RESEARCH ARTICLE
E ec o pa ien and ea men ac o s on
pe sis ence wi h an ihype ensi e ea men :
A popula ion-based s udy
Sa a MaloID
1,2
*, Isabel Aguila -PalacioID
1,2
, C is ina Feja
2,3
, Ma ı
´a Jesu
´s Lallana
2,4
,
Ja ie A mes o
4
, Ma ı
´a Jose
´Rabanaque
1,2
1Depa men o P e en i e Medicine and Public Heal h, Uni e si y o Za agoza, Fundacio
´n Ins i u o de
In es igacio
´n Sani a ia de A ago
´n (IIS A ago
´n), Za agoza, Spain, 2G upo de In es igacio
´n en Se icios
Sani a ios de A ago
´n (GRISSA), Za agoza, Spain, 3Di ec o a e o Public Heal h, Go e nmen o A agon,
Za agoza, Spain, 4A agon Heal hca e Se ice, Za agoza, Spain
*[email p o ec ed]
Abs ac
Pu pose
To analyze pa e ns o an ihype ensi e d ug use among new use s in a Sou he n Eu opean
popula ion, and iden i y pa ien - and ea men - ela ed ac o s ha in luence pe sis ence.
Me hods
This is a e ospec i e obse a ional s udy o new an ihype ensi e d ug use s aged �40
yea s in A ago
´n, Spain. In o ma ion on an ihype ensi e d ugs (2014–2016) p esc ibed and
dispensed a pha macies ia he public heal h sys em we e collec ed om a egional elec-
onic popula ion-based pha macy da abase. Pe sis ence was assessed using he gap
me hod. Kaplan-Meie and Cox eg ession analyses we e conduc ed o analyze pa e ns o
use and ac o s ha in luence pe sis ence.
Resul s
The 25,582 new an ihype ensi e d ug use s in A ago
´n du ing he s udy pe iod we e p e-
sc ibed an ihype ensi e d ugs in mono he apy (73.3%), ixed combina ion (13.9%), ee
combina ion (9.1%), o o he (3.7%). One in i e ecei ed an ihype ensi e d ugs wi hin 15
days o he p esc ip ion da e, bu no a e . Du ing he i s yea o ollow-up, 38.6% o he
s udy popula ion emained pe sis en . The likelihood o ea men discon inua ion was
highe o pa icipan s who we e male, aged �80 yea s, and ecei ed an an ihype ensi e
d ug in mono he apy compa ed wi h ixed combina ion.
Conclusion
O e all pe sis ence wi h an ihype ensi e he apy was poo , and was in luenced by he sex,
age and ype o he apy. Fixed combina ions appea o be a good choice o ini ial he apy,
especially in pa ien s wi h a highe isk o discon inua ion. None heless, ad e se d ug e ec s
and he pa ien ’s p e e ences and clinical p o ile should be aken in o accoun .
PLOS ONE
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 1 / 13
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Ci a ion: Malo S, Aguila -Palacio I, Feja C, Lallana
MJ, A mes o J, Rabanaque MJ (2021) E ec o
pa ien and ea men ac o s on pe sis ence wi h
an ihype ensi e ea men : A popula ion-based
s udy. PLoS ONE 16(1): e0245610. h ps://doi.o g/
10.1371/jou nal.pone.0245610
Edi o : Hans-Pe e B unne -La Rocca, Maas ich
Uni e si y Medical Cen e , NETHERLANDS
Recei ed: Ap il 14, 2020
Accep ed: Janua y 4, 2021
Published: Janua y 15, 2021
Pee Re iew His o y: PLOS ecognizes he
bene i s o anspa ency in he pee e iew
p ocess; he e o e, we enable he publica ion o
all o he con en o pee e iew and au ho
esponses alongside inal, published a icles. The
edi o ial his o y o his a icle is a ailable he e:
h ps://doi.o g/10.1371/jou nal.pone.0245610
Copy igh : ©2021 Malo e al. This is an open
access a icle dis ibu ed unde he e ms o he
C ea i e Commons A ibu ion License, which
pe mi s un es ic ed use, dis ibu ion, and
ep oduc ion in any medium, p o ided he o iginal
au ho and sou ce a e c edi ed.
Da a A ailabili y S a emen : Da a we e p o ided
by he egional Go e nmen o A agon, Spain, so
au ho s do no ha e pe mission o sha e he da a.
The pe mission ob ained om he A agon Heal h
In oduc ion
An ihype ensi e d ugs ha e p o en o signi ican ly educe ca dio ascula e en s by lowe ing
blood p essu e [1]. Howe e , heal h ou comes in pa ien s ea ed wi h hese d ugs do no
always co espond wi h expec a ions [2].
The Eu opean Asce aining Ba ie s o Compliance (ABC) p ojec de eloped a axonomy o
s anda dize bo h he medica ion- aking beha io e minology and he e alua ion o medica ion
adhe ence [3]. Acco ding o his axonomy, adhe ence is he p ocess by which pa ien s ake hei
medica ions as p esc ibed, and has h ee di e en componen s o phases: ini ia ion, implemen a-
ion and discon inua ion. Pe sis ence wi h he apy e e s o he las phase, and i is de ined as he
ime be ween ini ia ion o he apy and he las dose, which immedia ely p ecedes discon inua ion.
Poo ea men pe sis ence is he p ima y cause o loss o e ec i eness o p e en i e ca dio-
ascula medica ion [4,5], and con ibu es o inc eases in heal hca e u iliza ion and o e all
heal hca e cos s [6].
The li e a u e desc ibes an ihype ensi e d ug pe sis ence a es anging om 35% o 92%
a e 1 yea [7]. Howe e , accu a e compa ison o published esul s is hinde ed by di e ences
in pa ien popula ions, ollow-up du a ion, de ini ions o pe sis ence, and da a sou ces.
The use o pha macy eco ds as sou ces o in o ma ion o epidemiological d ug u iliza ion
s udies has ecen ly become widesp ead [8]. These sou ces can be used o in es iga e ela ion-
ships be ween pa e ns o d ug use and popula ion and ea men cha ac e is ics. In he case o
an ihype ensi e d ugs, assessmen o pe sis ence using his kind o da a poses some me hodo-
logical challenges, mainly owing o he a ie y o po en ial d ugs, doses, d ug combina ions,
and indica ions. Mo eo e , mos o he published s udies classi y an ihype ensi e d ug use s
whose egimens a e swi ched o added o du ing he ollow-up as discon inue s, p o iding an
un eal es ima e o pe sis ence.
The analysis o ac o s ha in luence pe sis ence is o pa icula in e es o he manage-
men o ca dio ascula disease, and he e is some con o e sy in he li e a u e ega ding he
in luence o he sex, age, ini ial d ug class and ype o he apy on pe sis ence o an ihype en-
si e d ug ea men [9–12].
In his s udy, we desc ibe pa e ns o an ihype ensi e d ug use among new use s in a Span-
ish popula ion and analyze he in luence o pa ien demog aphic cha ac e is ics and ea -
men - ela ed ac o s on pe sis ence.
Ma e ial and me hods
S udy design and da a sou ce
This e ospec i e obse a ional s udy o an ihype ensi e d ug use s was conduc ed in he
egion o A ago
´n in no heas e n Spain. Da a we e collec ed om an elec onic popula ion-
based pha macy da abase; he In o ma ion Sys em o Medica ion Consump ion in A agon. I
collec s he ollowing in o ma ion o all p esc ip ions dispensed in pha macies in A agon ia
he public Na ional Heal h Se ice: anonymous pa ien code, pa ien sex and bi h da e, dis-
pensing da e, Ana omic The apeu ic Chemical (ATC) code o he p esc ibed d ug, numbe o
De ined Daily Doses (DDD), and numbe o packages dispensed.
This sou ce is comple e, con ains alida ed da a, and has been used in p e ious d ug u iliza-
ion s udies [13].
D ugs we e iden i ied acco ding o he ATC/DDD classi ica ion [14], Wo ld Heal h O ga-
niza ion 2018. We included all he apeu ic subg oups indica ed o a e ial hype ension ea -
men (ATC codes C02 [an ihype ensi es], C03 [diu e ics], C07 [be a-blocking agen s], C08
[calcium channel blocke s], and C09 [agen s ac ing on he enin-angio ensin sys em]).
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 2 / 13
Sciences Ins i u e (IACS) imply he exclusi e use o
he da a by esea che s who au ho ed he p esen
s udy. Thus, his in o ma ion canno be published
o sha ed wi h o he ins i u ions. Da a access
eques s should be add essed o he IACS h ough
h ps://www.iacs.es/.
Funding: This wo k was suppo ed by he G upo
de In es igacio
´n en Se icios Sani a ios de A ago
´n
(GRISSA) [B09-17R] o he IIS A ago
´n, and unded
by he egional Go e nmen o A ago
´n, Spain and
he Eu opean Fund o Regional De elopmen
(FEDER). The unde s had no ole in s udy design,
da a collec ion and analysis, decision o publish, o
p epa a ion o he manusc ip .
Compe ing in e es s: The au ho s ha e decla ed
ha no compe ing in e es s exis .
S udy popula ion
The s udy popula ion consis ed o all indi iduals aged �40 yea s in he egion o A ago
´n who
began an ihype ensi e ea men o he i s ime (new use s) du ing 2015. Indi iduals <40
yea s we e no conside ed because o he low equency o hype ension a his age.
The popula ion in A ago
´n on Janua y 1
s
2015 was abou 1.3 million, o which abou
752,000 people we e aged �40 yea s and 99% we e co e ed by he public heal hca e sys em.
Ou come a iables
New use s we e de ined as hose ha had no been p esc ibed any an ihype ensi e d ug du -
ing he 365 days p eceding he index da e (i.e., he da e o he i s p esc ip ion). Fo each sub-
jec , age on he index da e was calcula ed.
New use s we e s a i ied in o ou coho s based on he ype o he apy ecei ed du ing a
15-day pe iod a e he index da e (index pe iod). The conside a ion o he index pe iod was
because an ihype ensi e he apy p esc ibing is ini ia ed in a g adual manne in ce ain
pa ien s, depending on hei needs and esponse. Taking in o accoun all d ugs p esc ibed du -
ing his i s pe iod may he e o e educe he isk o misclassi ica ion o an ihype ensi e d ug
use s:
a. Mono he apy: indi iduals ea ed wi h a single ac i e p inciple du ing he index pe iod.
b. Fixed combina ion: indi iduals ea ed wi h mul iple ac i e p inciples combined in a single
o mula ion du ing he index pe iod.
c. F ee combina ion: indi iduals ea ed wi h a combina ion o single ac i e p inciples in di -
e en o mula ions du ing he index pe iod.
d. Mul iple mix: indi iduals ha ecei ed a ea men o he han he a o emen ioned he a-
pies (e.g. mono he apy and a ixed combina ion) du ing he index pe iod.
Pe sis ence was de ined as he con inua ion o ea men om he index da e o he da e o
discon inua ion, and was e alua ed using he gap me hod. An ihype ensi e d ug use s we e
conside ed non-pe sis en o discon inue s i , du ing he ollow-up pe iod, he da abase
e ealed a gap mo e han wice he du a ion o he p eceding p esc ip ion du ing which no
medica ion was dispensed. This gap leng h was based on p e ious s udies [15,16]. The numbe
o days o a ailable medica ion was es ima ed based on he numbe o DDD in he mos
ecen ly dispensed p esc ip ion. The e o e, he maximum gap allowed was es ima ed as wice
he numbe o DDD. Subjec s we e censo ed i no new p esc ip ion was dispensed du ing
cou se o he pe mi ed gap, o upon eaching he end o he s udy pe iod. Thus, he yea 2014
co esponded o he “wash-ou ” pe iod, 2015 o he s udy “ ec ui men ” pe iod, and 2016 o
he “ ollow-up” pe iod.
An ihype ensi e d ug use s we e classi ied acco ding o hei pa e n o use as pe sis en ;
discon inue s;discon inue s who es a ed du ing he s udy pe iod; o spo use s. Spo use s
we e hose ha illed p esc ip ions exclusi ely du ing he index pe iod, i.e. wi hin 15 days o
he index da e. Upon iden i ica ion, spo use s we e excluded om subsequen analyses.
In bo h he mono he apy and ixed combina ion coho s, he ini ially p esc ibed he apy
could ha e been changed du ing he ollow-up pe iod. Swi ching was de ined as a swi ch om
a single ac i e p inciple o a di e en ac i e p inciple o om a ixed combina ion o a di e -
en ixed combina ion o a ee combina ion. Add-on was de ined as he addi ion o a sepa a e
d ug o he single ac i e p inciple o he ixed combina ion ini ially p esc ibed. A change om
a single ac i e p inciple o a ixed combina ion in which one o he cons i uen d ugs belonged
o he same he apeu ic subg oup was also conside ed add-on.
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 3 / 13
On he o he hand, he ype o coho in luenced he ype o pe sis ence assessed, i.e., class
pe sis ence (pe sis ence wi h a medica ion om he same an ihype ensi e d ug class) o he -
apy pe sis ence (pe sis ence wi h any an ihype ensi e medica ion). In line wi h p e iously
published me hodologies [8,10], we measu ed class pe sis ence in he mono he apy and ixed
combina ion coho s, and he apy pe sis ence in he ee combina ion coho .
New use s o ATC code C02 d ugs and hose assigned o he mul iple mix coho we e no
included in he pe sis ence no in subsequen analyses because o hei low equency and
he e ogenei y.
S a is ical analysis
The sex and age p o ile o he s udy popula ion and he mos p esc ibed d ugs in each o he
s udy coho s we e desc ibed. Pe sis ence was analyzed using he Kaplan-Meie me hod, and
di e ences be ween use g oups using he log- ank es . Median ime o discon inua ion was
es ima ed and Cox eg ession analyses we e pe o med o iden i y ac o s in luencing he
p obabili y o discon inua ion, conside ing he ollowing independen a iables: sex, age
g oup and ype o he apy.
A sensi i i y analysis was pe o med o accoun o he possible in luence o he selec ed
gap leng h and he numbe o DDD as a measu e o he daily dose on pa icipan classi ica ion.
The p opo ion o pe sis en use s and he median ime o discon inua ion we e ecalcula ed
o each coho by applying ixed gap leng hs o 30, 60 and 90 days, based on p e ious esea ch
[9,16,17].
All analyses we e pe o med using S a a e sion 12.1 (S a aCo p, College S a ion, TX,
USA).
E hical in o ma ion
The esea ch p o ocol o he s udy was app o ed by he A agon Heal h Sciences Ins i u e
(IACS) and he A agon Heal h Se ice, who p o ided us he co esponding pha macy claims
da a. All da a used we e anonymized and he esul s agg ega ed, making pa ien iden i ica ion
impossible. Nei he consen o pa icipa e no addi ional e hics app o al we e he e o e
needed.
Resul s
Cha ac e is ics o new an ihype ensi e d ug use s and p esc ibed
ea men s
In o al, 25,582 new an ihype ensi e d ug use s aged �40 we e iden i ied. The majo i y we e
assigned o he mono he apy coho (73.3%), ollowed by he ixed combina ion (13.9%), he
ee combina ion (9.1%) and he mul iple mix (3.7%) coho s (Table 1). The p opo ion o sub-
jec s ecei ing each ype o he apy di e ed acco ding o sex and age (p<0.001). The equency
o p esc ip ion o an ihype ensi e d ugs in ee combina ions as ini ial ea men inc eased
wi h age, om 7.3% in he 40–59 yea s’ age g oup o 13.5% in hose aged �80 yea s.
The egimens mos commonly p esc ibed in mono he apy o in ixed combina ion du ing
he index pe iod a e shown in Fig 1. The mos widely used egimen was angio ensin-con e -
ing-enzyme inhibi o s in mono he apy (31.3%), ollowed by diu e ics (21.9%), angio ensin
ecep o blocke s (13.2%), and be a blocke s (12.8%). On he con a y, he mos used d ugs in
he ee combina ion coho (Fig 2) we e diu e ics ( ep esen ing 32.7% o he o al numbe o
d ugs p esc ibed o his coho du ing he index pe iod), ollowed by angio ensin-con e ing-
enzyme inhibi o s (23.0%) and be a blocke s (20.0%).
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 4 / 13
Table 1. Dis ibu ion o new an ihype ensi e d ug use s aged �40 yea s in A ago
´n (Spain) by sex, age g oup, and ype o he apy du ing he i s 15 days o
ea men .
Sex Age g oup
Women n (%) Men n (%) 40–59 yea s n (%) 60–79 yea s n (%) �80 yea s n (%) To al n (%)
Mono he apy 8,840 (72.0%) 9,907 (74.5%) 7,659 (75.1%) 7,814 (72.4%) 3,274 (71.2%) 18,747 (73.3%)
Fixed combina ion 1,693 (13.8%) 1,857 (14.0%) 1,435 (14.1%) 1,571 (14.6%) 545 (11.9%) 3,551 (13.9%)
F ee combina ion 1,247 (10.2%) 1,097 (8.3%) 747 (7.3%) 978 (9.1%) 619 (13.5%) 2,344 (9.1%)
Mul iple mix 496 (4.0%) 444 (3.3%) 353 (3.5%) 427 (4.0%) 160 (3.5%) 940 (3.7%)
To al 12,276 (100.0%) 13,305 (100.0%) 10,194 (100.0%) 10,790 (100.0%) 4,598 (100.0%) 25,582 (100.0%)
Abb e ia ions: n, numbe ; %, pe cen age.
h ps://doi.o g/10.1371/jou nal.pone.0245610. 001
Fig 1. F equencies o he an ihype ensi e egimens mos commonly p esc ibed as mono he apy o ixed combina ion o new use s in A ago
´n (Spain). The alues
indica ed ep esen he pe cen age o new an ihype ensi e d ug use s ea ed wi h each an ihype ensi e egimen ei he as mono he apy o ixed combina ion du ing
he i s 15 days o he apy ela i e o he o al numbe o new use s.
h ps://doi.o g/10.1371/jou nal.pone.0245610.g001
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 5 / 13

Pa e n o use acco ding o sex, age g oup and ini ial ype o he apy
Compa ison o pe sis ence a es by sex (Fig 3) and by age g oup (Fig 4) e ealed signi ican di -
e ences (log- ank, p<0.001). Women we e pe sis en o longes compa ed wi h men, while
hose aged �80 discon inued he apy much ea lie han he o he wo age g oups. In pa icu-
la , 25.8% o use s aged �80 emained pe sis en o 1 yea om he index da e, ollowed by
use s aged 60–79 (40.6%) and use s aged 40–59 (42.1%) (Table 2).
The p opo ion o pe sis en use s also a ied depending on he ype o he apy p esc ibed
(Table 2). O e all, 5,771 use s (38.6%) s a ing an ihype ensi e he apy as mono he apy we e
classi ied as pe sis en . In he ixed combina ion coho , he e we e 1,263 (43.3%) pe sis en
use s. The ee combina ion coho had he lowes p opo ion o pe sis en use s (32.3%), as
well as he g ea es p opo ion o discon inue s who subsequen ly es a ed ea men (50.6%).
Analyses by age g oup and ype o he apy con i med ha use s aged �80 showed a lowe p o-
po ion o pe sis en use s independen ly o he ype o he apy (p<0.001). The e we e no sig-
ni ican di e ences in he p opo ion o pe sis en use s by ype o he apy among he 40–59
Fig 2. F equencies o he he apeu ic subg oups mos commonly p esc ibed as ee combina ion o new use s in A ago
´n (Spain). The alues indica ed ep esen he
pe cen age o p esc ip ions o each he apeu ic subg oup ela i e o he o al numbe o p esc ip ions p esc ibed as ee combina ion o new an ihype ensi e d ug use s
du ing he i s 15 days o he apy.
h ps://doi.o g/10.1371/jou nal.pone.0245610.g002
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 6 / 13
and he 60–79 age g oups, excep among he ee combina ion use s (39.2% e sus 32.8%,
p = 0.006).
O he o al new an ihype ensi e d ug use s analyzed, 5,005 (19.6%) we e spo use s. By
ype o he apy, dis ibu ion o spo use s was as ollows: 20.0% in he mono he apy coho ,
17.7% in he ixed combina ion coho and 13.7% in he ee combina ion g oup.
Among he pe sis en use s in he mono he apy coho , 765 (13.3%) swi ched o ano he
d ug du ing he ollow-up pe iod, and 902 (15.6%) we e p esc ibed an add-on medica ion.
Among he pe sis en use s in he ixed combina ion coho , 252 (20.0%) swi ched medica-
ion and 96 (7.6%) we e p esc ibed an add-on medica ion.
Di e ences in he median ime o discon inua ion we e obse ed depending on he ype o
he apy ecei ed: mono he apy coho , 171 days; ixed combina ion coho , 248 days; ee
combina ion coho , 148 days.
Mul i a ia e analysis esul s
Table 3 shows he esul s ob ained in bo h he bi a ia e and he mul i a ia e analyses. Men
we e mo e likely o discon inue an ihype ensi e he apy han women (HR, 1.20; 95%CI,
1.15–1.24). Compa ed wi h he 40–59 yea s’ g oup, he likelihood o discon inuing he apy
was signi ican ly highe o he oldes age g oup (�80 yea s) (adjus ed haza d a io [HR], 1.45;
95% con idence in e al [CI], 1.38–1.53). As ega ds he ype o he apy, indi iduals p esc ibed
Fig 3. An ihype ensi e d ug pe sis ence du ing he i s yea o ea men in use s om A ago
´n (Spain) acco ding o hei sex.
h ps://doi.o g/10.1371/jou nal.pone.0245610.g003
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 7 / 13
a ixed combina ion had in bo h he uni a ia e and he mul i a ia e analyses a lowe likelihood
o discon inua ion han hose ecei ing mono he apy. Con e sely, he associa ion be ween
p esc ibing o a ee combina ion and ea men discon inua ion did no each s a is ical
signi icance.
In he sensi i i y analyses (S1 Table), he p opo ions o new an ihype ensi e use s consid-
e ed ea men pe sis en we e 1,7%, 42.3% and 59.0% applying gaps o 30, 60 and 90 days,
espec i ely. These esul s should be compa ed wi h he 38.6% o pe sis en use s ob ained
when applying a gap o 2xDDD. Median ime o discon inua ion was o 46 and 228 days wi h
gaps o 30 and 60 days, espec i ely. Fo he gap o 90 days and also o he gap o 60 days in
he case o he ee combina ion coho , median ime could no be calcula ed since less han
50% o use s had discon inued 365 days a e ini ializing ea men . The p opo ion o pe sis-
en use s ob ained wi h he gap o 60 days was simila o he ob ained wi h he gap o 2xDDD
o all he ypes o he apy () excep o he ee combina ion coho , whe e 51.4% o use s
we e classi ied as pe sis en , ins ead o 32.3%.
Discussion
The indings p esen ed he e help u he ou unde s anding as o he ac o s ha in luence he
ini ial p esc ibing o an ihype ensi e d ugs and hei subsequen use.
In ou s udy popula ion, less han hal (38.6%) o he new an ihype ensi e d ug use s
emained pe sis en a e 1 yea o ea men . E idence sugges s ha measu emen o
Fig 4. An ihype ensi e d ug pe sis ence du ing he i s yea o ea men in use s om A ago
´n (Spain) acco ding o hei age g oup.
h ps://doi.o g/10.1371/jou nal.pone.0245610.g004
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 8 / 13
pe sis ence is a mo e accu a e means o cha ac e izing pa ien commi men and ea men
con inui y han o he indica o s usually calcula ed o desc ibe medica ion adhe ence [15].
Howe e , compa ison o hese indings wi h hose epo ed o o he popula ions is
Table 2. Classi ica ion o new an ihype ensi e d ug use s aged �40 yea s in A ago
´n (Spain) acco ding o he age g oup, he ini ially p esc ibed ype o he apy and
he pa e n o use o e 1 yea om he index da e.
Type o he apy Discon inue s n (%) Discon inue s who es a n (%) Pe sis en n (%) To al n (%)
Age g oup 40–59 yea s
Mono he apy 1,098 (18.2%) 2,424 (40.1%) 2,527 (41.8%) 6,049 (100.0%)
Fixed combina ion 246 (19.9%) 428 (33.7%) 564 (45.6%) 1,238 (100.0%)
F ee combina ion 84 (12.4%) 327 (48.4%) 265 (39.2%) 676 (100.0%)
To al 1,428 (17.9%) 3,179 (39.9%) 3,356 (42.1%) 7,963 (100.0%)
Age g oup 60–79 yea s
Mono he apy 1,126 (17.6%) 2,644 (41.4%) 2,614 (41.0%) 6,384 (100.0%)
Fixed combina ion 304 (23.8%) 415 (32.5%) 560 (43.8%) 1,279 (100.0%)
F ee combina ion 137 (17.1%) 403 (50.2%) 263 (32.8%) 803 (100.0%)
To al 1,567 (18.5%) 3,462 (40.9%) 3,437 (40.6%) 8,466 (100.0%)
Age g oup �80 yea s
Mono he apy 628 (25.0%) 1,257 (50.0%) 630 (25.0%) 2,515 (100.0%)
Fixed combina ion 122 (30.3%) 141 (35.1%) 139 (34.6%) 402 (100.0%)
F ee combina ion 121 (23.4%) 279 (54.1%) 116 (22.5%) 516 (100.0%)
To al 871 (25.4%) 1,677 (48.8%) 885 (25.8%) 3,433 (100.0%)
To al popula ion
Mono he apy 2,852 (19.1%) 6,325 (42.3%) 5,771 (38.6%) 14,948 (100.0%)
Fixed combina ion 672 (23.0%) 984 (33.7%) 1,263 (43.3%) 2,919 (100.0%)
F ee combina ion 342 (17.4%) 1,009 (50.6%) 644 (32.3%) 1,995 (100.0%)
To al 3,866 (19.5%) 8,318 (41.9%) 7,678 (38.6%) 19,862 (100.0%)
Abb e ia ions: n, numbe .
h ps://doi.o g/10.1371/jou nal.pone.0245610. 002
Table 3. Haza d a ios o discon inua ion o an ihype ensi e he apy in new an ihype ensi e d ug use s aged
�40 yea s in A ago
´n (Spain).
Unadjus ed HR (95%CI) Adjus ed HR (95%CI)��
Sex
Women 1 1
Men 1.23 (1.19–1.28)�1.20 (1.15–1.24)�
Age g oup
40–59 1 1
60–79 1.04 (1.00–1.08) 1.02 (0.98–1.06)
�80 1.51 (1.44–1.59)�1.45 (1.38–1.53)�
Type o he apy
Mono he apy 1 1
Fixed combina ion 0.79 (0.75–0.84)�0.80 (0.76–0.84)�
F ee combina ion 1.01 (0.95–1.08) 0.99 (0.93–1.06)
Abb e ia ions: HR, haza d a io; 95%CI, 95% con idence in e al.
N = 19,543.
�S a is ically signi ican esul .
��Model adjus ed o all o he a iables indica ed in he able.
h ps://doi.o g/10.1371/jou nal.pone.0245610. 003
PLOS ONE
De e minan s o pe sis ence wi h an ihype ensi e d ugs
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0245610 Janua y 15, 2021 9 / 13