g of

a lo

w-m

olec

ular

-wei

ght

hep

arin

in t

he p

atie

nt d

isch

arge

med

icat

ion

list

that

was

inte

ntio

nally

om

itted

in t

he d

isch

arge

lett

er b

ecau

se o

f the

d

evel

opm

ent

of h

epar

in-i

nduc

ed t

hrom

boc

ytop

aeni

a d

urin

g ho

spita

lisat

ion.

Ris

k fa

ctor

s:Th

e ex

plo

rativ

e m

ultiv

aria

te m

odel

ad

just

ed fo

r ag

e, s

ex, l

engt

h of

hos

pita

l sta

y an

d r

esid

entia

l situ

atio

n sh

owed

tha

t w

hen

the

dis

char

ge le

tter

con

tain

ed m

ore

than

five

dru

gs, t

he li

kelih

ood

of e

xper

ienc

ing

one

or m

ore

dru

g d

iscr

epan

cies

w

as 3

.22

(95%

CI 1

.40

to 7

.42,

p=

0.00

6) t

imes

hig

her

than

whe

n fiv

e or

few

er

dru

gs w

ere

men

tione

d.

Incr

easi

ng n

umb

ers

of d

rugs

in t

he d

isch

arge

med

icat

ion

list

(OR

1.1

9, 9

5% C

I 1.

07 t

o 1.

32, p

=0.

001)

and

dis

char

ge le

tter

(OR

1.1

8, 9

5% C

I 1.0

7 to

1.3

2,

p=

0.00

1) w

ere

asso

ciat

ed w

ith a

hig

her

risk

for

dis

crep

anci

es.

Dis

crep

anci

es

pre

vale

nce:

90

/189

=47

.6%

(95%

CI

40.5

to

54.7

)

*Err

or-r

elat

ed a

dve

rse

even

t

Pre

vent

able

AD

Es

58.

Fiel

d e

t al

, 200

777U

SA

Coh

ort

30 0

00 e

lder

ly ≥

65 y

ears

fr

om a

mb

ulat

ory

care

Pre

scrib

ed d

rug

AD

E r

esul

ting

from

pat

ient

s er

ror

and

ris

k fa

ctor

s.B

y el

ectr

onic

tra

ckin

g of

ad

min

istr

ativ

e d

ata,

rev

iew

of

med

ical

rec

ord

s, r

epor

ts fr

om

clin

icia

ns, h

osp

ital d

isch

arge

su

mm

arie

s an

d E

D v

isit.

Pre

vent

able

AD

E:

AD

E r

esul

ting

from

pat

ient

s’ e

rror

pre

vale

nce:

113

ind

ivid

uals

exp

erie

nced

AD

E

and

pot

entia

l AD

E.

Ris

k fa

ctor

:In

a m

ultiv

aria

te a

naly

sis,

the

re w

as a

dos

e–re

spon

se a

ssoc

iatio

n b

etw

een

pat

ient

err

ors

lead

ing

to A

DE

s an

d p

oten

tial A

DE

s an

d r

egul

arly

sch

edul

ed

med

icat

ions

; com

par

ed w

ith z

ero

to t

wo

med

icat

ions

, the

OR

for

thre

e to

four

m

edic

atio

ns w

as 2

.0 (9

5% C

I 0.9

to

4.2)

; for

five

to

six

med

icat

ions

was

3.1

(9

5% C

I 1.5

to

7.0)

; and

for

seve

n or

mor

e m

edic

atio

ns w

as 3

.3 (9

5% C

I 1.5

to

7.0

).Th

e st

rong

est

asso

ciat

ion

was

with

the

CC

I; co

mp

ared

with

a s

core

of 0

, the

O

R fo

r a

scor

e of

1–2

was

3.8

(95%

CI 2

.1 t

o 7.

0); f

or a

sco

re o

f 3–4

was

8.6

(9

5% C

I 4.3

to

17.0

); an

d fo

r a

scor

e of

5 o

r m

ore

was

15.

0 (9

5% C

I 6.5

to

34.5

).

AD

E r

esul

ting

from

p

atie

nts’

err

or

pre

vale

nce:

113

/30

000=

0.38

%

*AD

E r

esul

ting

from

p

atie

nts’

err

or

59.

Gan

dhi

et

al, 2

01022

Bos

ton

and

Ind

iana

pol

is,

US

AC

ross

-sec

tiona

l68

013

out

pat

ient

s, m

ean

age

48 a

nd 4

7 ye

ars

Pre

scrib

ed d

rug

AD

E.

Usi

ng e

lect

roni

c he

alth

rec

ord

sc

reen

ing,

cha

rt r

evie

w a

nd A

DE

m

onito

r.

Pre

vent

able

AD

E in

cid

ence

:Th

e ov

eral

l rat

e w

as 1

38 A

DE

s/10

00 p

erso

n-ye

ars

acro

ss t

he t

wo

site

s.

Pre

vent

able

AD

Es

rate

15/

1000

per

son-

year

s ac

ross

tw

o si

tes.

*Mos

t co

mm

on d

rugs

ass

ocia

ted

with

pre

vent

able

AD

E w

ere

AC

E in

hib

itors

an

d b

eta-

blo

cker

s.

Pre

vent

able

AD

Es

rate

15

/100

0 p

erso

n-ye

ars

acro

ss t

wo

site

s

*Pre

vent

able

AD

E

60.

Ob

reli-

Net

o et

al,

2012

28O

urin

hos

mic

rore

gion

,B

razi

lC

ohor

t p

rosp

ectiv

e43

3 el

der

ly a

ged

≥60

yea

rs

from

the

prim

ary

pub

lic

heal

th s

yste

m

Pre

scrib

ed d

rugs

b

oth

with

in a

nd

acro

ss p

resc

riptio

ns

DD

I-re

late

d A

DR

inci

den

ce a

nd

fact

ors.

Usi

ng p

hone

or

face

-to-

face

st

ruct

ured

inte

rvie

w.

DD

I scr

eeni

ng t

ool:

DD

I che

cker

p

rogr

amm

es (D

rugD

iges

t, D

rugs

, M

icro

med

ex a

nd M

edsc

ape)

.

Pre

vent

able

AD

E:

DD

I-re

late

d A

DR

inci

den

ce: o

ccur

red

in 3

0 p

atie

nts

(6.9

%).

Gas

troi

ntes

tinal

ble

edin

g oc

curr

ed in

37%

of t

he D

DI-

rela

ted

AD

R c

ases

, fo

llow

ed b

y hy

per

kala

emia

(17%

) and

myo

pat

hy (1

3%).

Sev

ente

en D

DI-

rela

ted

A

DR

s w

ere

clas

sifie

d a

s se

verit

y le

vel 2

, and

hos

pita

l ad

mis

sion

was

nec

essa

ry

in 1

1 ca

ses.

*War

farin

was

the

mos

t co

mm

only

invo

lved

dru

g (3

7% o

f cas

es),

follo

wed

by

acet

ylsa

licyl

ic a

cid

(17%

), d

igox

in (1

7%) a

nd s

piro

nola

cton

e (1

7%).

Ris

k fa

ctor

s:Th

e m

ultip

le lo

gist

ic r

egre

ssio

n sh

owed

tha

t th

e fo

llow

ing

wer

e as

soci

ated

with

th

e oc

curr

ence

of D

DI-

rela

ted

AD

Rs:

1. A

ge ≥

80 y

ears

(OR

4.4

, 95%

CI 3

.0 t

o 6.

1, p

<0.

01).

2. C

CI ≥

4 (O

R 1

.3, 9

5% C

I 1.1

to

1.8,

p<

0.01

).3.

Con

sum

ptio

n of

five

or

mor

e d

rugs

(OR

2.7

, 95%

CI 1

.9 t

o 3.

1, p

<0.

01).

4. U

se o

f war

farin

(OR

1.7

, 95%

CI 1

.1 t

o 1.

9, p

<0.

01).

Inci

den

ce o

f DD

I-re

late

d

AD

R: 3

0/43

3=6.

9%*E

rror

-rel

ated

ad

vers

e ev

ent

AC

OV

E, A

sses

sing

Car

e of

Vul

nera

ble

Eld

ers;

AD

E, a

dve

rse

dru

g ev

ent;

AD

I, ad

vers

e d

rug

inte

ract

ion;

AD

R, a

dve

rse

dru

g re

actio

n; C

CI,

Cha

rlson

Com

orb

idity

Ind

ex; D

DI,

dru

g–d

rug

inte

ract

ion;

ED

, em

erge

ncy

dep

artm

ent;

GP,

gen

eral

pra

ctiti

oner

s; H

ED

IS, H

ealth

Pla

n E

mp

loye

r D

ata

and

Info

rmat

ion

Set

; HR

QO

L, h

ealth

-rel

ated

qua

lity

of li

fe; I

DU

, ina

pp

rop

riate

dru

g us

e; IP

, ina

pp

rop

riate

pre

scrib

ing;

 IPE

T, im

pro

ved

pre

scrib

ing

in t

he e

lder

ly t

ool;

MA

I, M

edic

atio

n A

pp

rop

riate

Ind

ex; M

DA

PE

, med

icat

ion

dis

crep

ancy

ad

vers

e p

atie

nt e

vent

; ME

PS

, Med

ical

Exp

end

iture

Pan

el S

urve

y; N

OR

GE

P, N

orw

egia

n G

ener

al P

ract

ice;

OTC

, ove

r-th

e-co

unte

r; P

DD

I, p

oten

tial d

rug–

dis

ease

inte

ract

ion;

PIM

, pot

entia

lly in

app

rop

riate

med

icin

e; P

PO

, pot

entia

l pre

scrib

ing

omis

sion

s; S

TAR

T, S

cree

ning

Too

l to

Ale

rt d

octo

rs t

o R

ight

Tre

atm

ent;

STO

PP,

Scr

eeni

ng T

ool o

f Old

er P

erso

n’s

Pre

scrip

tions

.

Tab

le 1

C

ontin

ued

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

16 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Tab

le 2

A

Sys

tem

atic

rev

iew

qua

lity

asse

ssm

ent:

Joa

nna

Brig

gs In

stitu

te C

ritic

al A

pp

rais

al C

heck

list

for

des

crip

tive/

case

ser

ies

and

cro

ss-s

ectio

nal

12

34

56

78

9O

vera

ll ap

pra

ised

1R

amia

and

Zee

nny,

20

1471

Ad

ult

YY

NN

NA

NA

YY

YH

igh

Pat

ient

s w

ere

sub

ject

ed t

o a

que

stio

nnai

re

asse

ssin

g th

e ap

pro

pria

tene

ss o

f the

ir la

bor

ator

y-te

st m

onito

ring,

may

cau

se

reca

ll b

ias.

2S

oren

sen

et a

l, 20

0676

Ad

ult

YY

N, r

isk

fact

ors

rela

ted

to

pat

ient

no

t st

udie

d

YN

AN

AY

YY

Hig

h

3Vu

ong

and

 Mar

riott

, 20

0625

Ad

ult

UY

NY

NA

NA

NY

Y, p

erce

ntag

e w

as u

sed

but

st

atis

tics

was

no

t d

escr

ibed

in

the

full

text

.

Hig

hU

ncle

ar s

amp

ling

stra

tegy

.

4A

dam

s et

 al,

2009

72

Ad

ult

YY

Y (b

ut fo

r al

l ty

pes

of a

dve

rse

even

t)

N (s

elf-

rep

orte

d

adve

rse

even

ts)

NA

NA

NY

YH

igh

Ris

k of

rec

all b

ias

and

att

ribut

ion

with

sel

f-re

por

ted

ad

vers

e ev

ents

.

5G

and

hi e

t al

, 201

022

Ad

ult

UY

NY

YN

AN

AY

YH

igh

6Lu

and

 Rou

ghea

d,

2011

20

Ad

ult

YY

YN

(sub

ject

ive

pat

ient

-rep

orte

d

med

icat

ion

erro

r)

YN

AN

A (s

econ

dar

y an

alys

is)

N (t

elep

hone

su

rvey

, sel

f-re

por

ted

)

YH

igh

Ris

k of

rec

all b

ias

with

pat

ient

-rep

orte

d

med

icat

ion

erro

r.

7S

ears

et 

al, 2

01221

Ad

ult

YY

YN

(sub

ject

ive

self-

rep

orte

d

med

icat

ion

erro

r)

YN

AN

A (s

econ

dar

y an

alys

is)

N (t

elep

hone

su

rvey

, sel

f-re

por

ted

)

YH

igh

Ris

k of

rec

all b

ias

with

pat

ient

sel

f-re

por

ted

 med

icat

ion

erro

r.

8K

oper

et 

al, 2

01323

Ad

ult

N (c

onve

nien

ce)

YN

YN

AN

AN

A (1

00%

p

artic

ipan

ts)

YY

Hig

hS

elec

tion

bia

s.

9D

alle

nbac

h et

 al,

2007

24

Ad

ult-

DD

I

N (c

onse

cutiv

e)N

NY

NA

NA

NA

(re

tros

pec

tive)

YY

Mod

erat

e

10In

der

mitt

e et

 al,

2007

34

Ad

ult-

DD

I

Y (p

harm

acy

choo

se);

N (fi

rst

12

cust

omer

s)

YN

YN

AN

AY

YY

Hig

h

11M

ahm

ood

et

 al, 

2007

35

Ad

ult-

DD

I

YY

NY

NA

NA

NA

(re

tros

pec

tive)

YY

Hig

hP

atie

nts

may

act

ually

be

on o

ther

dru

gs s

o m

ay n

ot c

atch

all

the

DD

I.

12G

uthr

ie e

t al

, 201

539

Ad

ult-

DD

IY

YY

(but

for

bot

h ow

n ho

me

and

ca

re h

ome)

YY

NA

NA

(sec

ond

ary

anal

ysis

)Y

YH

igh

Ris

k fa

ctor

s fo

r b

oth

own

hom

e an

d c

are

hom

e.

13d

e O

livei

ra M

artin

s et

 al,

2006

41

Eld

erly

-PIM

N (fi

rst

cam

e to

p

harm

acy

carr

ying

p

resc

riptio

n fo

r tw

o or

mor

e d

rugs

)

YY,

but

not

all

YY

NA

NY

YH

igh

Sel

f-re

por

ted

dat

a fr

om e

lder

ly c

once

rnin

g d

rug

use

may

lead

to

info

rmat

ion

bia

s.

14P

ugh

et a

l, 20

0642

Eld

erly

-PIM

YY

YY

YN

AN

A (s

econ

dar

y d

ata

anal

ysis

)Y

YH

igh

May

und

eres

timat

e th

e ex

pos

ure

bec

ause

th

ey d

o no

t ac

coun

t fo

r O

TC.

15S

aab

et 

al, 2

00643

Eld

erly

-PIM

YY

YY

NA

NA

YY

YH

igh

Sel

f-re

por

ted

dat

a fr

om e

lder

ly c

once

rnin

g d

rug

use

may

dec

reas

e ac

cura

cy.

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

17Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

12

34

56

78

9O

vera

ll ap

pra

ised

16B

regn

høj e

t al

, 200

745

Eld

erly

-PIM

N (e

ach

GP

was

as

ked

to

recr

uit

six

pat

ient

s w

ho w

ere

rand

omly

sel

ecte

d)

YN

YN

AN

AY

YY

Hig

hS

elec

tion

bia

s.

17Jo

hnel

l and

 Fas

tbom

, 20

0846

Eld

erly

-PIM

YY

YY

YN

AY

YY

Hig

hD

id n

ot lo

ok fo

r co

mor

bid

ity a

s a

risk

fact

or b

ecau

se d

ata

wer

e fr

om S

wed

ish

Pre

scrib

ing

Dru

g R

egis

ter.

18H

aid

er e

t al

, 200

948

Eld

erly

-PIM

YY

YY

NA

NA

NA

YY

Hig

h

19La

i et 

al, 2

00949

Eld

erly

-PIM

YY

YY

NA

NA

NA

(sec

ond

ary

anal

ysis

)Y

YH

igh

Did

not

ad

dre

ss c

omor

bid

ity a

s a

risk

fact

or.

20R

yan 

et a

l, 20

0951

Eld

erly

-PIM

YY

YY

NA

NA

NY

YH

igh

May

und

eres

timat

e th

e ou

tcom

e b

ecau

se

they

do

not

acco

unt

for

OTC

.

21Z

aver

i et 

al, 2

01053

Eld

erly

-PIM

UY

YY

NA

NA

NY

YH

igh

Not

eno

ugh

info

rmat

ion

in t

he a

rtic

le.

22Le

ikol

a et

 al,

2011

56

Eld

erly

-PIM

YY

NY

NA

NA

NA

YY

Hig

hM

ay u

nder

estim

ate

the

outc

ome

bec

ause

d

atab

ase

lack

s d

iagn

ostic

pat

ient

dat

a,

ther

efor

e us

ed t

he B

eers

200

3 cr

iteria

in

dep

end

ent

of d

iagn

oses

and

the

dat

a p

rovi

de

no in

form

atio

n on

the

use

of P

IMs

that

are

not

rei

mb

ursa

ble

. Nin

e P

IMs

that

wer

e no

t re

imb

ursa

ble

in F

inla

nd in

20

07: t

riazo

lam

, bel

lad

onna

alk

aloi

ds,

d

iphe

nhyd

ram

ine,

hyd

roxy

zine

, fer

rous

su

lfate

, bis

acod

yl, n

itrof

uran

toin

and

cl

onid

ine.

23Li

n et

 al,

2011

57

Eld

erly

-PIM

UY

YY

NA

NA

NA

YY

Hig

h

24W

oelfe

l et 

al, 2

01170

Eld

erly

-PIM

YY

YY

NA

NA

NA

YY

Hig

h

25H

aasu

m e

t al

, 201

259

Eld

erly

-PIM

YY

NY

YN

AN

A (s

econ

dar

y d

ata

anal

ysis

)Y

YH

igh

26N

ybor

g et

 al,

2012

60

Eld

erly

-PIM

YY

YY

YN

AN

A (s

econ

dar

y d

ata

anal

ysis

)Y

YH

igh

27Ya

sein

et 

al, 2

01261

Eld

erly

-PIM

NY

NY

YN

AN

YY

Mod

erat

e

28C

and

ela 

Mar

roq

uín

et

al, 2

01219

Eld

erly

-PIM

N (c

onve

nien

ce

sam

ple

)Y

NY

NA

NA

NY

YM

oder

ate

Sam

plin

g st

rate

gy.

Sub

ject

ive

info

rmat

ion

on s

ocio

econ

omic

an

d c

linic

al v

aria

ble

s m

ay d

ecre

ase

accu

racy

.

29W

eng

et a

l, 20

1364

Eld

erly

-PIM

YY

YY

YN

AN

YY

Hig

hS

amp

ling

stra

tegy

.

30B

ald

oni e

t al

, 201

429

Eld

erly

-PIM

UY

YY

YN

AY

YY

Hig

h

Tab

le 2

A

Con

tinue

d

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

18 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

12

34

56

78

9O

vera

ll ap

pra

ised

31C

astil

lo-P

áram

o et

 al,

2014

65

Eld

erly

-PIM

YY

YY

YN

AY

YY

Hig

hE

lect

roni

c he

alth

rec

ord

use

lim

itatio

ns

(inco

mp

lete

rec

ord

and

qua

lity

of d

ata)

.

32Ve

zmar

Kov

ačev

ić

et a

l, 20

1466

Eld

erly

-PIM

YY

YY

NA

NA

NY

YH

igh

33N

obili

et 

al, 2

00938

Eld

erly

-DD

IY

YY

YN

AN

AN

A

(ad

min

istr

ativ

e d

atab

ase)

YY

Hig

hTh

e us

e of

ad

min

istr

ativ

e d

atab

ase

limits

lo

okin

g fo

r co

mor

bid

ity a

s a

conf

ound

er.

34S

ecol

i et 

al, 2

01030

Eld

erly

-DD

IU

YY

YN

AN

AN

AY

YH

igh

May

und

eres

timat

e th

e tr

ue D

DI

pre

vale

nce

bec

ause

the

y d

o no

t ac

coun

t fo

r O

TC.

35O

bre

li N

eto

et a

l, 20

1227

Eld

erly

-DD

I

YY

YY

NA

NA

NA

(dat

a fr

om p

rimar

y he

alth

care

sy

stem

)

YY

Hig

hM

ay u

nder

estim

ate

the

DD

I pre

vale

nce

bec

ause

(1) m

ost

inst

rum

ents

ava

ilab

le

for

asse

ssin

g D

DIs

con

sid

er o

nly

pai

rs o

f dru

gs a

nd d

o no

t ac

coun

t fo

r in

tera

ctio

ns in

volv

ing

com

bin

atio

ns o

f th

ree

or m

ore

dru

gs s

o (2

) did

not

acc

ount

fo

r O

TC.

36P

it et

 al,

2008

74

Eld

erly

YY

YY

NA

NA

YY

YH

igh

37Tu

lner

et 

al, 2

00931

Eld

erly

N (c

onse

cutiv

e)Y

YY

NA

NA

YY

YH

igh

Info

rmat

ion

on m

edic

atio

n d

escr

ibed

by

the

pat

ient

and

car

egiv

ers

may

not

alw

ays

be

accu

rate

.

38O

bre

li N

eto

et a

l, 20

1126

Eld

erly

-DD

I

YY

NY

NA

NA

NA

YY

Hig

h

39M

ira e

t al

, 201

373

Eld

erly

YY

YY

NA

NA

YY

YH

igh

Sel

f-re

por

ted

 med

icat

ion

erro

r fr

om e

lder

ly

conc

erni

ng d

rug

use

may

hav

e re

call

bia

s.

40M

and

et 

al, 2

01433

Eld

erly

YY

YY

NA

NA

NA

YY

Hig

h

1 W

as s

tud

y b

ased

on

a ra

ndom

or

pse

udo-

rand

om s

amp

le?

2 W

ere

the

crite

ria fo

r in

clus

ion

in t

he s

amp

le c

lear

ly d

efine

d?

3 W

ere

conf

ound

ing

fact

ors

iden

tified

and

str

ateg

ies

to d

eal w

ith t

hem

sta

ted

?4 

Wer

e ou

tcom

es a

sses

sed

usi

ng o

bje

ctiv

e cr

iteria

?5 

If co

mp

aris

ons

are

bei

ng m

ade,

was

the

re s

uffic

ient

des

crip

tions

of t

he g

roup

s?6 

Was

follo

w-u

p c

arrie

d o

ut o

ver

a su

ffici

ent

time

per

iod

?7 

Wer

e th

e ou

tcom

es o

f peo

ple

who

with

dre

w d

escr

ibed

and

incl

uded

in t

he a

naly

sis?

8 W

ere

outc

omes

mea

sure

d in

a r

elia

ble

way

?9 

Was

ap

pro

pria

te s

tatis

tical

ana

lysi

s us

ed?

DD

I, d

rug-

dru

g in

tera

ctio

n; G

P, g

ener

al p

ract

ition

er; N

, no;

NA

, not

ap

plic

able

; OTC

, ove

r-th

e-co

unte

r; P

IM, p

oten

tially

inap

pro

pria

te m

edic

atio

n; U

, unc

lear

; Y, y

es.

Tab

le 2

A

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

19Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Tab

le 2

B

Sys

tem

atic

rev

iew

qua

lity

asse

ssm

ent:

 Crit

ical

Ap

pra

isal

Ski

lls P

rogr

amm

e fo

r co

hort

stu

dy

Stu

dy

des

ign:

co

hort

Ref

eren

ceQ

ualit

y d

om

ains

12

34

5(a)

5(b

)6(

a)6(

b)

78

910

1112

Ove

rall

qua

lity

Are

the

res

ults

of t

he s

tud

y va

lid?

Wha

t ar

e th

e re

sults

?W

ill t

he r

esul

ts h

elp

loca

lly?

1M

aio

et a

l, 20

0640

PIM

YY

YY

Y, a

ge, g

end

er,

geog

rap

hica

l lo

catio

n, n

umb

er o

f m

edic

atio

n, n

umb

er

of c

hron

ic c

ond

ition

an

d in

com

e

NY

Y (1

yea

r)

retr

osp

ectiv

eP

IM p

reva

lenc

e: 1

8%.

Old

er a

ge, p

olyp

harm

acy

and

gre

ater

num

ber

of

chro

nic

cond

ition

s w

ere

sign

ifica

nt p

red

icto

rs o

f P

IM u

se.

P<

0.05

, 95

% C

IY

YY

–M

oder

ate

Non

e

2Z

ucke

rman

et

 al,

2006

44

PIM

YY

YY

Y, b

ut u

sed

for

irrel

evan

t ou

tcom

eY

YY

(2 y

ears

)In

app

rop

riate

med

icat

ion

use

pre

vale

nce:

41.

9%P

=0.

01,

99%

CI

YC

anno

t te

ll (g

ener

alis

abili

ty)

YLi

mite

d in

form

atio

n fr

om t

he

dat

abas

e.C

onfo

und

ing

fact

ors

wer

e fo

r th

e nu

rsin

g ho

me

adm

issi

on r

athe

r th

an fo

r P

IM.

Mod

erat

e

-

3Fi

eld

et 

al,

2007

77

Eld

erly

YY

YY

Y, a

ge, g

end

er,

com

orb

idity

, num

ber

of

med

icat

ions

YY

Y (1

yea

r)A

DE

res

ultin

g fr

om

pat

ient

s’ e

rror

pre

vale

nce:

0.

38%

P<

0.05

YY

YP

ossi

ble

dru

g-re

late

d in

cid

ence

fo

r w

hich

nec

essa

ry in

form

atio

n w

as n

ot d

ocum

ente

d in

th

e m

edic

al r

ecor

d w

as n

ot

cons

ider

ed.

Hig

h

Non

e

4G

agne

et 

al,

2008

36

DD

I

YY

YY

Y, a

ge, g

end

er,

geog

rap

hica

l lo

catio

n, c

omor

bid

ity,

num

ber

of m

edic

atio

n p

resc

ribed

YY

Y (1

yea

r)D

DI:

pre

vale

nce:

53%

95%

CI

YY

YA

pp

lyin

g th

e U

S li

st o

f clin

ical

ly

imp

orta

nt D

DI t

o Ita

ly m

ay

und

eres

timat

e th

e p

reva

lenc

e as

it

cap

ture

d o

nly

12 o

ut o

f the

25

DD

I orig

inal

list

. Una

ble

to

extr

act

risk

fact

ors

dat

a as

it is

 for

all a

ge

grou

ps.

Hig

h

Non

e

5B

erd

ot e

t al

, 20

0947

Eld

erly

-PIM

YY

YY

Y, b

ut fo

r irr

elev

ant

outc

ome

YY

Y (4

yea

rs)

PM

I pre

vale

nce:

 31.

6%95

% C

I, p

<0.

05Y

YY

Sel

f-re

por

t an

d d

ata

from

he

alth

care

insu

ranc

e p

lan

are

not

per

fect

for

actu

al d

rug

cons

ump

tion.

Rec

all b

ias.

Con

foun

din

g fa

ctor

s w

ere

for

the

risk

of fa

lls r

athe

r th

an fo

r P

IM.

Hig

h

–

6La

pi e

t al

, 20

0937

Eld

erly

-PIM

YY

YY

Y, c

omor

bid

ity,

pol

ypha

rmac

y, s

trok

e,

hear

t fa

ilure

YY

Y (1

yea

r)19

99:

IP p

reva

lenc

e: 5

.1%

Pot

entia

l DD

I pre

vale

nce:

30

.5%

Pot

entia

l maj

or D

DI:

5.6%

Pol

ypha

rmac

y w

as a

p

red

icto

r of

PIM

use

.

P<

0.05

, 95

% C

IY

NY

Sel

f-re

por

ted

dia

gnos

is a

nd

med

icat

ion

use

may

cau

se r

ecal

l b

ias.

Bee

rs li

st c

anno

t b

e fu

lly a

pp

lied

to

Ital

y; it

mos

t re

flect

s U

S d

rug

mar

ket.

Mod

erat

e

Age

, gen

der

7R

yan

et a

l, 20

0950

Eld

erly

-PIM

YY

YY

NC

anno

t te

llY

Y (6

mon

ths)

Med

icin

e p

resc

ribed

in

app

rop

riate

ly.

Bee

rs 2

003:

13%

IPE

T: 1

0.4%

Can

not

tell

YY

Y–

Low

–

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

20 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Stu

dy

des

ign:

co

hort

Ref

eren

ceQ

ualit

y d

om

ains

12

34

5(a)

5(b

)6(

a)6(

b)

78

910

1112

Ove

rall

qua

lity

8A

kaza

wa

et a

l, 20

1052

Eld

erly

-PIM

YY

YY

Y, a

ge, g

end

er,

pol

ypha

rmac

y (>

5 d

rugs

), ho

spita

lisat

ion,

co

mor

bid

ities

YY

Y (1

yea

r)P

reva

lenc

e of

PIM

: 43

.6%

.In

pat

ient

ser

vice

use

, p

olyp

harm

acy

and

co

mor

bid

ities

wer

e si

gnifi

cant

pre

dic

tors

of

PIM

use

.

95%

CI,

p<

0.05

YY

YM

edic

al in

form

atio

n ca

nnot

b

e ta

ken

from

cla

im d

ata,

un

obse

rved

con

foun

der

.P

IM n

ot a

ssoc

iate

d w

ith a

ge a

s se

vera

l oth

er s

tud

ies.

Hig

h

Non

e

9B

arne

tt e

t al

, 20

1154

Eld

erly

-PIM

YY

YY

Y, a

ge, s

ex,

pol

ypha

rmac

y an

d

pla

ce o

f res

iden

ce

YY

Y (2

 yea

rs)

PIM

pre

vale

nce:

30.

9%.

Pat

ient

s at

incr

ease

d r

isk

of r

ecei

ving

at

leas

t on

e P

IM if

the

y w

ere

youn

ger,

fem

ale

and

had

hig

her

pol

ypha

rmac

y.

95%

CI

YY

YC

omor

bid

ity n

ot a

ccou

nted

for.

Ris

k fa

ctor

s fo

r b

oth

care

hom

e an

d h

ome.

Hig

h

Com

orb

idity

10C

hang

et 

al,

2011

55

Eld

erly

-PIM

YY

YY

Y, a

ge, s

ex,

educ

atio

n, n

umb

er o

f ch

roni

c m

edic

atio

n,

num

ber

of c

hron

ic

cond

ition

s an

d

num

ber

of E

D v

isits

YY

Y (1

2,

24 w

eeks

)P

IM: 2

4%–7

3%N

umb

er o

f chr

onic

d

rugs

and

num

ber

of

chro

nic

cond

ition

s w

ere

a co

mm

on r

isk

fact

or in

all

crite

ria.

P<

0.05

YY

YM

ay u

nder

estim

ate

the

pre

vale

nce

bec

ause

sev

eral

dru

gs

in T

aiw

an w

ere

not

avai

lab

le in

the

se

x cr

iteria

.

Hig

h

Non

e

11Z

hang

et 

al,

2011

58

Eld

erly

-PIM

YY

YY

Y, r

ace,

gen

der

, fam

ily

inco

me,

ed

ucat

iona

l le

vel,

cens

us

regi

on, n

umb

er o

f p

resc

riptio

n, s

elf-

rate

d h

ealth

sta

tus

YY

Can

not

tell

Pre

vale

nce

of P

IM w

as

from

13.

84%

(95%

CI

12.5

2 to

15.

17) t

o 21

.3%

(9

5% C

I 19.

5 to

23.

1).

95%

CI,

p<

0.05

YY

YR

ecal

l bia

s d

ue t

o se

lf-re

por

ted

su

rvey

. Did

not

ass

ess

DD

I and

un

der

use

so m

ay u

nder

estim

ate

the

pre

vale

nce.

Mod

erat

e

Non

e

12C

ornu

et 

al,

2012

32

Eld

erly

YY

YY

Y, a

ge, g

end

er,

resi

den

tial s

ituat

ion

bef

ore

adm

issi

on,

resi

den

tial s

ituat

ion

afte

r d

isch

arge

, nu

mb

er o

f dru

gs in

th

e d

isch

arge

lett

er

or li

st

YY

Y (f

rom

ad

mis

sion

to

dis

char

ge)

Alm

ost

half

of t

hese

p

atie

nts

(47.

6% (9

5% C

I 40

.5 t

o 54

.7))

had

one

or

mor

e d

iscr

epan

cies

in

med

icat

ion

info

rmat

ion

at

dis

char

ge.

95%

CI,

p<

0.05

YC

anno

t te

llY

Was

don

e in

one

cen

tre

that

m

ay h

ave

diff

eren

t p

roce

dur

e of

d

isch

arge

.

Mod

erat

e

Com

orb

idity

Tab

le 2

B

Con

tinue

d

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

21Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Stu

dy

des

ign:

co

hort

Ref

eren

ceQ

ualit

y d

om

ains

12

34

5(a)

5(b

)6(

a)6(

b)

78

910

1112

Ove

rall

qua

lity

13M

oshe

r et

 al,

2012

75

Eld

erly

YY

YY

Y, h

ealth

lite

racy

YY

Y (3

and

12

mon

ths)

AD

Es

occu

rred

in 5

1p

atie

nts

(16.

5%) o

f th

e p

atie

nts

with

in

the 

first

3 m

onth

s of

the

st

udy,

 whi

ch in

crea

sed

to

 119

pat

ient

s (3

8.4%

) ov

er t

he fu

ll 12

-mon

th

follo

w-u

p p

erio

d.

P<

0.05

YC

anno

t te

llY

Res

ults

may

be

bia

sed

due

to

sam

plin

g st

rate

gy.

Mod

erat

e

Age

, num

ber

of

med

icat

ions

, co

mor

bid

ity

14O

bre

li-N

eto

et

al, 2

01228

DD

I

YY

YY

Y

YY

Y (4

mon

ths)

Inci

den

ce o

f DD

I-re

late

d

AD

R (6

.9%

)95

% C

I, p

<0.

05Y

YN

Rec

all b

ias

from

wee

kly

mee

ting

with

pat

ient

.M

ost

inst

rum

ents

ava

ilab

le

for

asse

ssin

g D

DIs

con

sid

er

only

pai

rs o

f dru

gs a

nd d

o no

t ac

coun

t fo

r in

tera

ctio

n in

volv

ing

com

bin

atio

ns o

f thr

ee o

r m

ore

dru

gs s

o th

e ris

k of

DD

I may

be

und

eres

timat

ed.

Mod

erat

e

Non

e

15B

lozi

k et

 al,

2013

62

Ad

ult

YY

YY

Y, g

end

erY

YY

(3 y

ears

)P

reva

lenc

e of

PIM

: 21.

1%95

% C

IY

YY

–H

igh

Age

, num

ber

of

med

icat

ions

, num

ber

of

dis

ease

16C

ahir

et a

l, 20

1463

Eld

erly

-PIM

YY

YY

Y, a

ge, g

end

er,

soci

oeco

nom

ic

stat

us, p

rivat

e he

alth

insu

ranc

e,

com

orb

idity

, num

ber

of

rep

eat

dru

g, s

ocia

l su

pp

ort

and

net

wor

k,

adhe

renc

e

YY

Y (6

mon

ths)

re

tros

pec

tive

stud

y

Pre

vale

nce

of p

oten

tially

IP

was

40.

5%.

95%

CI

YN

YR

ecal

l bia

s d

ue t

o se

lf-re

por

ted

A

DE

Mod

erat

e

Non

e

17Z

imm

erm

ann

et a

l, 20

1318

Eld

erly

-PIM

YY

YY

Y, g

end

er a

ge,

num

ber

of

med

icat

ions

, nu

mb

er o

f dis

ease

, d

epre

ssio

n, e

duc

atio

n

YY

Y (4

.5 y

ears

)A

t b

asel

ine

PIM

 pre

vale

nce

is 2

9%

(848

) acc

ord

ing

to t

he

PR

ISC

US

list

,w

hich

dec

reas

edto

25.

0% (4

64)

4.5

year

s la

ter

and

 21%

ac

cord

ing

to t

he

Bee

rs li

st d

ecre

asin

g af

ter 

4.5

year

s to

17

.1%

 (317

).

95%

CI,

p<

0.05

, O

R a

nd

CI f

or r

isk

fact

ors

YY

Y–

Hig

h

Non

e

Tab

le 2

B

Con

tinue

d

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

22 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Stu

dy

des

ign:

co

hort

Ref

eren

ceQ

ualit

y d

om

ains

12

34

5(a)

5(b

)6(

a)6(

b)

78

910

1112

Ove

rall

qua

lity

18A

mos

et 

al,

2015

67

Eld

erly

-PIM

YY

YY

Y, a

ge, g

end

er,

geog

rap

hica

l loc

atio

n,

num

ber

of m

edic

atio

n

YY

Y (1

yea

r)

retr

osp

ectiv

e st

udy

PIM

pre

vale

nce

28%

, and

ol

der

age

, fem

ale,

num

ber

of

med

icat

ions

incr

ease

ris

k of

PIM

95%

CI,

p<

0.05

YC

anno

t te

llY

May

und

eres

timat

e th

e tr

ue P

IM

pre

vale

nce

bec

ause

the

y d

o no

t ac

coun

t fo

r O

TC.

Mod

erat

e

Num

ber

of c

hron

ic

cond

ition

s

19H

edna

et 

al,

2015

68

Eld

erly

-PIM

YY

YY

NY

YY

(3 m

onth

s)

retr

osp

ectiv

eP

oten

tially

IP p

reva

lenc

e:

42%

AD

R c

ause

d b

y p

oten

tially

IP

.

95%

CI,

p<

0.05

YC

anno

t te

llY

Und

etec

ted

 con

foun

der

sM

oder

ate

Age

, gen

der

, num

ber

of

med

icat

ion,

nu

mb

er o

f chr

onic

co

nditi

on

20M

oria

rty

et a

l, 20

1569

Eld

erly

-PIM

YY

YY

Y, a

ge, g

end

er,

num

ber

of

med

icat

ion,

num

ber

of

chr

onic

con

diti

on,

leve

l of e

duc

atio

n

YY

Y (1

yea

r)P

IM p

reva

lenc

e: 3

6.7%

–64

.8%

.Fe

mal

e, a

ge a

nd h

ighe

r nu

mb

er o

f med

icin

es w

ere

asso

ciat

ed w

ith c

hang

e in

P

IM p

reva

lenc

e.A

ge a

nd h

ighe

r nu

mb

ers

of m

edic

ines

and

ch

roni

c co

nditi

ons

wer

e fo

und

to

be

asso

ciat

ed

with

cha

nge

in P

PO

p

reva

lenc

e.

95%

CI

YY

YLa

ck o

f inf

orm

atio

n on

OTC

from

th

e p

harm

acy

clai

m d

ata.

Hig

h

1 D

id t

he s

tud

y ad

dre

ss a

cle

arly

focu

sed

issu

e?2

Was

the

coh

ort

recr

uite

d in

an

acce

pta

ble

way

?3

Was

the

exp

osur

e ac

cura

tely

mea

sure

d t

o m

inim

ise

bia

s?4

Was

the

out

com

e ac

cura

tely

mea

sure

d t

o m

inim

ise

bia

s?5(

a) H

ave

the

auth

ors

iden

tified

all

imp

orta

nt c

onfo

und

ing

fact

ors?

 Lis

t th

e on

es y

ou t

hink

mig

ht b

e im

por

tant

, tha

t th

e au

thor

mis

sed

.5(

b) H

ave

they

tak

en a

ccou

nt o

f the

con

foun

din

g fa

ctor

s in

the

des

ign

and

/or

anal

ysis

?6(

a) W

as t

he fo

llow

-up

of s

ubje

cts

com

ple

te e

noug

h?6(

b) W

as t

he fo

llow

-up

of s

ubje

cts

long

eno

ugh?

7 W

hat

are

the

resu

lts o

f thi

s st

udy?

8 H

ow p

reci

se a

re t

he r

esul

ts?

9 D

o yo

u b

elie

ve t

he r

esul

ts?

10 C

an t

he r

esul

ts b

e ap

plie

d t

o th

e lo

cal p

opul

atio

n?11

Do

the

resu

lts o

f thi

s st

udy

fit w

ith o

ther

ava

ilab

le e

vid

ence

?12

Wha

t ar

e th

e im

plic

atio

ns o

f thi

s st

udy

for

pra

ctic

e?A

DE

, ad

vers

e d

rug

even

t; A

DR

, ad

vers

e d

rug

reac

tion;

ATC

, Ana

tom

ical

The

rap

eutic

Che

mic

al ;

DD

I, d

rug–

dru

g in

tera

ctio

n; E

D, e

mer

genc

y d

epar

tmen

t; IP

, ina

pp

rop

riate

pre

scrib

ing;

IPE

T, im

pro

ved

p

resc

ribin

g in

the

eld

erly

too

l; N

, no;

OTC

, ove

r-th

e-co

unte

r; P

IM, p

oten

tially

inap

pro

pria

te m

edic

atio

n; P

PO

, pot

entia

l pre

scrib

ing

omis

sion

, U, u

ncle

ar ;Y

, yes

.

Tab

le 2

B

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

23Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Person’s Prescriptions criteria. Johnell and Fastbom46 and Haider et al mentioned two other specific crite-ria.46 48

B. The prevalence of potential prescribing omission (PPO) was measured in five studies for the elder-ly age group only (≥65 years), ranging from 23% to 57%.19 51 65 66 69 PPO was detected by the Screening Tool to Alert doctors to Right Treatment and Assess-ing Care of Vulnerable Elders.

Dosing errorsKoper et al23 found that overdosing and/or underdosing was found in 44% of patients.23

Monitoring errorsMonitoring errors were measured in one study by Ramia and Zeenny,71 who found that 73% of patients had incom-plete therapeutic/safety laboratory-test monitoring tests.71

Other errors: discrepancyOne study found that at least one discrepancy between the medication lists from the pharmacy, the GP or the patient was present in 86.7% of patients.31 In another study, almost half of the patients (47.6%; 95% CI 40.5 to 54.7) had one or more discrepancies in medication infor-mation at discharge.32

The reported point or period prevalence of medica-tion errors in the community settings, including self-re-ported medication errors, prescribing errors (indication, drug–disease interaction, DDI, dosing error and inappro-priate prescribing), monitoring error and discrepancies, had a very wide range from 2% to 94%. Figure 2 shows the medication errors prevalence estimates stratified according to the settings. The highest prevalence was in primary healthcare or general practice (94%).

rIsK fACtOrsRisk factors for medication errors were either related to patients, healthcare professionals and/or medications.

Patient-related risk factorsPatient-related risk factors for the develop-ment of medication errors were discussed in 33 studies.18 20 27 29–33 37 38 40–43 48 49 51–53 55 57 58 60 62 64–67 69 70 73–75

Seven risk factors related to patients were addressed in the included studies: polypharmacy, increased age, number of diseases or comorbidities, female, low level of education, hospital admission and middle family income (table 3).

Several definitions of polypharmacy existed, ranging from prescription of at least three to six medications concurrently. Twenty-six studies showed a positive association between medication error and polyphar-macy,18 27 29–33 37 38 40–42 49 51–53 55 57 58 64–67 69 70 74 of which 18 mentioned the estimated OR ranging from 1.06 to 11.45.18 27 29 30 32 33 37 38 40 42 49 52 57 64–67 69

Older age (≥75 years) was associ-ated with medication errors in 13 studies,18 27 33 38 40 48 49 51 57 65–67 69 of which 10 mentioned the OR ranging from 1.02 to 4.03.18 27 33 38 40 49 57 66 67 69

healthcare professional-related risk factorsNine risk factors related to healthcare professionals for the development of medication errors were identi-fied: more than one physician involved in their care, family medicine/GP specialty, age ≥51 years, male GP, frequent changes in prescription, not considering the prescription of other physicians, inconsistency in the information and outpatient clinic visits (see table 4).27 31 42 49 52 60 67 73 74

Figure 2 Medication errors prevalence estimates according to settings.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

24 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Tab

le 3

M

edic

atio

n er

rors

pat

ient

-rel

ated

ris

k fa

ctor

s

Ris

k fa

cto

rS

tud

ies

wit

h p

osi

tive

as

soci

atio

n (n

)C

ont

rolle

d

stud

ies 

(n)

Co

ntro

lled

fo

rS

pec

ific

info

rmat

ion

OR

or

RR

(95%

or

99%

CI)

p v

alue

s

Age

 ≥75

yea

rs13

(24,

33,

37,

42,

44,

52,

53, 5

5, 6

1, 6

9–71

, 73)

10N

A≥8

0 ye

ars

OR

1.0

21 (9

5% C

I 1.0

18 t

o 1.

023)

p<

0.00

149

Ad

just

ed fo

r ag

e, s

ex, n

umb

er o

f reg

ular

m

edic

ine

and

dia

gnos

ed c

hron

ic c

ond

ition

Old

er a

geO

R 1

.03

(95%

CI 1

.02

to 1

.04)

p<

0.05

69

NA

Old

er a

geO

R 1

.05

(95%

CI 1

to

1.09

) p=

0.04

657

NA

Old

er a

geO

R 1

.06

(95%

CI 1

.0 t

o 1.

13) p

=0.

03718

NA

≥75

year

sO

R 1

.10

(95%

CI 1

.05

to 1

.15)

p<

0.00

133

NA

≥85

year

sO

R 1

.18

(95%

CI 1

.16

to 1

.20)

p<

0.05

40

Ad

just

ed fo

r se

x, a

ge a

nd n

umb

er o

f chr

onic

d

rugs

≥85

year

sO

R 1

.52

(95%

CI 1

.46

to 1

.6)38

NA

≥85

year

sO

R 1

.53

(95%

CI 1

.5 t

o 1.

55) p

<0.

0167

NA

≥85

year

sO

R 1

.79

(95%

CI 1

.19

to 2

.83)

p=

0.00

966

Ad

just

ed fo

r se

x an

d a

ge≥7

5 ye

ars

OR

4.0

3 (9

5% C

I 3.7

9 to

4.2

8) p

<0.

00127

Com

orb

idity

or

num

ber

of

dis

ease

or

chro

nic

cond

ition

dru

g gr

oup

(C

CD

G) s

core

 ≥4

10 (2

4, 2

6, 3

3, 4

4, 4

7, 5

6,

59, 7

3, 7

7, 7

8)3

Ad

just

ed fo

r ag

e, s

ex, n

umb

er o

f reg

ular

m

edic

ines

and

dia

gnos

ed c

hron

ic c

ond

ition

Hig

her

num

ber

of c

hron

ic

cond

ition

sP

PO

: OR

1.4

7 (9

5% C

I 1.3

9 to

1.5

6) p

<0.

0569

NA

CC

DG

sco

re ≥

4O

R 1

.76

(95%

CI 1

.72

to 1

.81)

p<

0.05

40

Ad

just

ed fo

r ag

e an

d s

exD

iagn

osed

dis

ease

 ≥3

OR

6.4

3 (9

5% C

I 3.2

5 to

12.

44) p

<0.

00127

CC

I3

(52,

55,

69)

1N

AC

CI <

2R

R 2

.885

(95%

CI 1

.972

to

4.22

) p=

065

Fem

ale

gend

er10

(33,

35,

47,

52,

53,

 62,

64

, 66,

71,

73)

4A

dju

sted

for

age,

sex

, num

ber

of r

egul

ar

med

icin

es a

nd d

iagn

osed

chr

onic

con

diti

onP

IM: O

R 1

.27

(95%

CI 1

.07

to 1

.5) p

<0.

0569

Ad

just

edO

R 1

.6 (9

9% C

I 1.5

8 to

1.6

4)60

Ad

just

ed fo

r ag

e, s

ex, e

duc

atio

n le

vel,

par

tner

ship

, per

cap

ita in

com

e an

d

occu

pat

ion

Bee

rs 2

003:

OR

2.5

(95%

CI 1

.9 t

o 3.

5)B

eers

201

2: O

R 1

.8 (9

5% C

I 1.3

to

2.5)

29

Ad

just

ed fo

r se

x an

d a

geO

R 2

.49

(95%

CI 2

.29

to 2

.75)

p<

0.00

127

Hea

lth li

tera

cy o

r lo

w e

duc

atio

n2

(52,

79)

1A

dju

sted

for

age,

sex

, typ

e of

res

iden

tial

area

and

com

orb

idity

OR

1.0

9 (9

5% C

I 1.0

7 to

1.1

7)48

Hos

pita

l ad

mis

sion

2 (2

6, 5

6)1

NA

OR

3.3

5 (9

5% C

I 2.4

3 to

4.6

2) p

<0.

0552

Mid

dle

fam

ily in

com

e1

(62)

NA

NA

Con

tinue

d

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

25Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Ris

k fa

cto

rS

tud

ies

wit

h p

osi

tive

as

soci

atio

n (n

)C

ont

rolle

d

stud

ies 

(n)

Co

ntro

lled

fo

rS

pec

ific

info

rmat

ion

OR

or

RR

(95%

or

99%

CI)

p v

alue

s

Pol

ypha

rmac

y26

(22–

24, 3

3, 3

5–37

, 41,

42

, 44–

46, 5

3, 5

5–57

, 59,

61

, 62,

68–

71, 7

3, 7

4, 7

8)

18N

AH

ighe

r nu

mb

er o

f pre

scrib

ed

med

icat

ions

OR

1.0

6 (9

5% C

I 1.3

9 to

1.9

8) p

<0.

00157

Ad

just

ed fo

r ag

e, s

ex, n

umb

er o

f reg

ular

m

edic

ines

and

dia

gnos

ed c

hron

ic c

ond

ition

Hig

her

num

ber

of p

resc

ribed

m

edic

atio

nsP

IM: O

R 1

.2 (9

5% C

I 1.1

7 to

1.2

4) p

<0.

05P

PO

: OR

1.0

4 (9

5% C

I 1.0

1 to

1.0

7) p

<0.

0569

NA

≥4 m

edic

atio

nsO

R 1

.91

(95%

CI 1

.83

to 2

.0) p

<0.

00133

NA

Hig

her

num

ber

of p

resc

ribed

m

edic

atio

nsO

R 1

.99

(95%

CI 1

.80

to 2

.18)

p=

0.00

018

Ad

just

ed fo

r ag

e, s

ex, e

duc

atio

n le

vel,

par

tner

ship

, per

cap

ita in

com

e an

d

occu

pat

ion

≥5 m

edic

atio

nsB

eers

200

3: O

R 2

.9 (9

5% C

I 2.1

to

3.8)

Bee

rs 2

012:

OR

2.7

(95%

CI 2

to

3.6)

29

Ad

just

ed fo

r d

isab

ility

, cor

onar

y ar

tery

dis

ease

, hea

rt fa

ilure

and

oth

er

com

orb

iditi

es

≥5 m

edic

atio

nsIP

: OR

2.9

(95%

CI 1

.5 t

o 5.

8)P

oten

tial m

ajor

DD

I: 3.

8 (9

5% C

I 1.7

to

8.2)

37

Ad

just

ed fo

r ag

e, s

ex, n

umb

er o

f chr

onic

co

nditi

ons

and

num

ber

or

dru

g co

nsum

ed≥3

med

icat

ions

OR

3.2

1 (9

5% C

I 2.7

8 to

3.5

9) p

<0.

00127

Ad

just

ed fo

r ag

e, s

ex, l

engt

h of

hos

pita

l sta

y an

d r

esid

entia

l situ

atio

n≥5

med

icat

ions

OR

3.2

2 (9

5% C

I 1.4

0 to

7.4

2) p

=0.

00632

NA

≥6 m

edic

atio

nsO

R 3

.37

(95%

CI 2

.08

to 5

.48)

p<

0.00

130

NA

≥7 m

edic

atio

nsO

R 4

.528

(95%

CI 4

.52

to 4

.54)

p<

0.00

149

Ad

just

ed fo

r ag

e, s

ex, C

CI,

hist

ory

of

card

iova

scul

ar d

isor

der

and

 his

tory

of

dig

estiv

e d

isor

der

≥5 m

edic

atio

nsO

R 5

.4 (9

5% C

I 3 t

o 9.

7) p

<0.

00164

Ad

just

ed fo

r se

x, a

ge a

nd n

umb

er o

f chr

onic

d

rugs

≥6 m

edic

atio

nsO

R 5

.59

(95%

CI 5

.39

to 5

.80)

38

NA

≥5 m

edic

atio

nsO

R 5

.69

(95%

CI 5

.0 t

o 6.

48) p

<0.

0552

NA

≥6 m

edic

atio

nsS

TOP

P: R

R 6

.837

(95%

CI 4

.155

to

11.2

47)

STA

RT:

RR

2.0

51 (9

5% C

I 1.2

5 to

3.3

67)65

NA

≥10

med

icat

ions

OR

7.3

3 (9

5% C

I 7.1

5 to

7.5

1) p

<0.

0540

NA

≥9 m

edic

atio

nsO

R 7

.43

(95%

CI 3

.20

to 1

7.23

) p<

0.00

166

NA

≥10

med

icat

ions

Mal

e: O

R 8

.2 (9

5% C

I 8 t

o 8.

4)Fe

mal

e: O

R 9

.6 (9

5% C

I 8.2

to

11.2

)42

NA

≥10

med

icat

ions

OR

11.

45 (9

5% C

I 11.

2 to

11.

7) p

<0.

0167

Tab

le 3

C

ontin

ued

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

26 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

Medication-related risk factorsMedication-related risk factors for the development of medication error were multiple medication storage loca-tions used, expired medication present, discontinued medication repeats retained, hoarding of medications, therapeutic duplication,25 no medication administra-tion routine, poor adherence and patients confused by generic and trade names.76 In one study by Johnell and Fastbom,46 multidose drug dispensing users (ie, medi-cines machine-packed into unit-dose bags for each time of administration) were more exposed to all indicators of potentially inappropriate drug.46

Receiving anticoagulant therapy (OR 2.38, 95% CI 2.15 to 2.64) was strongly associated in one study to potential drug–disease interactions.33

The use of OTC and/or prescribed drugs was a risk factor in two additional studies.29 43 The use of OTC medi-cations was associated with PIM; the OR after adjusting for

age, sex, education level, partnership, per capita income and occupation was 2.5 (95% CI 1.7 to 3.6) using Beers 2003 and 1.8 (95% CI 1.2 to 2.5) using Beers 2012.29

errOr-relAteD ADverse eventsError-related adverse events or preventable ADEs were mentioned in six studies.22 28 29 31 32 77 The most frequently reported consequences were ED visits and hospitalisation.

Two methods for detecting ADE were applied: an ADE monitor (ie, using computerised programs composed of rules that identified incidents suggesting that an ADE might be present)22 and using trigger tools to detect ADEs.77

Incidence and/or prevalenceOne study estimated preventable ADE incidence as 15/1000 person-years.22 ACE inhibitors and beta-blockers

Table 4 Medication errors healthcare professional-related risk factors

Risk factor

Studies with positive association (n)

Controlled studies (n) Adjusted for

OR or RR or beta (95% or 99% CI) p values

Age ≥51 years 2 (53, 71) 2 NA OR 1.03 (95% CI 1.01 to 1.06) p<0.0167

NA OR 1.238 (95% CI 1.235 to 1.242) p<0.00149

More than one physician involved in their care

5 (22, 33, 64, 77, 78)

3 NA Beta 0.7 (95% CI 0.5 to 1.0) p=0.03473

Adjusted for age, sex, number of chronic conditions and number or drug consumed

OR 1.39 (95% CI 1.17 to 1.67) p<0.00127

Adjusted for age and number of prescriber

OR 3.52 (99% CI 3.44 to 3.60)60

Male general practitioner 2 (53, 71) 2 NA OR 1.07 (95% CI 1.05 to 1.10) p<0.0167

NA OR 1.206 (95% CI 1.202 to 1.210) p<0.00149

Frequent changes in prescription 1 (77) 1 NA Beta 0.4 (95% CI 0.2 to 0.9) p=0.01973

Not considering the prescription of other physicians

1 (77) 1 NA Beta 1.9 (95% CI 1.1 to 3.2) p=0.01373

Inconsistency in the information 1 (77) 1 NA Beta 4.4 (95% CI 1.3 to 14.8) p=0.01373

Outpatient clinic visit 1 (46) 1 NA 1.4 (male 95% CI 1.3 to 1.4) (female 95% CI 1.3 to 1.6)42

Family medicine/general practice specialty

3 (53, 56, 71) 3 NA OR 1.06 (95% CI 1.03 to 1.10) p<0.0167

NA OR 1.267 (95% CI 1.265 to 1.269) p<0.00149

NA OR 1.46 (95% CI 1.28 to 1.65) p<0.0552

CCI, Charlson Comorbidity Index; IP, inappropriate prescribing; NA, not applicable; PIM, potentially inappropriate medication; PPO, potential prescribing omission; START, Screening Tool to Alert doctors to Right Treatment; STOPP, Screening Tool of Older Person’s Prescriptions.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

27Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

were the most common medications associated with preventable ADE.22 The estimate of the prevalence of preventable ADE was calculated from five studies as detailed below.28 29 31 32 77

All stages of medicines’ management processField et al found the prevalence of error caused by patients leading to an adverse event to be 0.38%, that is, less than 1% of the overall population experienced a medica-tion-related adverse event. They found that the majority of patient errors-related adverse events (n=129) occurred in modifying the medication regimen (42%), admin-istering the medication (32%) or not following clinical advice about medication use (22%).77 The medications associated with more than 10 preventable ADEs were anti-coagulants/antiplatelets, cardiovascular drugs, diuretics, hypoglycaemics and non-opioid analgesics.77

errOr-relAteD ADverse events ACCOrDIng tO MeDICInes’ MAnAgeMent PrOCessPrescribing errorsDrug–drug interactionObreli-Neto et al28 found that DDI-related adverse drug reaction (ADR) occurred in 7% of patients. Warfarin, digoxin, spironolactone and acetylsalicylic acid were the drugs most commonly associated with DDI-related ADRs.28

Potentially inappropriate medicationForty-six per cent of participants reported complaints related to ADEs by interview; 95% of these were caused by prescribed medications.29

Use of inappropriate drugs was associated with an increased risk of nursing home admission, hospitalisa-tion, more outpatient visit days, ED visits and having ADEs or ADRs.44 52 63 68

Other errorsAdverse events (undertreatment due to deletions, ADR due to additions and DDI) related to discrepancy between the medication lists from the patient, the GP or the phar-macy were identified in 24% of patients.31 Two discrep-ancies were categorised as having the potential to cause severe patient harm.32

rIsK fACtOrsRisk factors for the error-related adverse events were discussed in three studies only.28 31 77

Patient-related risk factorsField et al found that the number of regularly sched-uled medications (seven or more medications) (OR 3.3, 95% CI 1.5 to 7.0) and a Charlson Comorbidity Index (CCI) score of 5 or more (OR 15.0, 95% CI 6.5 to 34.5) were both associated with higher risk of patient error leading to preventable ADE.77 Obreli-Neto et al28 found that an age of 80 years or more (OR 4.4, 95 % CI 3.0 to

6.1, p<0.01), a CCI of 4 or more (OR 1.3, 95% CI 1.1 to 1.8, p<0.01) and consumption of five or more medications (OR 2.7, 95% CI 1.9 to 3.1, p<0.01) were associated with the occurrence of DDI-related ADRs. In addition, Tulner et al31 found that the number of medications was signifi-cantly positively correlated with medication discrepancy adverse patient events.

Medication-related risk factorsThe use of medication with narrow therapeutic indices such as warfarin was associated with an increased risk of DDI-related ADRs (OR 1.7, 95% CI 1.1 to 1.9, p<0.01).28

DIsCussIOnsummary of main findingsWe sought to critically review previous studies conducted in the community of the incidence/prevalence of medi-cation errors and associated adverse events and to identify the main risk factors. We identified 60 studies carried out in various countries providing a comprehensive assessment of the available evidence on the epidemiology of medication errors and error-related ADEs in community settings.

No relevant studies on the incidence of medication errors in these settings were found. The reported point or period prevalence of medication errors in community settings had a very wide range (ie, 2%–94%). This wide range appears, at least in part, to be due to the inconsis-tency in the definitions of the medication errors used in the studies, differences in populations studied, methodol-ogies employed for error detection and different outcome measures. More than half (37 studies) of the resulting studies were regarding the prescription of inappropriate drugs within the prescribing error stage in an elderly age group using different criteria. The comparison of those criteria is challenging due to the difference in medication use, consumption and availability of those medications to patients between countries. Further work is needed to review errors occurring at administration and dispensing stages of the medicines’ management process.

As for preventable ADEs, which may in some cases occur as a result of medication errors, only one study reported error-related adverse events incidence, measured as 15/1000 person-years.22 The prevalence of preventable ADE was further reported in five other studies and varied according to the medication error type that resulted in the adverse event.

The most common patient-related risk factors for both medication errors and preventable ADEs mentioned were the number of medications used by the patient and the increased age of patients.

strengths and limitationsThe main strength of this systematic review is that a rigorous and transparent process has been employed, which included no language restrictions, an indepen-dent screening of titles and abstracts, independent data extraction and critical appraisal of included studies by

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

28 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

two reviewers. It is the first review undertaken within community settings. The use of the ICPS conceptual framework,17 which provides a comprehensive defini-tion of each concept and type of error in the medicines’ management process, is a further strength.

However, several limitations need to be considered. First, despite the thorough process, no data were found regarding the dispensing error stage. This might be due to the lack of a ‘dispensing error’ key term in our search strategy, although ‘medication therapy management’ as a key term was included. However, 10 studies on dispensing errors were excluded because they failed to satisfy the inclusion criteria on one or more counts. Second, no data were found regarding the administration error stage. However, 14 studies on administration errors were also excluded for the same previous reason. Third, this system-atic review had different outcomes reported in a variety of ways using different tools and methodology, which made combining results in one meta-analysis difficult. Lastly, the studies addressed risk factors adjusted for different confounders, which makes it difficult to generate compa-rable estimates and/or make causal inferences about whether the harm resulted from the medication error.

Comparison of the findings with previous studiesThe definitional variation issue is supported by another two reviews.78 79 Other systematic reviews focusing on the safety of primary care contexts only have identified studies with vastly different prevalence estimates of the rates of medication errors. These reflect differences in definitions, sampling strategy and populations studied; none have investigated the risk factors for medication errors.80 81

Implications for research, policy and practiceThere is a need for (1) improvement in the quality of research in this area—it is important that all researchers provide a standardised set of outcome measures of medi-cation errors or internationally accepted terminology and definitions of key concepts; (2) training and moni-toring of healthcare professionals with the involvement of medication safety pharmacists in the community; (3) empowering and educating the patients and the public, particularly those with chronic diseases and polyphar-macy, to increase their knowledge of medication safety with a record of the current medication list for each patient; (4) patient use of tools and technology particu-larly for monitoring and follow-up; and (5) encourage the reporting of medication errors, administration errors and dispensing errors.82 This would strengthen the quality of research, improve the development of strategies to detect and prevent these errors, and provide a safer environ-ment for the community to self-care safely.

COnClusIOnsWe found a very wide variation in the medication error and error-related adverse events rate between studies, which, at least in part, reflects differences in their definitions,

methodologies employed for error detection or clinical heterogeneity, that is, differences in populations studied and different outcome measures. Most of the studies were conducted on elderly populations in economically devel-oped countries. There is therefore clearly a need to extend this work to low-income and middle-income countries, particularly give the WHO’s recent launch of a Global Medication Safety Challenge.82 83 Furthermore, most studies focused only on inappropriate prescribing with relatively little attention to other stages such as admin-istration and dispensing. The most common patient and medication-related risk factors for both medication errors and preventable ADEs were the number of medications used by the patient, increased age and receiving anticoag-ulant therapy. The most common healthcare profession-al-related risk factor for medication error was when more than one practitioner was involved in the care of patients and care provision by family medicine and GP specialities.

This study has identified important limitations and discrepancies in the methodology used to study medica-tion errors and error-related ADEs in community settings. These findings need to be considered in the context of designing future research related to medication safety. More research is needed in the areas of incidence of medication errors, administration error and dispensing errors and reporting. Researchers should use a more consistent set of definitions and outcomes in order to facilitate collation and synthesis of data.

ethICs AnD DIsseMInAtIOnThe systematic review protocol was published in BMJ Open on 31 August 2016 and is registered with PROS-PERO, an international prospective register of systematic reviews.11 12 It is reported using PRISMA. Trial registration number: CRD42016048126.

Author affiliations1Centre for Population Health Sciences, The University of Edinburgh, Edinburgh, UK2Department of Clinical Pharmacy, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia3Department of Paediatrics, Faculty of Medicine, King Saud University, Riyadh, Saudi Arabia4Department of Pharmacy, Pharmacology and Postgraduate Medicine, School of Life and Medical Sciences, University of Hertfordshire, Hertfordshire, UK5College of Pharmacy, Clinical Pharmacy Department, Taibah University, Madinah, Al Madinah, Saudi Arabia6The Global Health Academy, Centre for Population Health Sciences, The University of Edinburgh, Edinburgh, UK7Saudi Food and Drug Authority, Riyadh, Saudi Arabia8Usher Institute of Population Health Sciences and Informatics, The University of Edinburgh, Edinburgh, UK

Acknowledgements We are grateful to Marshall Dozier for her help with formulating the search strategy; Kathrin Cresswell and Andrea Fuentes Pacheco for non-English studies' translation; and the experts in the field for unpublished and in progress work and experts within the Farr Institute.

Contributors GAA conceived the idea for this review, conducted the systematic literature search, study inclusion, data extraction and quality assessment. NAS participated in the study inclusion, data extraction and quality assessment. MAM participated in data extraction. NA participated in data extraction and quality

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

29Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

assessment. GAA led the writing and drafting of the manuscript, and this was commented on critically by AS, EG, HA and NAS.

funding The systematic review protocol is part of GAA’s PhD study at The University of Edinburgh. King Saud University, College of Pharmacy funded the scholarship. AS is supported by the Farr Institute. The project was financially supported through Prince Abdullah bin Khalid Celiac Disease Research Chair, Vice Deanship of Research Chairs, King Saud University and The University of Edinburgh.

Competing interests None declared.

Patient consent Not required.

Provenance and peer review Not commissioned; externally peer reviewed.

Data sharing statement All available data can be obtained by contacting the corresponding author.

Open Access This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http:// creativecommons. org/ licenses/ by- nc/ 4. 0/

© Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

referenCes 1. Mark S, Little J, Geller S. et alPrinciples and practices of medication

safety. In: DiPiro JT TR, Yee GC, Matzke GR, Wells BG, Posey L, . eds. Pharmacotherapy: a pathophysiologic approach. New York: McGraw-Hill, 2011.

2. Einarson TR. Drug-related hospital admissions. Ann Pharmacother 1993;27(7-8):832–40.

3. Krähenbühl-Melcher A, Schlienger R, Lampert M, et al. Drug-related problems in hospitals: a review of the recent literature. Drug Saf 2007;30:379–407.

4. Kongkaew C, Hann M, Mandal J, et al. Risk factors for hospital admissions associated with adverse drug events. Pharmacotherapy 2013;33:827–37.

5. Aitken M, Gorokhovich L. Advancing the responsible use of medicines: applying levers for change. SSRN Electronic Journal 2012.

6. In: Kohn LT, Corrigan JM, Donaldson MS, eds. To err is human: building a safer health system. Washington (DC): National Academies Press, 2000.

7. Sheikh A, Panesar SS, Larizgoitia I, et al. Safer primary care for all: a global imperative. Lancet Glob Health 2013;1:e182–e183.

8. Cresswell KM, Panesar SS, Salvilla SA, et al. Global research priorities to better understand the burden of iatrogenic harm in primary care: an international delphi exercise. PLoS Med 2013;10:e1001554.

9. Monitor. Moving healthcare closer to home: Literature review of clinical impacts. 2015 https://www. gov. uk/ government/ uploads/ system/ uploads/ attachment_ data/ file/ 459268/ Moving_ healthcare_ closer_ to_ home_ clinical_ review. pdf

10. Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;349:g7647.

11. Assiri GA, Grant L, Aljadhey H, et al. Investigating the epidemiology of medication errors and error-related adverse drug events (ADEs) in primary care, ambulatory care and home settings: a systematic review protocol. BMJ Open 2016;6:e010675.

12. Assiri GA, Grant L, Aljadhey H, et al. Investigating the epidemiology of medication errors and error-related adverse drug events (ADEs) in primary care, ambulatory care and home settings: a systematic review protocol. BMJ Open 2016;6:e010675–8.

13. Citical Appraisal Skills Programme checklist for cohort studies. 2015 http://www. casp& uk. net/ wp& content/ uploads/ 2011/ 11/ CASP_ Cohort_ Appraisal_ Checklist_ 14oct10. pdfwebcite

14. Joanna Briggs Institute. Checklist for critical appraisal of descriptive studies. 2015 ht tp://jo annab rigg s.org/asset s/do cs/j bc/o pera tions/ criticalAppraisalForms/ JB C_ Form_ Crit Ap_ DescCase. pdf

15. Thomsen LA, Winterstein AG, S⊘ndergaard B, et al. Systematic review of the incidence and characteristics of preventable adverse drug events in ambulatory care. Ann Pharmacother 2007;41:1411–26.

16. Taché SV, Sönnichsen A, Ashcroft DM. Prevalence of adverse drug events in ambulatory care: a systematic review. Ann Pharmacother 2011;45(7-8):977–89.

17. Wolrd Health Organization. The conceptual framework for the international classification for patient safety. Final Technical report. Geneva: World Health Organization, 2009.

18. Zimmermann T, Kaduszkiewicz H, Van Den Bussche H, et al. Potentially inappropriate medication in elderly primary care patients: A retrospective, longitudinal analysis. [German]. Bundesgesundheitsbl 2013;56:941–9.

19. Candela Marroquín E, Mateos Iglesia N, Palomo Cobos L. [Adequacy of medication in patients 65 years or older in teaching health centers in Cáceres, Spain]. Rev Esp Salud Publica 2012;86:419–34.

20. Lu CY, Roughead E. Determinants of patient-reported medication errors: a comparison among seven countries. Int J Clin Pract 2011;65:733–40.

21. Sears K, Scobie A, Mackinnon NJ. Patient-related risk factors for self-reported medication errors in hospital and community settings in 8 countries. Can Pharm J 2012;145:88–93.

22. Gandhi TK, Seger AC, Overhage JM, et al. Outpatient adverse drug events identified by screening electronic health records. J Patient Saf 2010;6:91–6.

23. Koper D, Kamenski G, Flamm M, et al. Frequency of medication errors in primary care patients with polypharmacy. Fam Pract 2013;30:313–9.

24. Dallenbach MF, Bovier PA, Desmeules J. Detecting drug interactions using personal digital assistants in an out-patient clinic. QJM 2007;100:691–7.

25. Vuong T, Marriott JL. Unnecessary medicines stored in homes of patients at risk of medication misadventure. Journal of Pharmacy Practice and Research 2006;36:16–20.

26. Obreli Neto PR, Vieira JC, Teixeira DRA, et al. Potential risks in drug prescriptions to elderly: A cross-sectional study in the public primary health care system of Ourinhos micro-region, Brazil. Latin American Journal of Pharmacy 2011;30:629.

27. Obreli Neto PR, Nobili A, Marusic S, et al. Prevalence and predictors of potential drug-drug interactions in the elderly: a cross-sectional study in the brazilian primary public health system. Journal of Pharmacy & Pharmaceutical Sciences 2012;15:344–54.

28. Obreli-Neto PR, Nobili A, de Oliveira Baldoni A, et al. Adverse drug reactions caused by drug–drug interactions in elderly outpatients: a prospective cohort study. Eur J Clin Pharmacol 2012;68:1667–76.

29. Baldoni AdeO, Ayres LR, Martinez EZ, et al. Factors associated with potentially inappropriate medications use by the elderly according to Beers criteria 2003 and 2012. Int J Clin Pharm 2014;36:316–24.

30. Secoli SR, Figueras A, Lebrão ML, et al. Risk of potential drug-drug interactions among Brazilian elderly: a population-based, cross-sectional study. Drugs Aging 2010;27:759–70.

31. Tulner LR, Kuper IMJA, Frankfort SV, et al. Discrepancies in reported drug use in geriatric outpatients: Relevance to adverse events and drug-drug interactions. Am J Geriatr Pharmacother 2009;7:93–104.

32. Cornu P, Steurbaut S, Leysen T, et al. Discrepancies in medication information for the primary care physician and the geriatric patient at discharge. Ann Pharmacother 2012;46(7-8):983–91.

33. Mand P, Roth K, Biertz F, et al. Drug-disease interaction in elderly patients in family practice. Int. Journal of Clinical Pharmacology and Therapeutics 2014;52:337–45.

34. Indermitte J, Reber D, Beutler M, et al. Prevalence and patient awareness of selected potential drug interactions with self-medication. J Clin Pharm Ther 2007;32:149–59.

35. Mahmood M, Malone DC, Skrepnek GH, et al. Potential drug-drug interactions within Veterans Affairs medical centers. Am J Health Syst Pharm 2007;64:1500–5.

36. Gagne JJ, Maio V, Rabinowitz C. Prevalence and predictors of potential drug–drug interactions in Regione Emilia-Romagna, Italy. J Clin Pharm Ther 2008;33:141–51.

37. Lapi F, Pozzi C, Mazzaglia G, et al. Epidemiology of suboptimal prescribing in older, community dwellers: a two-wave, population-based survey in Dicomano, Italy. Drugs Aging 2009;26:1029–38.

38. Nobili A, Pasina L, Tettamanti M, et al. Potentially severe drug interactions in elderly outpatients: results of an observational study of an administrative prescription database. J Clin Pharm Ther 2009;34:377–86.

39. Guthrie B, Makubate B, Hernandez-Santiago V, et al. The rising tide of polypharmacy and drug-drug interactions: population database analysis 1995–2010. BMC Med 2015;13:74.

40. Maio V, Yuen EJ, Novielli K, et al. Potentially inappropriate medication prescribing for elderly outpatients in emilia romagna, Italy. Drugs Aging 2006;23:915–24.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

30 Assiri GA, et al. BMJ Open 2018;8:e019101. doi:10.1136/bmjopen-2017-019101

Open Access

41. de Oliveira Martins S, Soares MA, Foppe van Mil JW, et al. Inappropriate drug use by Portuguese elderly outpatients--effect of the Beers criteria update. Pharm World Sci 2006;28:296–301.

42. Pugh MJ, Hanlon JT, Zeber JE, et al. Assessing potentially inappropriate prescribing in the elderly Veterans Affairs population using the HEDIS 2006 quality measure. J Manag Care Pharm 2006;12:537–45.

43. Saab YB, Hachem A, Sinno S, et al. Inappropriate medication use in elderly lebanese outpatients: prevalence and risk factors. Drugs Aging 2006;23:743–52.

44. Zuckerman IH, Langenberg P, Baumgarten M, et al. Inappropriate drug use and risk of transition to nursing homes among community-dwelling older adults. Med Care 2006;44:722–30.

45. Bregnhøj L, Thirstrup S, Kristensen MB, et al. Prevalence of inappropriate prescribing in primary care. Pharmacy World & Science 2007;29:109–15.

46. Johnell K, Fastbom J. Multi-dose drug dispensing and inappropriate drug use: A nationwide register-based study of over 700 000 elderly. Scand J Prim Health Care 2008;26:86–91.

47. Berdot S, Bertrand M, Dartigues J-F, et al. Inappropriate medication use and risk of falls – A prospective study in a large community-dwelling elderly cohort. BMC Geriatr 2009;9:30.

48. Haider SI, Johnell K, Weitoft GR, et al. The influence of educational level on polypharmacy and inappropriate drug use: a register-based study of more than 600,000 older people. J Am Geriatr Soc 2009;57:62–9.

49. Lai H-Y, Hwang S-J, Chen Y-C, et al. Prevalence of the prescribing of potentially inappropriate medications at ambulatory care visits by elderly patients covered by the Taiwanese National Health Insurance program. Clin Ther 2009;31:1859–70.

50. Ryan C, O’Mahony D, Kennedy J, et al. Appropriate prescribing in the elderly: an investigation of two screening tools, Beers criteria considering diagnosis and independent of diagnosis and improved prescribing in the elderly tool to identify inappropriate use of medicines in the elderly in primary care in Ireland. J Clin Pharm Ther 2009;34:369–76.

51. Ryan Cristà n, O'Mahony D, Kennedy J, et al. Potentially inappropriate prescribing in an Irish elderly population in primary care. Br J Clin Pharmacol 2009;68:936–47.

52. Akazawa M, Imai H, Igarashi A, et al. Potentially inappropriate medication use in elderly Japanese patients. Am J Geriatr Pharmacother 2010;8:146–60.

53. Zaveri HG, Mansuri SM, Patel VJ. Use of potentially inappropriate medicines in elderly: A prospective study in medicine out-patient department of a tertiary care teaching hospital. Indian J Pharmacol 2010;42:94–8.

54. Barnett K, McCowan C, Evans JMM, et al. Prevalence and outcomes of use of potentially inappropriate medicines in older people: cohort study stratified by residence in nursing home or in the community. BMJ Qual Saf 2011;20:275–81.

55. Chang C-B, Chen J-H, Wen C-J, et al. Potentially inappropriate medications in geriatric outpatients with polypharmacy: application of six sets of published explicit criteria. Br J Clin Pharmacol 2011;72:482–9.

56. Leikola S, Dimitrow M, Lyles A, et al. Potentially inappropriate medication use among Finnish non-institutionalized people aged ≥65 years: a register-based, cross-sectional, national study. Drugs Aging 2011;28:227–36.

57. Lin Y-J, Peng L-N, Chen L-K, et al. Risk factors of potentially inappropriate medications among older patients visiting the community health center in rural Taiwan. Arch Gerontol Geriatr 2011;53:225–8.

58. Zhang Y-J, Liu W-W, Wang J-B, et al. Potentially inappropriate medication use among older adults in the USA in 2007. Age Ageing 2011;40:398–401.

59. Haasum Y, Fastbom J, Johnell K. Institutionalization as a risk factor for inappropriate drug use in the elderly: a swedish nationwide register-based study. Ann Pharmacother 2012;46:339–46.

60. Nyborg G, Straand J, Brekke M. Inappropriate prescribing for the elderly—a modern epidemic? Eur J Clin Pharmacol 2012;68:1085–94.

61. Yasein NA, Barghouti FF, Irshaid YM, et al. Elderly patients in family practice: poly pharmacy and inappropriate prescribing - Jordan. International Medical Journal 2012;19:302–6.

62. Blozik E, Rapold R, von Overbeck J, et al. Polypharmacy and potentially inappropriate medication in the adult, community-dwelling population in switzerland. Drugs Aging 2013;30:561–8.

63. Cahir C, Bennett K, Teljeur C, et al. Potentially inappropriate prescribing and adverse health outcomes in community dwelling older patients. Br J Clin Pharmacol 2014;77:201–10.

64. Weng M-C, Tsai C-F, Sheu K-L, et al. The impact of number of drugs prescribed on the risk of potentially inappropriate medication among outpatient older adults with chronic diseases. QJM: An International Journal of Medicine 2013;106:1009–15.

65. Castillo-Páramo A, Clavería A, Verdejo González A, et al. Inappropriate prescribing according to the STOPP/START criteria in older people from a primary care setting. Eur J Gen Pract 2014;20:281–9.

66. Vezmar Kovačević S, Simišić M, Stojkov Rudinski S, et al. Potentially inappropriate prescribing in older primary care patients. PLoS One 2014;9:e95536.

67. Amos TB, Keith SW, Del Canale S, et al. Inappropriate prescribing in a large community-dwelling older population: a focus on prevalence and how it relates to patient and physician characteristics. J Clin Pharm Ther 2015;40:7–13.

68. Hedna K, Hakkarainen KM, Gyllensten H, et al. Potentially inappropriate prescribing and adverse drug reactions in the elderly: a population-based study. Eur J Clin Pharmacol 2015;71:1525–33.

69. Moriarty F, Bennett K, Fahey T, et al. Longitudinal prevalence of potentially inappropriate medicines and potential prescribing omissions in a cohort of community-dwelling older people. Eur J Clin Pharmacol 2015;71:473–82.

70. Woelfel JA, Patel RA, Walberg MP, et al. Use of potentially inappropriate medications in an ambulatory medicare population. The Consultant Pharmacist 2011;26:913–9.

71. Ramia E, Zeenny R. Completion of therapeutic and safety monitoring tests in Lebanese outpatients on chronic medications: a cross-sectional study. Patient Prefer Adherence 2014;8:1195–204.

72. Adams RJ, Tucker G, Price K, et al. Self-reported adverse events in health care that cause harm: a population-based survey. Med J Aust 2009;190:484–8.

73. Mira JJ, Orozco-Beltran D, Perez-Jover V, et al. Physician patient communication failure facilitates medication errors in older polymedicated patients with multiple comorbidities. Fam Pract 2013;30:56–63.

74. Pit SW, Byles JE, Cockburn J. Prevalence of self-reported risk factors for medication misadventure among older people in general practice. J Eval Clin Pract 2008;14:203–8.

75. Mosher HJ, Lund BC, Kripalani S, et al. Association of health literacy with medication knowledge, adherence, and adverse drug events among elderly veterans. J Health Commun 2012;17:241–51.

76. Sorensen L, Stokes JA, Purdie DM, et al. Medication management at home: medication risk factor prevalence and inter-relationships. J Clin Pharm Ther 2006;31:485–91.

77. Field TS, Mazor KM, Briesacher B, et al. Adverse drug events resulting from patient errors in older adults. J Am Geriatr Soc 2007;55:271–6.

78. Alsulami Z, Conroy S, Choonara I. Medication errors in the middle east countries: a systematic review of the literature. Eur J Clin Pharmacol 2013;69:995–1008.

79. Karthikeyan MBT, Khaleel MI, Sahl M, et al. A systematic review on medication errors. International Journal of Drug Development and Research 2015:7–4.

80. Olaniyan JO, Ghaleb M, Dhillon S, et al. Safety of medication use in primary care. Int J Pharm Pract 2015;23:3–20.

81. Panesar SS, deSilva D, Carson-Stevens A, et al. How safe is primary care? A systematic review. British Medical Journal Quality and Safety 2015;0:1–10.

82. Donaldson LJ, Kelley ET, Dhingra-Kumar N, et al. Medication without harm: who's third global patient safety challenge. Lancet 2017;389:1680–1.

83. Sheikh A, Dhingra-Kumar N, Kelley E, Kieny MP, et al. The third global patient safety challenge: tackling medication-related harm. Bull World Health Organ 2017;95:546–546A.

84. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE prevention study group. JAMA 1995;274:29–34.

85. What is a Medication Error? National coordinating council for medication error reporting and prevention. http://www. nccmerp. org/ about- medication- errors

86. U.S. Food and Drug Administration. What are over-the-counter (OTC) drugs and how are they approved? http://www. fda. gov/ AboutFDA/ Transparency/ Basics/ ucm194951. htm

87. Gallagher P, Barry P, O'Mahony D. Inappropriate prescribing in the elderly. J Clin Pharm Ther 2007;32:113–21.

88. Moher D, Liberati A, Tetzlaff J, et al. PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097.

on June 3, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2017-019101 on 5 May 2018. D

ownloaded from

1BMJ Open 2019;9:e019101corr1. doi:10.1136/bmjopen-2017-019101corr1

Open access

Correction: What is the epidemiology of medication errors, error-related adverse events and risk factors for errors in adults managed in community care contexts? A systematic review of the international literature

Assiri GA, Shebl NA, Mahmoud MA, et al. What is the epidemiology of medication errors, error-related adverse events and risk factors for errors in adults managed in community care contexts? A systematic review of the international literature. BMJ Open 2018;8:e019101. doi: 10.1136/bmjopen-2017-019101

The author, Ghadah Asaad Assiri would like to change the affiliation order in this article.

At present, the order of affiliation for Ghadah Asaad Assiri is:

1. Centre for Population Health Sciences, The University of Edinburgh, Edinburgh, UK.

2. Department of Clinical Pharmacy, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia. The revised order of affiliation is:

1. Department of Clinical Pharmacy, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia.

2. Centre for Population Health Sciences, The University of Edinburgh, Edinburgh, UK.

As per the revised order, the affiliation for Nouf Aloudah is

1. Department of Clinical Pharmacy, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia.

Open access This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http:// creativecommons. org/ licenses/ by- nc/ 4. 0/.

© Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

BMJ Open 2019;9:e019101corr1. doi:10.1136/bmjopen-2017-019101corr1

Correction