RESEARCH ARTICLE Open Access

Concurrent use of alcohol interactivemedications and alcohol in older adults:a systematic review of prevalence andassociated adverse outcomesAlice E. Holton1*, Paul Gallagher1, Tom Fahey2 and Gráinne Cousins1

Abstract

Background: Older adults are susceptible to adverse effects from the concurrent use of medications and alcohol.The aim of this study was to systematically review the prevalence of concurrent use of alcohol and alcohol-interactive (AI)medicines in older adults and associated adverse outcomes.

Methods: A systematic search was performed using MEDLINE (PubMed), Embase, Scopus and Web of Science (January1990 to June 2016), and hand searching references of retrieved articles. Observational studies reporting on the concurrentuse of alcohol and AI medicines in the same or overlapping recall periods in older adults were included. Twoindependent reviewers verified that studies met the inclusion criteria, critically appraised included studies andextracted relevant data. A narrative synthesis is provided.

Results: Twenty studies, all cross-sectional, were included. Nine studies classified a wide range of medicines as AI usingdifferent medication compendia, thus resulting in heterogeneity across studies. Three studies investigated any medicationuse and eight focused on psychotropic medications. Based on the quality assessment of included studies, the mostreliable estimate of concurrent use in older adults ranges between 21 and 35%. The most reliable estimate of concurrentuse of psychotropic medications and alcohol ranges between 7.4 and 7.75%. No study examined longitudinal associationswith adverse outcomes. Three cross-sectional studies reported on falls with mixed findings, while one study reported onthe association between moderate alcohol consumption and adverse drug reactions at hospital admission.

Conclusions: While there appears to be a high propensity for alcohol-medication interactions in older adults, there is alack of consensus regarding what constitutes an AI medication. An explicit list of AI medications needs to be derived andvalidated prospectively to quantify the magnitude of risk posed by the concurrent use of alcohol for adverse outcomes inolder adults. This will allow for risk stratification of older adults at the point of prescribing, and prioritise alcohol screeningand brief alcohol interventions in high-risk groups.

Keywords: Alcohol, Drug interaction, Alcohol interactive, Older adult, Psychotropic medicines, Adverse outcomes

* Correspondence: aliceholton@rcsi.ie1School of Pharmacy, Royal College of Surgeons in Ireland (RCSI), Dublin 2,IrelandFull list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Holton et al. BMC Geriatrics (2017) 17:148 DOI 10.1186/s12877-017-0532-2

BackgroundBy 2050, older adults aged ≥60 years are expected to ac-count for 34% of the population in Europe [1]. While al-cohol consumption changes over the life-course, with adecline in consumption in older age, recent evidencefrom nine UK based prospective cohort studies, haveshown that drinking occasions tend to become more fre-quent among older adults [2]. There is also evidence of acohort effect, with successive birth cohorts reporting anincrease in alcohol consumption across all age-groups,including among older adults [3].Even at relatively low levels of alcohol consumption,

older adults can be vulnerable to harm, with physio-logical changes exacerbating these harms [4, 5]. Further-more, older adults experience a disproportionate burdenof alcohol related-harm; in England between 2009 and2010, adults aged ≥65 years accounted for approximately44% (461,400) of alcohol-related hospital admissions yetcomprised of only 17% of the population [6, 7]. Alcohol-related deaths were also highest among those aged 55 to74 years [4].Furthermore, the use of multiple medicines is increas-

ing in older adults [8–11]. A recent Irish study reportedan increase in polypharmacy from 17.8 to 60.4% betweenthe years 1997–2012 in older adults aged ≥65 years [11].Certain medications have the potential to interact withalcohol; these medications are referred to as alcoholinteractive (AI) medications [12]. They may interact withalcohol by altering the metabolism (pharmacokinetic) oreffects (pharmacodynamic) of alcohol and/or the medi-cation [12]. Certain interactions may occur with any al-cohol consumption, whereas other interactions mayfollow a dose response, with the risk or severity of aninteraction increasing with increasing levels of alcohol[13]. AI medications, when combined with alcohol, in-crease the risk of medical complications such ashypoglycaemia, hypotension, sedation, gastrointestinalbleeds and liver damage, in older adults [5, 12]. For ex-ample, older adults are vulnerable to the sedating effectsof alcohol and when combined with central nervous sys-tem (CNS) agents, such as psychotropic medications,older adults have an increased risk of sedation anddrowsiness [5]. Psychotropic medicines include antide-pressants, sedatives/hypnotics, stimulants and neurolep-tics, all of which act on the CNS and are commonlyprescribed to older adults [14]. Similarly, concurrent useof alcohol with cardiovascular agents, such as vasodila-tory agents, increases the risk of hypotension in olderadults [5], with concurrent use with non-steroidal anti-inflammatory drugs (NSAIDs) increasing the risk ofgastrointestinal bleeds [12].While a recent systematic review has assessed the

prevalence of concurrent alcohol use and prescriptionsedative-hypnotic medicines in middle-aged and older

adults, [15] there have been no systematic reviews onthe prevalence of concurrent use of alcohol and alcoholinteractive medications beyond psychotropic medica-tions and associated adverse outcomes in older adults.An older review did investigate the potential risk ofcombining alcohol with medications in older adultshowever the focus of the review was on the pharmacol-ogy and mechanism of action involved in alcohol-medication interactions and potential clinical implica-tions of these interactions [5]. Therefore this study aimsto systematically review the prevalence of concurrentuse of alcohol and alcohol interactive medications inolder adults and associated adverse outcomes.

MethodsThis systematic review was performed according to Pre-ferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines [16].

Search strategyA comprehensive systematic search was performed usingMEDLINE (PubMed), Embase, Scopus and Web of Sci-ence. A combination of the following keywords and MeSHterms were used: “ethanol”, “alcohol”, “drug interactions”,“drug alcohol interaction” and “aged”. This search wassupplemented by a search in Google Scholar and by handsearching references of retrieved articles. The search wasrestricted to English language articles and articles pub-lished since January 1990 to June 2016.

Study selection and data extractionStudies were included if they met the following eligibilitycriteria: Observational studies reporting on the concur-rent use of alcohol and alcohol interactive (AI) medi-cines in the same or overlapping recall periods in olderadults. Studies also had to report on the quantity or fre-quency of alcohol consumption. We excluded studieswhich exclusively sampled patients with specific ill-nesses, or those seeking treatment for alcohol use disor-ders (AUD) or illicit drug use.Title and abstracts of identified studies were reviewed

by one reviewer (AH) to determine potential eligibility.Full text articles were then reviewed by two reviewers(AH/GC) for those studies considered eligible from title/abstract, or when it was unclear whether a study met theinclusion criteria. The following data were extracted bytwo reviewers (AH/GC): year of publication, country,study sample, study design, measurement and definitionof alcohol interactive (AI) medications, measurement ofalcohol consumption, prevalence of alcohol use and AImedication use and prevalence of concurrent alcoholand AI medication use. Adverse outcomes associatedwith concurrent use of alcohol and AI medications werealso extracted if reported. Any uncertainty in relation to

Holton et al. BMC Geriatrics (2017) 17:148 Page 2 of 16

study eligibility and data extraction was resolved throughdiscussion between two reviewers (GC/AH).

Critical appraisalThe risk of bias was evaluated, by two reviewers (AH/GC), using an adapted form of the Newcastle Ottawa co-hort scale (NOS) [17]. This amended NOS scale allowedfor the evaluation of cross-sectional studies, focussingon the risk of selection bias and information bias, specif-ically misclassification bias of both exposures (alcoholand AI medications) and the outcome (concurrent useof alcohol and AI medications) across included studies.

ResultsOf the 546 citations identified from this search strategy,108 full text articles were assessed for eligibility, with 20studies meeting the inclusion criteria [14, 18–36] (Fig. 1).

Study characteristicsTen of the included studies were conducted in Europe[14, 19, 21–23, 25, 27–29, 36], eight in North America[18, 20, 24, 30–32, 34, 35] and two in Australia [26, 33].(Table 1) All studies were cross-sectional [14, 18–36].Study settings varied across studies with; community-dwelling [14, 19, 21–27, 31, 34–36], both communitydwelling or living independently in care facilities [32],general populations [20, 28, 33], hospital setting [29], re-tirement communities [18] and participants signed up toa pharmaceutical assistance contract for the elderly [30]reported.Sample sizes varied from 311 to 83,321 participants

[18, 30] . Men and women were included in all studies,with the exception of one Australian study which only in-cluded men [26]. Nine studies reported on a wide range ofprescription and/or over the counter (OTC) medicines withpotential to interact with alcohol [18–21, 24, 27, 29–31] .Three studies investigated any medication use during therecall period [23, 33, 35], and a further eight focused onpsychotropic medications [14, 22, 25, 26, 28, 32, 34, 36]. Ofthe eight studies focusing on psychotropic medications, fiveinvestigated psychotropic medications alone [22, 25, 26, 34,36] and three studies also included analgesics [14, 28, 32].Of the nine studies focusing on a wide range of AI med-

ications, all studies classified central nervous system(CNS) agents as AI medications (Table 1). Consistent withthose studies investigating psychotropic medications[14, 22, 25, 26, 28, 32, 34, 36], the following drugclasses were classified as AI medications, sedatives/hypnotics [18–21, 24, 27, 30, 31], antidepressants[18–21, 27, 30, 31], opioids/narcotics [18–21, 27, 30,31], anticonvulsants [19–21, 24, 27, 30, 31] and anti-psychotics [19–21, 27, 29, 30]. After CNS agents, car-diovascular medicines (CVS) were the most commonAI medicines [18–21, 24, 27, 29–31], followed by

antidiabetic drugs [18, 20, 21, 24, 27, 29–31], warfarin[18–21, 24, 27, 30, 31], gastrointestinal agents [18–21,24, 30, 31], non-steroidal anti-inflammatory drugs(NSAIDs) [18–21, 29–31], antibiotics/anti-infectives [21,24, 27, 29–31] and anti-histamines [19–21, 30].

Quality assessmentThe methodological quality of the included studies is de-tailed in Table 2. The external validity was high in 4studies, as they reported on random population samplesof community dwelling older adults aged ≥60 or≥65 years [21, 24, 27] or in the case of Breslow et al., arandom population sample with oversampling of olderadults aged ≥60 years [20]. Ten studies were consideredto have moderate external validity; five studies sampledrandom community dwelling older adults but with agerestrictions which may have introduced selection bias[14, 19, 25, 26, 31] such as including adults aged be-tween 53 and 75 years [25] or only including adults≥75 years [19]. A further five studies reported on ran-dom population samples of community dwelling adultswith subgroup analysis of older adults, however they didnot report oversampling of older adults [22, 23, 28, 33, 34].The six remaining studies were considered to have poorexternal validity, as the risk of selection bias is consideredto be high [18, 29, 30, 32, 35, 36]. For example, Pringle etal. recruited enrollees in the Pennsylvania pharmaceuticalassistance contract for the elderly, which may not be repre-sentative of all older adults as the members are older(mean age 78.8 years) and more likely to be female andwhite with multiple chronic conditions [30].Internal validity was assessed by evaluating the potential

risk of misclassification bias for both exposure and out-come across studies. Ten studies were considered to have alow risk of misclassifying exposure to medications whichhave the potential to interact with alcohol, as they usedprescription claims data or in house inventories where theyrecorded details from the labels of medication containersor prescriptions and provided references supporting the in-clusion of medicines as potentially alcohol interactive(Table 2) [14, 20, 21, 25, 26, 28, 30–32, 34]. The potentialfor misclassification bias was considered high in theremaining studies as they relied on self-report for medica-tion exposure [18, 22, 23, 27, 29, 33, 35, 36] and/or did notprovide references supporting the inclusion of medicinesas potentially alcohol interactive [19, 23, 24, 29, 33, 35].While all studies relied on self-reported alcohol con-

sumption, thus introducing potential biases in recall andreporting, 11 studies were considered to have a lowerrisk of bias as they reported on both quantity and fre-quency of alcohol consumption within a specified recallperiod, ranging from 1 week to 12 months (Table 2)[14, 18–23, 25, 26, 31, 34]. The risk of misclassificationbias, specifically underestimating exposure to alcohol, was

Holton et al. BMC Geriatrics (2017) 17:148 Page 3 of 16

considered to be higher in the remaining studies as theyused quantity or frequency measures alone [24, 32, 35],did not specify the recall period [24, 27, 30, 35], restrictedmeasurements to wine consumption [29, 36] and used avery narrow recall period “last working day or last week-end” [28]. Ascertainment of exposure to alcohol was un-clear in one study [33].In relation to outcome assessment, no study directly

measured the concurrent use of alcohol and AI medica-tions, rather all studies inferred concurrent use. Althoughpossible for all studies, the risk of misclassifying the out-come of concurrent use was considered lowest in thosestudies identified as having a low risk of misclassifying

exposure, who used the same recall period for both expo-sures or inferred concurrent use based on alcohol con-sumption within a specific recall period, ranging between1 week and 12 months, and current or regular medicationuse (Table 2) [14, 20, 21, 25, 26, 31, 34].

Summary of findingsAs noted in Table 1, all estimates relate to studies fromNorth America, Europe and Australia.

Alcohol consumptionThe prevalence of alcohol consumption ranged between57 and 63% in studies reporting on nationally representative

Fig. 1 Flow diagram of studies included in this review

Holton et al. BMC Geriatrics (2017) 17:148 Page 4 of 16

Table

1Characteristicsof

includ

edstud

ies

Reference

Setting

Participants:N

,sex,m

ean

age(±sd),rang

eStud

yde

sign

(surveymod

e)Measuremen

t&D

efinition

:

AIM

edication(s)

Alcoh

olCon

sumptionQuantity/Frequ

ency;

thresholds

applied

Adams1995

[18]

UnitedStates

(US),

retirem

entcommun

ityreside

nts

N=311

23%

men

83(±

6years)

NR

Cross

sectional(Mailedself-

repo

rted

survey)

Regu

laror

occasion

aluseof

“highrisk”

med

sin

last

6mon

ths:NSA

IDS,aspirin

,sed

atives,narcotics,

antid

epressants,anti-h

ypertensives,antacids,H2

blockers,w

arfarin

&med

sforcong

estivehe

artfailure,

gout

ordiabetes.Referen

cesource

notrepo

rted

Khavariq

uestionn

aire:q

uantity

–fre

quen

cy,last6mon

ths,Non

e,1–6drinks/w

eekor

≥7drinks/w

eek

1drink=12

oz.ofbe

er,5

oz.ofwine,

3oz.offortified

wineor

1.5oz.of

hard

liquo

r.

Aira

2005

[19]

Finland,

commun

itydw

ellingolde

radults

N=521

27%

men

81(±

4.4years)

Rang

e:75–95.7years

Cross

sectional(Nurse

interview;

prescriptio

ns&containe

rs)

Current

useof

med

ications

having

potential

interactions

with

alcoho

l:Acetaminop

hen

(Paracetam

ol),anticon

vulsants,antidep

ressants,TCAs,

antih

istamines,b

enzodiazep

ines,H

2receptor

antago

nists,ne

urolep

tics,nitrates,N

SAIDs,op

iates&

warfarin

.Referen

cesource

notrepo

rted

Beverage

specificqu

antity-fre

quen

cy,

last12

mon

ths:Non

e,1–7un

its/w

eek,

>7un

its/w

eek

1unit=11-12gof

alcoho

l

Breslow

2015

[20]

US,ge

neralp

opulation

≥20

years

≥65

years

N=7183

≥20

yearsN=26,657

51%

men

NR

Cross

sectional(Interviewsin-

home,med

icationcontaine

rs)

Use

ofAIp

rescrip

tionmed

ication,pastmon

th;

iden

tifiedusing(i)

Drugs.com

,(ii)Carem

ark.com,(iii)

Health

line.com,(iv)DailyMed

databases&(v)

references

from

3pu

blications

[12,43,44]

includ

ing:

cardiovascular

agen

ts,C

NSagen

ts,coagu

latio

nmod

ifiers,GI,metabolic,p

sychothe

rape

utic&

respiratory

agen

ts

Quantity

–fre

quen

cyin

last12

mon

ths:

Non

e,forwom

enof

allage

s&men

>65

years:mod

erateconsum

ption:>0–7drinks/

week&he

avierdrinkers(>

7drinks/w

eek).

Formen

20–64yearsmod

erate:>0–14

drinks/w

eek&he

avier>14drinks/w

eek

Cou

sins

2013

[21]

Ireland

,com

mun

itydw

ellingolde

radults

≥60

years

N=3815

46.6%

males

69.7(±7.3years)

Rang

e:60–99years

Cross

sectional(Nurse

led

interviewsin-hom

e;med

ications

review

ed&self-completed

questio

nnaire)

Current

orregu

laruseof

med

ications

with

potential

tointeract

with

alcoho

lide

ntified

usingStockley’s

DrugInteractions,BritishNationalFormularyand

Irish

Med

icines

Form

ulary:cardiovascular

agen

ts,

CNSagen

ts,antihistamines,b

lood

agen

ts,

antid

iabe

ticagen

ts,anti-infectives,G

Idrugs,

immun

omod

ulators&musclerelaxants

Quantity-freq

uency,past6mon

ths:Non

e,Ligh

t/mod

erate(≤

4drinks/day

or10

drinks

/week)andhe

avydrinkers(>4drinks/day

or10

drinks/week)

1drink=10

gof

alcoho

l

DelRio

1996

[23]

Spain,commun

itydw

ellingadults

≥16

years

≥66

yearsN=3003

≥16

years

N=21,084

48%

men

NR

Cross

sectional(survey

with

interview)

Med

icationuse,past2weeks.N

oreferencesource

repo

rted

Beverage

specificqu

antity-fre

quen

cy,p

ast

2weeks

DelRio

2002

[22]

Spain,commun

itydw

ellingadults

≥16

years

≥66

years=1025

≥16

yearsN=6396

48%

men

NR

Cross

sectional(survey

with

interview)

Benzod

iazepine

use,past2weeks

Beverage

specificqu

antity-fre

quen

cypast2

weeks:Low

(men

:≤21

units/w

eek,wom

en:

≤14

units/w

eek),m

oderate(m

en:22–50

units/w

eek,wom

en:15–35

units/w

eek)&

high

consum

ption(m

en:>

50un

its/w

eek&

wom

en:>

35un

its/w

eek)

1un

it=10

gof

alcoho

l

Du2008

[14]

Germany,commun

itydw

ellingolde

radults

N=1605

45.2%

men

NR

Rang

e:60–79years

Cross

sectional(survey

&interview;p

rescrip

tions

ororiginalcontaine

rs)

Any

psycho

trop

icmed

icationusein

past7days:A

TCne

rvou

ssystem

drug

scode

dN00.(Exclud

eddrug

scode

dN02Baspirin

&paracetamol,excep

tfor

N02BA

71)O

piatecode

ines

used

asantitussiveswere

Beverage

specificqu

antity-fre

quen

cy,p

ast

12mon

ths:Prob

lem

use(riskydrinking

):daily

consum

ption≥10

gforwom

en&≥

20gformen

Holton et al. BMC Geriatrics (2017) 17:148 Page 5 of 16

Table

1Characteristicsof

includ

edstud

ies(Con

tinued)

merge

dwith

N02A&op

iatesforGIcon

ditio

ns(A07D)wereno

trecorded

Forster

1993

[24]

US,commun

itydw

ellingolde

radults

N=667

39.9%

men

74.1(±6.6years)

NR

Cross

sectional(survey

&

interview;m

edicationcontaine

rs)

Prescriptio

nor

OTC

med

ications

curren

tlyor

past

mon

th:A

Idrugs

defined

bystud

yclinicalteam

:OTC

painkillers,anti-h

ypertensives,d

iuretics,OTC

cold

prep

arations,arthritismed

ications,heartmed

ications,

antib

iotics,mindalterin

gmed

ications,che

stpain

med

ications,p

ainmed

ications,d

iabe

tesmed

ications,

ulcermed

ications,sleep

ingpills,O

TCnasalsprays,

steroids,b

lood

agen

ts,insulin,seizure

med

ications

&OTC

asthmamed

ication

Freq

uency:ne

ver,rarely,som

etim

es&

regu

larly

Ilomaki

2008

[25]

Finland,

commun

itydw

ellingolde

radults

N=1,774

48.1%

men

63years(NR)

Rang

e:53-73years

Cross

sectional(mailedsurvey

&interview;p

rescrip

tions

brou

ght

ifanyprob

lems)

Regu

laruseof

psycho

trop

icdrug

sat

timeof

interview:antipsychotics,anxiolytics,hypn

otics,

sedatives,antidep

ressants&combinatio

ns

Beverage

specificqu

antity-fre

quen

cypast

12mon

ths:fre

quen

t≥2tim

es/w

eek,bing

e(m

en≥5un

its/occasion:wom

en≥4un

its/

occasion

)&he

avydrinking

(men

:>14

units

/week:wom

en>7un

its/w

eek

1un

it=12

gof

alcoho

l

Ilomaki

2013

[26]

Australia,com

mun

itydw

ellingolde

rmen

≥70

yearsN=1,705

100%

men

NR

NR

Cross

sectional(Interview:

med

ications

brou

ght)

Antidep

ressants(includ

ingSSRIS,TC

As,MAOIsand

othe

rs)&

SADs(includ

ingbe

nzod

iazepine

sand

benzod

iazepine

-like

hypn

otics)

Beverage

specificqu

antity-fre

quen

cypast

12mon

ths:daily

(7days/w

eek),b

inge

(≥5

drinks

atleaston

ce/m

onth),he

avy(>2

drinks/day)&prob

lem

drinkers(CAGEscore

≥2).N

ondrinkers:former

orne

verdrinkers

1drink=

10gof

alcoho

l

Immon

en2012

[27]

Finland,

commun

itydw

ellingolde

radults

≥65

yearsN=1,395

35.5%

men

78.7years(NR)

NR

Cross

sectional(mailedsurvey)

Current

prescribed

med

ications:Swed

ish,Finn

ish

InteractionX-referencing(SFINX)

interaction

database

iden

tifiedsign

ificant

drug

-alcoh

olinteractions:m

etronidazole,tinidazole,disulfiram,

griseo

fulvin,p

razosin,metform

in&tacrolim

us.C

NS

agen

ts,hypog

lycaem

icsandwarfarin

Beverage

specificqu

antity-fre

quen

cy:A

trisk

(>7drinks/week,or

≥5drinks

onatypical

drinking

dayor

≥3drinks

severaltim

es/

week),m

oderatedrinkers(atleast1drink/

mon

thbu

t<7drinks/w

eek)&minim

al/non

-users(<1drink/mon

th)

1drink=12

gof

alcoho

l

John

2007

[28]

Germany,ge

neral

popu

latio

n≥20

years

NR

≥20

yearsN=4290

49%

men

NR

Rang

e:20-79years

Cross

sectional;(Interview;

containe

ror

participant

provided

inform

ationon

curren

tmed

ications)

Use

ofsedative,hypn

oticsor

anxiolytics(SHA)

med

icines

includ

ing:

barbitu

rates(&

derivatives),

benzod

iazepine

s(&

derivatives),carbam

ates,

pipe

ridined

ione

derivatives,hypno

ticsor

anxiolytics

&combinatio

nsof

sedatives

andhypn

otics.SO

grou

p:includ

edSH

Aandop

ioid,p

ast7days

Quantity

ofbe

verage

specificalcoho

lconsum

edon

lastworking

dayandlast

weekend

:Riskdrinking

(men

:>30

g/day&

wom

en>20g/day)

Lagn

aoui

2001

[36]

France,com

mun

itydw

ellingolde

radults

≥65

yearsN=3767

41.7%

men

NR

NR

Cross

sectional;(Interview

atho

me;adho

cqu

estio

nnaire;

visualinspectio

n)

Benzod

iazepine

use

Quantity:D

ailywineconsum

ptiondivide

dinto:I)no

neii)mod

erateup

to0.25

lday

-1

&iii)he

avy>0.25

lday

-1(ld

ay−1 =

litres

consum

eddaily)

Ond

er2002

[29]

Italy,hospital

admission

sam

ong

olde

radults

N=22,778

45%

men

70.3(±

16years)

NR

Cross

sectional(qu

estio

nnaire

atadmission

s)Med

ications

takenpriorto

admission

,during

hospitalstay&prescribed

atdischarge:drug

sclassifiedaccordingto

riskof

causingadversedrug

reactio

nsusingNaranjo

algo

rithm

Quantity

ofdaily

consum

ptionpriorto

hospital

admission

:(daily

wineun

its).Non

-drin

kers&

mod

eratedrinkerson

ly,heavy

alcoho

lusers

wereexclud

ed

Pringle2005

[30]

US,Penn

sylvania

Pharmaceutical

N=83,321

19%

men

Cross

sectional(survey

&prescriptio

nclaimsdata)

Filledprescriptio

ns,p

ast45

days:d

rugs

with

potentialfor

alcoho

linteractio

ns(using

First

Current

quantity-fre

quen

cy:

Ligh

t:1-7drinks/m

onth

Holton et al. BMC Geriatrics (2017) 17:148 Page 6 of 16

Table

1Characteristicsof

includ

edstud

ies(Con

tinued)

AssistanceCon

tract

fortheElde

rlyprescriptio

ndrug

users

78.8(±

6.9years)

Rang

e:65-106

years

DatabankInc.Drugs)with

oneof

thefollowing

warning

sconsidered

AI:

1.May

causedrow

sine

ss;alcoh

olmay

intensify

this

effect

2.Dono

tdrinkalcoho

lwhe

ntaking

thismed

ication

3.May

causedrow

sine

ssanddizziness;alcoho

lmay

intensify

thiseffect

4.Limitalcoho

lwhiletaking

thismed

ication;daily

useof

alcoho

lmay

increase

riskof

stom

achbleeds

Mod

erate:8-30

drinks/m

onth

Heavy:>

30drinks/m

onth

Qato2015

[31]

US,commun

itydw

ellingolde

radults

N=2975

48.6%

men

NR

Rang

e:57-85years

Cross

sectional(In

house

interviews&ob

servation

med

icationbo

ttles)

Current

orregu

laruse(dailyor

weekly)with

potentialtointeract

with

alcoho

l(Thom

son

Micromed

exdatabase).Drugs

catego

rised

according

toseverityof

interaction:

1.Con

traind

icated

:Drugs

contraindicatedforuse

with

alcoho

l2.Major:m

aybe

life-threaten

ingor

requ

iremed

ical

interven

tionto

preven

tserio

usadverseeven

ts3.Mod

erate:may

resultin

exacerbatio

nof

individu

als

cond

ition

orrequ

irealternativetherapy

4.Minor:lim

itedclinicaleffects

Quantity-freq

uencylast3mon

ths:

Non

-reg

ular

drinkers:(no

neor

<1

drink/week)

Ligh

t–reg

ular:(≥1drink/week&1

drink/day)

Heavy-reg

ular:(≥1drink/week&2-3

drinks/day)

Bing

e-regu

lar:(≥

1drink/week&≥4

drinks/day)

Sheahan

1995

[32]

US,commun

itydw

elling&

inde

pend

entliving

cong

regate

care

facilityolde

rreside

nts

≥55

yearsN=1028

26.9%

men

73.5(±

9.47

years)

NR

Cross

sectional(In

house

interviews&med

ication

containe

rlabe

ls)

Psycho

trop

icmed

ications

inpastyear:sed

atives/

hypn

otics,anxiolytics/tranqu

ilizers,antidep

ressants

&prescriptio

npain

med

ications

Freq

uency,pastyear.A

verage

numbe

rof

times/w

eek;mon

thor

year

consum

edalcoho

l.Num

berof

drinking

days

inpastyear

calculated

Veldhu

izen

2009

[34]

Canada,commun

itydw

elling≥15

years

NR

≥15

years:N=36,984

49%

men

N

Cross

sectionalstudy

(Survey

with

interview,m

edication

containe

rs)

Use

ofbe

nzod

iazepine

s(N03AE,N05BA

andN05CD

),no

n-be

nzod

iazepine

hypn

otics(zop

iclone

,eszopiclon

e,zolpidem

andzaleplon

)du

ringpast2

days

amon

gthosewho

repo

rted

usein

past

12mon

ths

Quantity-freq

uencyon

each

oftheprevious

7days

amon

gthosedrinking

inpastyear:

Nodrinking

,mod

erate&he

avydrinking

(>14

drinks/w

eekformen

/>9drinks/w

eek

forwom

en)o

rbing

edrinking

(>4drinks/

dayforwom

enor

5drinks/day

formen

)1drink=

1bo

ttleor

canof

beer/glassof

draft,1glassof

wine/cooler

or1½

ozof

liquo

r

Won

g2016

[35]

America,commun

itydw

ellingolde

radults

N=2444

33%

men

76.84(±8.13

years)

Rang

e:60-103

years

Cross

sectional(Face

toface

interviews)

Prescriptio

nsandOTC

med

ications

takendaily.

Num

berof

med

ications:low

≤1,mod

erate2-4&

high

≥5

Quantity-freq

uency

Abstainers:no

toalcoho

l&0drinkers/

mon

th,Light:yes

toalcoho

l&<29

drinks/

mon

th,M

oderate:yesto

alcoho

l&30

drinks

/mon

th,H

eavy:yes

toalcoho

l&>31

drinks/

mon

th1drink=

14gof

alcoho

l

Holton et al. BMC Geriatrics (2017) 17:148 Page 7 of 16

samples of community dwelling older adults (Table 3)[21, 24, 27]. Prevalence estimates of alcohol consump-tion were generally lower in studies sampling restrictedage-groups, for example estimates ranged between 33.7and 44% for studies restricted to older adults aged>70 years [26] and >75 years [19]. Similarly, prevalence es-timates of alcohol consumption were generally lower inthose studies considered to have a high risk of selectionbias [18, 29, 30, 32, 35]. For example, 20% of older adultsregistered on the Pennsylvania pharmaceutical assistancecontract for the elderly reported alcohol consumption[30]. All studies reporting on gender differences identifieda higher prevalence of alcohol consumption in men[19–21, 25, 29, 31, 36].

Alcohol interactive (AI) medicinesExposure to alcohol interactive medications varied acrossstudies; nationally representative studies of communitydwelling older adults using objective measures of AI ex-posure estimate exposure at between 72 and 79% amongthe total study samples (Table 3) [20, 21]. Studies with ob-jective measures of exposure to psychotropic medicationsin community dwelling older adults reported prevalenceestimates of between 11.5–20% among the total studysamples [14, 25, 26]. In contrast to alcohol consumption,

use of psychotropic medications was significantly higherin women than men [14, 22, 25, 28, 32].

Concurrent use of alcohol and AI medications amongolder adultsEleven studies reported on the concurrent use of alcoholand alcohol interactive medications among the total studysamples (Table 3) [14, 18–20, 23, 24, 31–33, 35, 36]. Eightof the 11 studies reported on alcohol consumption and awide range of medicines with potential to interactwith alcohol, with the prevalence of concurrent useranging between 18 and 39% among the total studysamples [18–20, 23, 24, 31, 33, 35]. Only two of thesestudies were considered to have a low risk of mis-classification bias for concurrent use [20, 31] . Bre-slow et al. estimated the prevalence of concurrent useamong older adults at 35% [20]. Concurrent use washighest for cardiovascular agents (28%) followed bymetabolic agents (17%), CNS agents (10%) and coagu-lation modifiers (10%). Approximately 7.75% of allolder adults were identified as concurrent users of al-cohol and psychotropic medications [20]. Qato et al.reported a prevalence estimate of 21%, with potentialalcohol-medication interactions in older adults agedbetween 57 and 84 years being significantly more

Table 2 Results of the critical appraisal of included studies

Study Representativenessof sample

Ascertainment ofalcohol consumption

Ascertainment ofAI medications(classified)

Ascertainment ofAI medications(measured)

Assessment ofthe outcome(concurrent use)

Study total:

a) Cross Sectional Studies Outcome of Interest Concurrent use of alcohol and medications:

Adams et al. 1995 [18] ++ + +++

Aira et al. 2005 [19] + ++ ++ +++++

Breslow et al. 2015 [20] ++ ++ ++ ++ ++ ++++++++++

Cousins et al. 2014 [21] ++ ++ ++ ++ ++ ++++++++++

Del Rio et al. 1996 [23] + ++ +++

Del Rio et al. 2002 [22] + ++ ++ +++++

Du et al. 2008 [14] + ++ ++ ++ ++ +++++++++

Forster et al. 1995 [24] ++ + ++ +++++

Ilomaki et al. 2008 [25] + ++ ++ ++ ++ +++++++++

Ilomaki et al. 2013 [26] + ++ ++ ++ ++ +++++++++

Immonen et al. 2012 [27] ++ + ++ +++++

John et al. 2007 [28] + + ++ ++ ++++++

Lagnaoui et al. 2001 [36] ++ ++

Onder et al. 2002 [29] + +

Pringle et al. 2005 [30] + ++ ++ +++++

Sheahan et al. 1995 [32] + ++ ++ +++++

Qato et al. 2015 [31] + ++ ++ ++ ++ +++++++++

Swift et al. 2007 [33] + +

Veldhuizen et al. 2009 [34] + ++ ++ ++ ++ +++++++++

Wong et al. 2016 [35]

Holton et al. BMC Geriatrics (2017) 17:148 Page 8 of 16

Table

3Summaryof

results;p

revalenceof

alcoho

lcon

sumption,

alcoho

linteractivemed

icationuseandconcurrent

useam

ongolde

radults

Stud

yPrevalen

ceof

alcoho

lconsum

ptionin

olde

radults(gen

der);

Heavy/Problem

drinking

(gen

der)

Prevalen

ceof

alcoho

linteractivemed

icationuse:

Con

curren

tuserepo

rted

amon

g:Totalsam

pleof

olde

radults

Con

curren

tuserepo

rted

amon

g:AIm

edicationusers

Con

curren

tuserepo

rted

amon

g:Current

drinkers

Stud

iesrepo

rtingon

awiderang

eof

alcoho

linteractive(AI)med

icines

(n=13

stud

ies)

Adams

1995

[18]

47%

drankalcoho

linprevious

6mon

ths

Heavy

(>7drinks/w

eek):8%

80%

used

oneor

moreof

thefollowingin

thelast6

mon

ths:NSA

IDS,aspirin

,sedatives,narcotics,

antid

epressants,anti-

hype

rten

sives,antacids,H

2

blockers,w

arfarin

&med

sforcong

estivehe

artfailure,

gout

ordiabetes

38%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications,6%

repo

rted

concurrent

heavyalcoho

lconsum

ptionandAI

med

ications

Overalldrinkers:80%

used

anAIm

edication(50%

used

anti-

hype

rten

sives,27%

used

aspirin

,20%

used

NSA

IDs,18%

used

chroniche

artfailure

drug

s,11%

used

sedatives;5%

used

narcotics,

5%used

warfarin

,4%

used

diabeticdrug

s,3%

used

antid

epressants,3%

used

drug

sforgo

ut)

Heavy

drinkers:80%

used

anAI

med

ication(48%

used

anti-

hype

rten

sives,28%

used

aspirin

,16%

used

NSA

IDs,16%

used

chroniche

artfailure

drug

s,4%

used

sedatives,4%

used

narcotics,

4%used

warfarin

,8%

used

diabetic

drug

s,12%

used

antid

epressants,

4%used

drug

sforgo

ut)

Aira

2005

[19]

44%

drankalcoho

linprevious

12mon

ths(66%

ofmen

&37%

ofwom

en)

Heavy

(>7un

its/w

eek):7%

ofmen

&0%

ofwom

en

90%

used

oneor

moreof

thefollowingregu

larly

oras

need

ed:acetaminop

hen,

anticon

vulsants,antidep

ressants,

antih

istamines,b

enzodiazep

ines,

histam

ineH2receptor

agon

ist,

neurolep

tics,nitrates,N

SAIDs,

opiatesor

warfarin

39%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications,1.9%

repo

rted

concurrent

heavyalcoho

lconsum

ptionandAI

med

ications.C

oncurren

talcoho

land

specificAI

med

ications:7.5%

acetam

inop

hen,0.19%

anticon

vulsants,5.16%

antid

epressants,0.76%

antih

istamines,11%

benzod

iazepine

s,1.15%

histam

ineH2receptor

agon

ist,2.29%

neurolep

tics,

23%

nitrates,29%

NSA

IDs,

3.8%

opiates&3.6%

warfarin

OverallAIm

edication

users:44%

drankalcoho

l(36%

acetam

inop

hen

users,17%

anticon

vulsants

users,40%

antid

epressants

users,21%

antih

istamines

users,38%

benzod

iazepine

users,43%

histam

ineH2

receptor

agon

istusers,

25%

neurolep

ticusers,

43%

nitrateusers,46%

NSA

IDusers,38%

opiate

users&40%

warfarin

users)

Overalldrinkers:88%

used

AI

med

ications

Breslow

2015

[20]

47%

drankalcoho

linprevious

12mon

ths(55%

ofmen

&39.7%

ofwom

en)

78.6%

used

oneor

moreof

the

following:

cardiovascular

agen

ts,

CNSagen

ts,coagu

latio

nmod

ifiers,GIage

nts,metabolic

agen

ts,p

sychothe

rape

utic

agen

ts,respiratory

agen

ts

35%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications.C

oncurren

talcoho

land

specificAI

med

ications:28%

cardiovascular

agen

ts,10%

OverallAIm

edication

users:45%

drankalcoho

l(44%

cardiovascular

agen

tsusers,40%

CNS

agen

tusers(34%

anticon

vulsants,40%

Overalldrinkers:77.8%

took

AI

med

ications

(61.3%

used

cardiovascular

agen

ts,

22%

used

CNSagen

t(3.9%

used

anticon

vulsants,4.6%

used

anxiolytic/sed

ative/hypn

otic,4.8%

Holton et al. BMC Geriatrics (2017) 17:148 Page 9 of 16

Table

3Summaryof

results;p

revalenceof

alcoho

lcon

sumption,

alcoho

linteractivemed

icationuseandconcurrent

useam

ongolde

radults(Con

tinued)

CNSagen

ts(1.8%

anticon

vulsants,2%

anxiolytic/sed

ative/hypn

otic,

2%narcotics,2.2%

NSA

IDs),

3%coagulationmod

ifiers,

2.2%

GIage

nts,16.9%

metabolicagen

ts,3.9%

psycho

therapeutic

agen

ts(3.8%

antid

epressants),2.1%

respiratory

agen

ts

anxiolytic/sed

ative/

hypn

otic,43%

NSA

ID),

44%

coagulationmod

ifier

users,43%

GIage

ntusers,

43%

metabolicagen

tusers,

41%

psycho

therapeutic

agen

tusers(42%

antid

epressants),48%

respiratory

agen

tusers)

used

NSA

ID),6%

used

coagulation

mod

ifier,4.7%

used

GIage

ntusers,

36.5%

used

metabolicagen

t,9.6%

used

psycho

therapeutic

agen

t(9.2%

used

antid

epressants),4.6%

used

respiratory

agen

t)

Cou

sins

2013

[21]

62.8%

drankalcoho

lpreviou

s6mon

ths(72%

ofmen

&59%

ofwom

en)

Heavy

(>4drinks/day

or>10

drinks/w

eek):20%

(32%

ofmen

&11%

ofwom

en)

CAGE:8%

(12.2%

ofmen

&4%

ofwom

en)

72%

took

oneor

moreof

the

following:

cardiovascular

agen

ts,

CNSagen

ts,antihistamines,

Bloo

d,antid

iabe

ticagen

ts,anti-

infectives,G

Iage

nts,

immun

omod

ulatorsor

muscle

relaxants

OverallAIm

edication

usersconcurrent

useof

alcoho

l:60%

drankalcoho

l(60%

cardiovascular

agen

tusers,53.5%

CNSagen

tusers(59%

ofNSA

IDusers,

54%

hypn

oticusers,44%

anxiolyticusers,52.9%

antid

epressantusers),

66.9%

antih

istamineusers,

58.5%

bloo

dmed

ication

users,54%

antid

iabe

ticagen

tusers,47%

anti-

infectiveusers,50%

GI

agen

tusers,51%

immun

omod

ulator

users

&80.3%

musclerelaxant

users)

AIm

edicationusers

concurrent

heavyalcoho

lconsum

ption:25%

antih

istamineusers,20%

cardiovascular

agen

tusers,

20%

bloo

d(anti-coagu

lant

oranti-platelet)users,20%

anti-diabeticagen

tusers,

16%

CNSagen

tusers(13%

usersof

anti-ep

ileptic

agen

ts;13%

antip

sychotic

agen

ts;13%

hypn

oticusers

&18%

antid

epressantusers)

DelRio

1996

[23]

App

roximately20%

drank

alcoho

latleaston

cepe

rday

inthepast2weeks

75–80%

took

oneor

more

med

icationin

previous

2weeks

18%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications

Forster

1993

[24]

57.1%

repo

rted

usingalcoho

l16.9%

admitted

todrinking

enou

ghto

become“ligh

theade

d”

Not

Repo

rted

25%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications.C

oncurren

talcoho

land

specificAI

Holton et al. BMC Geriatrics (2017) 17:148 Page 10 of 16

Table

3Summaryof

results;p

revalenceof

alcoho

lcon

sumption,

alcoho

linteractivemed

icationuseandconcurrent

useam

ongolde

radults(Con

tinued)

med

ications:19%

OTC

analge

sics,6.9%

antih

ypertensives,5.4%

diuretics;4.3%

OTC

cold

prep

arations,2.1%

mind

alterin

gdrug

s,1.5%

diabetes

pills,1.5%

prescriptio

npain

med

ication,

1.2%

sleeping

pills,0.6%

prescriptio

nbloo

dthinne

rs,

0.6%

insulin

&0.3%

seizure

med

ications

Immon

en2013

[27]

62.6%

drankalcoho

lAtriskdrinking

(>7drinks/w

eek,

or≥5drinks

ontypicald

rinking

day,or

≥3drinks

severaltim

es/

week):7.9%

42%

took

oneor

moreof

the

following:

metronidazole,

tinidazole,disulfiram,g

riseo

fulvin,

prazosin,m

etform

in&tacrolim

us.

CNSagen

ts,hypog

lycaem

icsand

warfarin

OverallAIm

edication

users:62.2%

drankalcoho

lHeavy

orat

riskdrinkers:42.2%

%took

AIm

edications

(2.2%

used

antip

sychotics,4.4%

used

anti-

depressant,6.7%

used

anxiolytics,

11.1%

used

hypn

otics/sedatives,

5.6%

used

anti-ep

ileptics,3.3%

used

opioids,11.1%

used

warfarin

&13.3%

used

metform

in)

Ond

er2002

[29]

54.2%

drank≤40

gof

wine/day

priorto

hospitaladm

ission

(68.

1%of

men

&42.8%

ofwom

en)

27%

used

diuretics,23%

digo

xin,

17.7%

calcium

channelb

lockers,

16%

ACEinhibitors,15%

aspirin

&anti-platelets,9%

oral

hypo

glycaemicagen

ts,6%

NSA

IDs,

6%antib

iotics,5.2%

nitrates,5%

insulin,4%

steroids

&3.2%

antip

sychotics

Overallwinedrinkers:26%

used

diuretics,3.8%

oralhypo

glycaemic

agen

ts,2.6%

antip

sychotics&1.8%

insulin

Pringle

2005

[30]

20.3%

drankalcoho

lHeavy

(>30

drinks/m

onth):1.2%

77.4%

used

onor

moreAI

med

ication

OverallAIm

edication

usersconcurrent

useof

alcoho

l:19%

drankalcoho

l(18.4%

cardiovascular

agen

tusers,18%

CNSagen

tusers

(20%

ofNSA

IDusers,16.8%

anxiolytic/hypno

tic/sed

ative

users&

16%

antid

epressant

users),20%

antih

istamine

users,14%

bloo

dmed

icationusers,13%

antid

iabe

ticagen

tusers,

16%

anti-infectiveusers,

14%

GIage

ntusers&16%

musclerelaxant

users)

Qato2015

[31]

41%

wereregu

lardrinkersin

the

past3mon

ths(59.3%

ofmen

&40.7%

ofwom

en)

Heavy

(2–3

drinks/day):19.7%

57.7%

used

atleaston

eAIm

edication.

21%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications

Overalldrinkers(Reg

ular

drinkers):

51%

used

AIm

edications

(8.4%

used

antid

iabe

ticagen

ts,6.6%

used

analge

sics,2.4%

used

narcotics,5.3%

used

acetam

inop

hen,26.7%

used

Holton et al. BMC Geriatrics (2017) 17:148 Page 11 of 16

Table

3Summaryof

results;p

revalenceof

alcoho

lcon

sumption,

alcoho

linteractivemed

icationuseandconcurrent

useam

ongolde

radults(Con

tinued)

aspirin

,18.9%

used

psycho

trop

icmed

ication,8.5%

used

antid

epressants,6%

used

anxiolytic/sed

ative/hypn

otics&

4%used

warfarin

)

Swift

2007

[33]

18%

drankalcoho

ldailyin

past

12mon

ths

87.3%

used

atleaston

eAIm

edicationin

last

24h

35.4%

repo

rted

concurrent

useof

alcoho

land

AI

med

ications

inprevious

24h

Won

g2016

[35]

38%

consum

edalcoho

lHeavy

(>31

drinks/m

onth):6%

83%

repo

rted

anymed

ication

use

31%

repo

rted

concurrent

useof

alcoho

land

med

ications&1.4%

repo

rted

concurrent

heavyalcoho

lconsum

ptionandmed

ication

use

Stud

iesrepo

rtingon

psycho

trop

icmed

icines

(n=8stud

ies)

DelRio

2002

[22]

Not

Repo

rted

13.4%

used

benzod

iazepine

sin

previous

2weeks

Overallbe

nzod

iazepine

usersconcurrent

useof

alcoho

l:23%

(56–66

years),

approx.15%

(66–75

years)

&approx.10%

(>75

years)

Du2008

[14]

47.3%

drankalcoho

latleaston

cein

thelast-w

eek

Heavy

drinking

(≥10

g/dayfor

wom

en/≥20

g/dayformen

):14.8%

20%

repo

rted

useof

atleaston

epsycho

trop

icmed

ication.

7.4%

repo

rted

concurrent

useof

alcoho

land

psycho

trop

icmed

ications

&2.4%

repo

rted

concurrent

heavyalcoho

lcon

sumption

andpsycho

trop

icmed

ication

use

Overallpsycho

trop

icmed

icationusersconcurrent

useof

alcoho

l:37.5%

Overalldrinkers:16%

used

psycho

trop

icmed

ication

Ilomaki

2008

[25]

76.7%

drankalcoho

linprevious

year

(87.5%

ofmen

&68.9%

ofwom

en)

Heavy

drinking

(>14

units/w

eek

men

&>7un

its/w

eekwom

en):

12.6%

(16.6%

ofmen

&7.9%

ofwom

en)

11.5%

repo

rted

useof

atleaston

epsycho

trop

icmed

ication

Overallpsycho

trop

icmed

icationusersconcurrent

useof

alcoho

l:38.9%

ofmaleusers&14.7%

offemale

userswerefre

quen

tdrinkers,

25.9%

ofmaleusers&8.3%

offemaleuserswerehe

avy

drinkers

Ilomaki

2013

[26]

33.7%

drankalcoho

ldailyin

the

past12

mon

ths

Heavy

drinking

(>2drinks/day):

19.2%

Prob

lem

drinkers(CAGE):11%

13.6%

repo

rted

useof

atleaston

epsycho

trop

icmed

ication(8%

repo

rted

antid

epressantuse,

5.7%

sedativeor

anxiolyticsuse&1.6%

both

drug

classes)

Overallantid

epressantusers:

27.1%

consum

edalcoho

ldaily

&15%

heavydrinkers

Sedativeor

hypn

oticusers:

42.7%

consum

edalcoho

ldaily

&26%

heavydrinkers

John

2007

[28]

Prevalen

ceno

trepo

rted

Riskydrinking

(>20

g/daywom

en&>30

g/daymen

):15.1%

ofmen

&3.2%

ofwom

en(exclude

susers

Men

:3.8%

used

sedative,hypn

otic,anxiolytic;1%

used

opioids

Wom

en:6.8%

used

sedative,hypn

otic,anxiolytic;

1.6%

used

opioids

Men

:Riskdrinkers(non

-smoker)

useof

sedative,hypn

oticor

anxiolytic:4.3%

of60–79year

olds

Holton et al. BMC Geriatrics (2017) 17:148 Page 12 of 16

Table

3Summaryof

results;p

revalenceof

alcoho

lcon

sumption,

alcoho

linteractivemed

icationuseandconcurrent

useam

ongolde

radults(Con

tinued)

ofsedative-hypn

otic-anxiolytics

andop

ioids)

Men

:Riskdrinkers(smokers)useof

sedative,hypn

oticor

anxiolytic:

0%of

60–79year

olds

Wom

en:Riskdrinker(non

-smoker)

useof

sedative,hypn

oticor

anxiolytic:13%

of60–79year

olds

Wom

en:Riskdrinkers(smokers)

useof

sedative,hypn

oticor

anxiolytic:0%

of60–79year

olds

Lagn

aoui

2001

[36]

56.3%

drankwine(77.2%

ofmen

&41.1%

ofwom

en)

Heavy

drinking

>0.25

lwine/day:

15.3%

(32.3%

ofmen

&3.1%

ofwom

en)

32%

used

benzod

iazepine

s15.7%

repo

rted

concurrent

useof

alcoho

l(wine)

and

benzod

iazepine

s&2.9%

repo

rted

concurrent

heavy

alcoho

lcon

sumption(wine)

andbe

nzod

iazepine

use

Overallbe

nzod

iazepine

users:49%

consum

edwine

&9.2%

heavydrinkers

Overallwinedrinkers:28.1%

used

benzod

iazepine

s&he

avywine

drinkers:5.3%

used

benzod

iazepine

s

Sheahan

1995

[32]

38%

drankalcoho

linthepast

12mon

ths

28%

repo

rted

useof

atleast

onepsycho

trop

icmed

ication

2%repo

rted

concurrent

useof

alcoho

land

psycho

trop

icmed

ication

Veldhu

izen

2009

[34]

Not

repo

rted

Not

repo

rted

Overallbe

nzod

iazepine

users:33.9%

consum

edalcoho

l&5.1%

werehe

avy

drinkers

Holton et al. BMC Geriatrics (2017) 17:148 Page 13 of 16

likely among men, white respondents, wealthier re-spondents, and those with higher education levels andgreater comorbidities [31]. The prevalence of concur-rent use was lower in studies which focused on psy-chotropic medications, ranging between 2 and 15.7%[14, 32, 36]. Only one study was identified as havinga low risk of misclassifying concurrent use in olderadults [14], with an estimated prevalence of 7.4%among the total study sample and 2.4% for concur-rent heavy alcohol consumption and psychotropicmedications. Older age (70–79 years), residing inrural or small town areas, living alone, higher socialstatus, polypharmacy and a poor social network wereindependently associated with concurrent use of alco-hol and psychotropic medications [14]. While ahigher proportion (15.7%) of older adults in a Frenchstudy [36] were identified as concurrent users of wineand benzodiazepines, the methods of ascertaining ex-posure to benzodiazepines was unclear, as was the as-sessment of concurrent use. Similarly, the lowerestimate of 2% for concurrent use of alcohol and psy-chotropic medications in Sheahan et al. [32] is likelyan artefact of their sample including patients in con-gregate care facilities.Five studies assessed the prevalence of alcohol con-

sumption among users of a broad range of alcohol inter-active medications (Table 3) [19–21, 27, 30]. Breslow[20] and Cousins [21] were identified as having a lowrisk of misclassifying concurrent use, and reportedprevalence estimates of 45 and 60% respectively. Cousinset al. [21] found that older adults using AI medicationswere significantly less likely to report alcohol consump-tion compared to those unexposed to AI medications.Younger age (60–64 years), men, urban dwelling, higherlevels of education and a history of smoking were inde-pendently associated with concurrent use of alcohol andAI medications [21]. A further six studies focused onpsychotropic medication users [14, 22, 25, 26, 34, 36],four of which were identified as having a low risk of mis-classification bias [14, 25, 26, 34]. Du et al. estimated theprevalence of alcohol consumption among users of psy-chotropic medications at 37.5% [14]. Similarly, Ilomakiet al. [25] reported that 38.9% of male psychotropicmedication users, and 14.7% of their female users, con-sumed alcohol, with 26% of male and 8.3% of femaleusers identified as heavy drinkers. Male psychotropicmedication users consumed greater quantities of alcohol,and more often, than non-users. This pattern was notobserved among women [25]. Between one-third and ahalf of all sedative or hypnotic users (34–54%) have beenshown to consume alcohol [20, 21, 26, 34] with between5 and 13% drinking heavily [21, 34]. Consistent withthese findings, Ilomaki et al. [15] found that 26% of menaged greater than 70 years who use sedatives or

hypnotics drink heavily; heavy drinking and daily drink-ing was significantly higher among male sedative or hyp-notic users compared to non-users [26]. Concurrent useof alcohol among older adults taking antidepressantsranged between 42 and 53%, [20, 21] with 27% of maleusers aged greater than 70 years reporting concurrentuse [26].

Adverse outcomes associated with concurrent use ofalcohol and alcohol interactive medicationsOnly four studies reported on adverse outcomes. Threestudies reported on falls, all three studies were cross-sectional [27, 32, 35]. A study by Immonen et al. [27] of2100 older adults in Finland found that falls and injurieswhen a person has consumed alcohol in the past12 months were more common among at risk drinkers(>7 drinks/week, or ≥5 drinks drinking days or ≥3 drinksseveral times/week) using AI medications (13.8%) com-pared to AI medication-users who were not consideredas at risk drinkers (4.1%) (p < 0.001) [27]. In contrast tothese findings, Sheahan et al.’s [32] study of older adultsin America, which included patients in congregate carefacilities, found that although the number of psycho-tropic drugs was associated with an increased odds offalling, the concurrent use of alcohol and psychotropicdrugs was not. Similar non-significant associations werereported in Wong et al.’s [35] convenience sample ofolder adults in the US. One study examined the associ-ation between moderate alcohol consumption (≤ 40 g ofwine per day) and adverse drug reactions among olderadults at the point of admission to one of 81 acute carehospitals in Italy [29]. Among 22,778 participants, 3.9%were identified as having one or more adverse drug reac-tions. Moderate alcohol consumption was associatedwith a 24% increase in the odds of having an adversedrug reaction [29].

DiscussionOverall, the results of this review suggest that betweenone-in-five and one-in-three older adults are potentiallysusceptible to alcohol-medication interactions, withmore than half of AI medication users reporting alcoholconsumption [20, 21, 31]. However, these estimates needto be interpreted with caution as studies differed intheir classification of AI medications. In the absenceof an explicit list of alcohol-interactive medications,multiple drug reference sources were used acrossstudies [20, 21, 27, 30, 31]. Use of different medica-tion compendia led to a lack of consistency in the inclu-sion of AI medications across studies and may have led tothe over or under inclusion of medications [31, 37], result-ing in an over-or-under estimate of concurrent use. Therewas however consensus with regards to CNS agents; allstudies that reported on AI medicines classified CNS

Holton et al. BMC Geriatrics (2017) 17:148 Page 14 of 16

agents as alcohol interactive, specifically psychotropicmedications [14, 18–22, 24–32, 34, 36]. Based on thequality assessment of studies, the most reliable esti-mates for the concurrent use of psychotropic medica-tions and alcohol ranged between 7.4% [14] and7.75% [20].Despite the high prevalence of concurrent use among

older adults, no study examined longitudinal associa-tions with adverse outcomes. Three cross-sectional stud-ies reported on falls with mixed findings [27, 32, 35];with one study reporting on an association betweenmoderate alcohol consumption and adverse drug reac-tions among older adults at the point of admission tohospital [29]. An evidence based list of medicationswhich have a significant risk of harm to older patientswhen combined with alcohol would be useful in a clin-ical setting, allowing for the identification of older adultswhose alcohol consumption places them at increasedrisk and who would benefit from a preventative inter-vention. While recent studies have shown that cliniciansrarely undertake screening and brief interventions to re-duce alcohol consumption [38–40], flagging patients atthe point of prescribing an alcohol-interactive medica-tion may facilitate targeted screening and interventionsto reduce harm. Brief alcohol interventions in primarycare are effective in significantly reducing weekly alcoholconsumption [41]. Additionally, educating older adultsin relation to the risks associated with concurrent use ofalcohol and medications has been shown to increaseolder adults’ awareness of potential risks [42]. However,whether this intervention results in a behaviour changefor those at risk is unclear [42].This is the first systematic review to examine the

prevalence of concurrent use of alcohol and alcoholinteractive medications and associated adverse out-comes in older adults. An explicit and robust method-ology was applied to identify, critically appraise andsynthesise the study findings. However, the findings ofthe review need to be interpreted in the context of thestudy limitations. The risk of misclassification bias forboth alcohol and AI medications was high across manystudies, undermining internal validity. Furthermore, nostudy directly measured the concurrent use of alcoholand AI medications, rather all studies inferred concur-rent use. Finally, heterogeneity across studies in relationto classifying medications as alcohol interactive and inthe methods used to quantify alcohol consumption,prevent statistical pooling of data from existing studies.Variation in study setting and age restrictions, alsomake it difficult to compare prevalence of concurrentuse across studies. Furthermore, due to the current gapin the literature, the available evidence for this reviewwas restricted to three continents, Europe, NorthAmerica and Australia.

ConclusionsWhile there appears to be a high propensity for alcohol-medication interactions in community dwelling olderadults, there is a lack of consensus regarding what con-stitutes an alcohol interactive medication. An explicit listof alcohol interactive medications needs to be derived,and validated prospectively to quantify the magnitude ofrisk posed by the concurrent use of alcohol and alcoholinteractive medications for adverse outcomes in olderadults. This will allow for risk stratification of olderadults at the point of prescribing, and prioritise alcoholscreening and brief alcohol interventions in high-riskgroups.

Additional files

Additional file 1: Database search: this document describes theretrieval process of studies for the systematic review. (DOCX 14 kb)

Additional file 2: Search strategy: This file describes the searchstrategies used in Embase, PubMed, Web of Science and Scopus in orderto identify studies for this systematic review. (DOCX 15 kb)

Additional file 3: Newcastle-Ottawa Scale (NOS) adapted for cross-sectional studies. (DOC 35 kb)

Additional file 4: Table S1. Most common alcohol interactive (AI)medicines across included studies. (DOCX 281 kb)

AbbreviationsADRs: Adverse drug reactions; AI: Alcohol interactive; AUD: Alcohol usedisorder; NSAIDs: Nonsteroidal anti-inflammatory drugs; OTC: Over thecounter

AcknowledgementsNot applicable.

FundingThis study was funded by the Royal College of Surgeons in Ireland (RCSI),School of Pharmacy. The funder had no role in the design of the study andcollection, analysis, and interpretation of data or in writing the manuscript.

Availability of data and materialsAll data generated or analysed during this study are reported in themanuscript (and its Additional files 1, 2, 3 and 4).

Authors’ contributionsAH defined the research question and the methodology including thesearch strategy, performed the literature search, and reviewed all includedstudies, extracted data, performed quality assessment and drafting of themanuscript. GC conceived the research study, defined the research questionand methodology and reviewed full text articles, extracted data, and performedquality assessment. GC was also involved in writing of the paper and approvedthe final manuscript. TF defined the research question, and the methodology,reviewed the manuscript and approved this manuscript to be published. PGdefined the research question, and the methodology, reviewed the manuscriptand approved the manuscript to be published. All authors read and approvedthe final manuscript.

Ethics approval and consent to participateNot applicable.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Holton et al. BMC Geriatrics (2017) 17:148 Page 15 of 16

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.

Author details1School of Pharmacy, Royal College of Surgeons in Ireland (RCSI), Dublin 2,Ireland. 2HRB Centre for Primary Care Research, Department of GeneralPractice, Royal College of Surgeons in Ireland (RCSI), Dublin 2, Ireland.

Received: 18 January 2017 Accepted: 4 July 2017

References1. United Nations, Department of Economic and Social Affairs PD. World

Population Prospects: The 2015 Revision, Key Findings and Advance Tables.New York: 2015.

2. Britton A, Ben-Shlomo Y, Benzeval M, Kuh D, Bell S. Life course trajectoriesof alcohol consumption in the United Kingdom using longitudinal datafrom nine cohort studies. BMC Med. 2015;13:47.

3. Meng Y, Holmes J, Hill-McManus D, Brennan A, Meier PS. Trend analysis andmodelling of gender-specific age, period and birth cohort effects onalcohol abstention and consumption level for drinkers in great Britain usingthe general lifestyle survey 1984–2009. Addiction. 2014;109

4. Wadd S, Papadopoulos C. Drinking behaviour and alcohol-related harm amongstolder adults: analysis of existing UK datasets. BMC Res Notes. 2014;7:741.

5. Moore AA, Whiteman EJ, Ward KT. Risks of combined alcohol/medicationuse in older adults. Am J Geriatr Pharmacother. 2007;5(1):64–74.

6. NHS Information Centre: Statistics for Alcohol : England 2010. 2011.7. Office for National Statistics: 2010 Based Subnational Population Projections

for England 2012.8. Hovstadius B, Hovstadius K, Astrand B, Petersson G. Increasing polypharmacy -

an individual-based study of the Swedish population 2005-2008. BMC ClinPharmacol. 2010;10:16.

9. Sumukadas D, McMurdo ME, Mangoni AA, Guthrie B. Temporal trends inanticholinergic medication prescription in older people: repeated cross-sectionalanalysis of population prescribing data. Age Ageing. 2014;43(4):515–21.

10. Richardson K, Moore P, Peklar J, Galvin R, Bennett K, RA. K. Polypharmacy inadults over 50 in Ireland: Opportunities for cost saving and improvedhealthcare. A report from The Irish Longitudinal Study on Ageing (TILDA).TILDA Trinity College Dublin 2012 December 2012. Report No.

11. Moriarty F, Hardy C, Bennett K, Smith SM, Fahey T. Trends and interaction ofpolypharmacy and potentially inappropriate prescribing in primary care over 15years in Ireland: a repeated cross-sectional study. BMJ Open. 2015;5(9):e008656.

12. Weathermon R, Crabb DW. Alcohol and medication interactions. AlcoholRes Health. 1999;23(1):40–54.

13. Adams WL. Interactions between alcohol and other drugs. Int J Addict.1995;30(13–14):1903–23.

14. Du Y, Scheidt-Nave C, Knopf H. Use of psychotropic drugs and alcoholamong non-institutionalised elderly adults in Germany. Pharmacopsychiatry.2008;41(6):242–51.

15. Ilomaki J, Paljarvi T, Korhonen MJ, Enlund H, Alderman CP, Kauhanen J, et al.Prevalence of concomitant use of alcohol and sedative-hypnotic drugs inmiddle and older aged persons: a systematic review. Ann Pharmacother.2013;47(2):257–68.

16. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reportingitems for systematic reviews and meta-analyses: the PRISMA statement. AnnIntern Med. 2009;151(4):264–9. W64

17. Herzog R, Alvarez-Pasquin MJ, Diaz C, Del Barrio JL, Estrada JM, Gil A. Arehealthcare workers’ intentions to vaccinate related to their knowledge,beliefs and attitudes? A systematic review. BMC Public Health. 2013;13:154.

18. Adams WL. Potential for adverse drug-alcohol interactions amongretirement community residents. J Am Geriatr Soc. 1995;43(9):1021–5.

19. Aira M, Hartikainen S, Sulkava R. Community prevalence of alcohol use andconcomitant use of medication - a source of possible risk in the elderlyaged 75 and older? Int J Geriatr Psychiatry. 2005;20(7):680–5.

20. Breslow RA, Dong C, White A. Prevalence of alcohol-interactive prescriptionmedication use among current drinkers: United States, 1999 to 2010. AlcoholClin Exp Res. 2015;39(2):371–9.

21. Cousins G, Galvin R, Flood M, Kennedy M-C, Motterlini N, Henman MC, et al.Potential for alcohol and drug interactions in older adults: evidence fromthe Irish longitudinal study on ageing. BMC Geriatr. 2014;14

22. Del Rio MC, Prada C, Alvarez FJ. Do Spanish patients drink alcohol whileundergoing treatment with benzodiazepines? Alcohol. 2002;26(1):31–4.

23. Del Río MC, Prada C, Alvarez FJ. The use of medication and alcohol amongthe Spanish population. Br J Clin Pharmacol. 1996;41(3):253–5.

24. Forster LE, Pollow R, Stoller EP. Alcohol use and potential risk for alcohol-related adverse drug reactions among community-based elderly. J CommunityHealth. 1993;18(4):225–39.

25. Ilomaki J, Korhonen MJ, Enlund H, Hartzema AG, Kauhanen J. Risk drinkingbehavior among psychotropic drug users in an aging Finnish population:the FinDrink study. Alcohol. 2008;42(4):261–7.

26. Ilomaeki J, Gnjidic D, Hilmer SN, Le Couteur DG, Naganathan V, CummingRG, et al. Psychotropic drug use and alcohol drinking in community-dwelling older Australian men: the CHAMP study. Drug Alcohol Review.2013;32(2):218–22.

27. Immonen S, Pitkälä K. The prevalence of potential alcohol-drug interactionsin older adults. Eur Geriatr Med. 2012;3:S134.

28. John U, Baumeister SE, Volzke H, Meyer C, Ulbricht S, Alte D. Sedative,hypnotic, anxiolytic and opioid medicament use and its co-occurrence withtobacco smoking and alcohol risk drinking in a community sample. BMCPublic Health. 2007;7:337.

29. Onder G, Landi F, Della Vedova C, Atkinson H, Pedone C, Cesari M, et al.Moderate alcohol consumption and adverse drug reactions among olderadults. Pharmacoepidemiol Drug Safety. 2002;11(5):385–92.

30. Pringle KE, Ahern FM, Heller DA, Gold CH, Brown TV. Potential for alcoholand prescription drug interactions in older people. J Am Geriatr Soc. 2005;53(11):1930–6.

31. Qato DM, Manzoor BS, Lee TA. Drug-alcohol interactions in older US adults.J Am Geriatr Soc. 2015;63(11):2324–31.

32. Sheahan SL, Coons SJ, Robbins CA, Martin SS, Hendricks J, Latimer M.Psychoactive medication, alcohol use, and falls among older adults. J BehavMed. 1995;18(2):127–40.

33. Swift W, Stollznow N, Pirotta M. The use of alcohol and medicines amongAustralian adults. Aust N Z J Public Health. 2007;31(6):529–32.

34. Veldhuizen S, Wade TJ, Cairney J. Alcohol consumption among Canadianstaking benzodiazepines and related drugs. Pharmacoepidemiol Drug Saf.2009;18(3):203–10.

35. Wong H, Heuberger R, Logomarsino J, Hewlings S. Associations betweenalcohol use, polypharmacy and falls in older adults. Nurs Older People.2016;28(1):30–6.

36. Lagnaoui R, Moore N, Dartigues JF, Fourrier A, Begaud B. Benzodiazepineuse and wine consumption in the French elderly. Br J Clin Pharmacol. 2001;52(4):455–6.

37. Vitry AI. Comparative assessment of four drug interaction compendia. Br JClin Pharmacol. 2007;63(6):709–14.

38. Brown J, West R, Angus C, Beard E, Brennan A, Drummond C, et al.Comparison of brief interventions in primary care on smoking and excessivealcohol consumption: a population survey in England. Br J Gen Pract. 2016;66(642):e1–9.

39. Hamilton FL, Laverty AA, Gluvajic D, Huckvale K, Car J, Majeed A, et al. Effectof financial incentives on delivery of alcohol screening and brief intervention(ASBI) in primary care: longitudinal study. J Public Health. 2014;36(3):450–9.

40. Deehan A, Templeton L, Taylor C, Drummond C, Strang J. Low detectionrates, negative attitudes and the failure to meet the “health of the nation”alcohol targets: findings from a national survey of GPs in England andWales. Drug Alcohol Rev. 1998;17(3):249–58.

41. Kaner EF, Beyer F, Dickinson HO, Pienaar E, Campbell F, Schlesinger C, et al.Effectiveness of brief alcohol interventions in primary care populations.Cochrane Database Syst Rev. 2007;2:CD004148.

42. Benza AT, Calvert S, McQuown CB. Prevention BINGO: reducing medicationand alcohol use risks for older adults. Aging Ment Health. 2010;14(8):1008–14.

43. National Consumers League (2013). Avoid Food Drug Interactions Food andDrug Administration Publication No. (FDA) CDER 10–1933.

44. Alcoholism NIoAAa (2014). Harmful Interactions: mixing alcohol withmedicines National Institutes of Health, US Department of Heath andHuman Services 2014.

Holton et al. BMC Geriatrics (2017) 17:148 Page 16 of 16