Academic Edi o : Gianca lo Tonolo
Recei ed: 16 Feb ua y 2025
Re ised: 14 Ap il 2025
Accep ed: 21 Ap il 2025
Published: 23 Ap il 2025
Ci a ion: Nascimen o, T.; And ade,
A.; Pin o, E.; Cab i a, C.; Pais, S.;
Pue a, R.d.l. Medica ion Adhe ence
and Glycemic Con ol in Olde Adul s
wi h Type 2 Diabe es: A C oss-
Sec ional S udy in a Communi y
Se ing. Diabe ology 2025,6, 33.
h ps://doi.o g/10.3390/
diabe ology6050033
Copy igh : © 2025 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license
(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
A icle
Medica ion Adhe ence and Glycemic Con ol in Olde Adul s
wi h Type 2 Diabe es: A C oss-Sec ional S udy in a
Communi y Se ing
Tânia Nascimen o 1,2,* , Amanda And ade 1, Ezequiel Pin o 1,2 , Ca a ina Cab i a 1, Sand a Pais 3
and Rocío de la Pue a 4,*
1Escola Supe io de Saúde, Uni e sidade do Alga e (ESSUAlg), Campus de Gambelas, Edi ício 1,
8005-139 Fa o, Po ugal; [email p o ec ed] (A.A.); [email p o ec ed] (E.P.); [email p o ec ed] (C.C.)
2Alga e Biomedical Cen e Resea ch Ins i u e (ABC-RI), Uni e sidade do Alga e, Campus de Gambelas,
Edi ício 2, 8005-139 Fa o, Po ugal
3Comp ehensi e Heal h Resea ch Cen e, Uni e sidade de É o a, Palácio do Vimioso, Gabine e 256, La go
Ma quês de Ma ial a, Apa . 94, 7002-554 É o a, Po ugal; [email p o ec ed]
4Facul y o Pha macy, Pha macology Depa men , Uni e si y o Se ille, S P o esso Ga cía Gonzalez n◦2,
41012 Se ille, Spain
*Co espondence: [email p o ec ed] (T.N.); [email p o ec ed] (R.d.l.P.)
Abs ac : Backg ound/Objec i es: Glycemic con ol is essen ial o p e en ing bo h sho -
and long- e m complica ions o ype 2 diabe es (T2D), equi ing s ic adhe ence o pha -
macological he apy. Medica ion adhe ence di ec ly in luences he apeu ic e ec i eness,
making i s assessmen in clinical p ac ice c ucial. This s udy aimed o e alua e medi-
ca ion adhe ence in elde ly pa ien s wi h T2D and i s associa ion wi h glycemic con ol.
Me hods: A desc ip i e c oss-sec ional s udy was conduc ed in he Alga e, Po ugal,
in ol ing 133 elde ly pa ien s (
≥
60 yea s) wi h T2D. Ca diome abolic pa ame e s and
medica ion adhe ence (global, in en ional, and unin en ional) we e assessed. S a is ical
analyses we e pe o med using IBM SPSS S a is ics 28.0. Resul s: The s udy popula ion
had a mean age o 71.7
±
5.7 yea s, wi h a p edominance o male pa icipan s (57.9%)
and a high p e alence o dyslipidemia and/o hype ension. Ca diome abolic con ol was
gene ally poo , wi h only 26.3% achie ing blood p essu e a ge s (
≤
140/90 mmHg), 8.5%
main aining as ing glycemia wi hin he ecommended ange (70–110 mg/dL), and 13.6%
a aining glyca ed hemoglobin (HbA1c) alues
≤
7%. Despi e his, medica ion adhe ence
was no ably high (97.7%), wi h no signi ican associa ion wi h ca diome abolic con ol
(p> 0.05). Unin en ional non-adhe ence beha io s, such as o ge ulness and inconsis en
medica ion schedules, we e he mos equen ly epo ed. Conclusions: Al hough elde ly
pa ien s wi h T2D demons a ed high medica ion adhe ence a es, hei ca diome abolic
con ol emained subop imal. Unin en ional non-adhe ence beha io s may con ibu e
o poo glycemic con ol. Howe e , medica ion adhe ence alone does no ully explain
hese ou comes, highligh ing he need o assess adhe ence o o he sel -ca e beha io s,
pa icula ly die a y and physical ac i i y pa e ns. Fu u e in e en ions should in eg a e
comp ehensi e li es yle modi ica ions alongside pha macological managemen o enhance
o e all disease con ol.
Keywo ds: ype 2 diabe es; glycemic con ol; medica ion adhe ence; c oss-sec ional s udy
Diabe ology 2025,6, 33 h ps://doi.o g/10.3390/diabe ology6050033
Diabe ology 2025,6, 33 2 o 15
1. In oduc ion
1.1. Backg ound
Diabe es is a ch onic disease ha a ec ed app oxima ely 828 million people in 2022,
acco ding o he NCD Risk Fac o Collabo a ion [
1
]. In 2021, he In e na ional Diabe es
Fede a ion es ima ed a p e alence o 537 million indi iduals aged 20 o 79 yea s, wi h ype
2 diabe es accoun ing o 90% o diagnosed cases [
2
]. These igu es highligh he inc easing
p e alence o he condi ion and i s g owing global impac .
The managemen o diabe es equi es complex pha macological and
non-pha macological in e en ions, which a e o en inadequa ely ollowed. Pha maco-
logical he apy is pa icula ly in ica e due o he high numbe o medica ions p esc ibed
no only o diabe es bu also o i s associa ed como bidi ies. The complexi y o he a-
peu ic egimens poses challenges o bo h pa ien s and heal hca e p o ide s, con ibu ing
o issues, such as non-adhe ence, he apeu ic ailu e, and ad e se d ug eac ions, which
signi ican ly impai he pa ien s’ quali y o li e [
3
,
4
]. A s udy analyzing he complexi y o
diabe es ea men be ween 1995 and 2003 indica ed ha he apeu ic egimens had become
inc easingly in ica e, wi h a g ea e numbe o pa ien s ecei ing mul i-componen ca e [
5
].
Adhe ence o he apy is a key de e minan o disease con ol and p ognosis in
ch onic condi ions. I can be de ined as he ex en o which a pa ien accep s and ol-
lows medical ecommenda ions, encompassing bo h medica ion use and necessa y li es yle
modi ica ions [6].
Adhe ence can be assessed h ough di ec o indi ec me hods, including se um d ug
concen a ion measu emen s, pa ien sel - epo ing, pha macy e ill eco ds, and elec onic
moni o ing. In Po ugal, gi en he cu en d ug dispensing amewo k, he mos easible
and accessible me hod o e alua ing adhe ence elies on indi ec sel - epo ing ools [7].
Non-adhe ence o ea men is a widesp ead issue in clinical p ac ice, pa icula ly
among asymp oma ic pa ien s wi h ch onic diseases, such as diabe es, hype ension, and
hype choles e olemia. I is associa ed wi h an inc eased isk o complica ions, disease
p og ession, hospi aliza ions, p ema u e disabili y, mo ali y, and a subs an ial economic
bu den [
8
]. The phenomenon o non-adhe ence esul s om a complex in e play be ween
he pa ien , heal hca e p o essionals, and he b oade social en i onmen [
9
]. Despi e
i s c i ical ole in disease managemen , adhe ence is seldom sys ema ically measu ed in
clinical p ac ice. Howe e , i s assessmen , alongside s a egies, such as enhanced pa ien
educa ion and imp o ed communica ion ega ding ea men egimens, has been shown o
posi i ely in luence medica ion adhe ence [10–14].
Adhe ence o he apy in diabe es is c ucial o p e en ing mic o ascula and mac o as-
cula complica ions, such as diabe ic e inopa hy and ca dio ascula disease, espec-
i ely [
15
,
16
]. These complica ions con ibu e o inc eased hospi aliza ion and mo ali y
a es, leading o a subs an ial ise in di ec heal hca e cos s, no only o na ional heal h
sys ems bu also o indi idual pa ien s [17]. Acco ding o he In e na ional Diabe es Fed-
e a ion (IDF), global heal hca e expendi u es ela ed o diabe es ha e su ged by 316% o e
he pas 15 yea s, inc easing om USD 232 billion in 2007 o USD 966 billion in 2021 [2].
P e ious s udies ha e epo ed wide a ia ions in he p e alence o ea men adhe -
ence among indi iduals wi h diabe es, anging om 15% o 93.1% [
18
,
19
]. A me a-analysis
es ima ed an a e age adhe ence a e o o al an idiabe ic medica ions o 67.9%, whe eas
ea men pe sis ence—de ined as he p opo ion o pa ien s emaining on he apy h ough-
ou he s udy pe iod o he du a ion o con inuous medica ion use wi hou p olonged
in e up ion—was lowe , a 56.2% [
20
]. Adhe ence and pe sis ence end o be highe o
o al an idiabe ic agen s compa ed o injec able he apies. The p esence and se e i y o
ad e se d ug eac ions ha e been linked o lowe adhe ence and pe sis ence. Fu he mo e,
Diabe ology 2025,6, 33 3 o 15
sociodemog aphic ac o s, como bidi ies, and glycemic con ol appea o in luence bo h
adhe ence o he apy and ea men pe sis ence in diabe es [21].
1.2. Objec i es
P ima y Objec i e: To assess sel - epo ed medica ion adhe ence in olde adul s
(≥60 yea s) wi h T2D and in es iga e i s associa ion wi h glycemic con ol.
Seconda y Objec i es: To de e mine he p e alence o polypha macy, desc ibe unin-
en ional and in en ional non-adhe ence beha io s, and explo e po en ial implica ions o
clinical p ac ice in e ms o li es yle- ocused in e en ions.
2. Ma e ials and Me hods
This s udy ollowed he STROBE (S eng hening he Repo ing o Obse a ional S ud-
ies in Epidemiology) guidelines o c oss-sec ional esea ch [22] (Supplemen a y File S1).
2.1. S udy Design and Pa icipan s
We conduc ed a c oss-sec ional, desc ip i e s udy, in a non- andom sample o commu-
ni y li ing, au onomous, olde adul s p e iously diagnosed wi h T2D and esiding in he
Alga e egion, Po ugal. This s udy is a seconda y analysis o da a om an obse a ional
s udy i led Sa copenia Sc eening and Heal h-Rela ed Issues in he Region o Alga e
(unpublished), as ou sample included only he olde adul s wi h T2D ha we e pa o
main s udy.
Sample size calcula ion o he main obse a ional s udy, acco ding o me hods p o-
posed by Hulley e al. [
23
], conside ing a 95% con idence le el, a 5% ma gin o e o ,
and an expec ed p e alence o sa copenia o 11.2%, yielded a minimum sample size o
153 pa icipan s. Accoun ing o a 10% a e o d op-ou s, he minimum sample size was
se a 170 indi iduals. The sa copenia p e alence was ex ac ed om he Po uguese Na-
ional Heal h and Physical Ac i i y Su ey [
24
], which included a na ionally ep esen a i e
sample o Po uguese adul s.
Pa icipan s we e ec ui ed be ween Janua y and Decembe h ough in i a ions and
in o ma ional ma e ials dissemina ed ia o mal and in o mal ins i u ions and commu-
ni y g oups o olde indi iduals, such as pa ish councils, municipal councils, and o he
o ganiza ions p o iding se ices o indi iduals aged 60 yea s o olde . Da a collec ion
was conduc ed h ough indi idual, ace- o- ace in e iews wi h all hose who ag eed
o pa icipa e.
2.2. Va iables and Da a Collec ion Tools
Pa icipan s p o ided in o ma ion on gende , age, educa ional le el, como bidi ies,
and cu en medica ion use. Addi ionally, hey comple ed a s uc u ed ques ionnai e
inco po a ing se e al alida ed ins umen s, including he Mini Nu i ional Assessmen
o e alua e nu i ional s a us, he WOMAC Index o assess sel -pe cei ed pain, s i ness,
and unc ional impai men due o os eoa h i is, and he T ea men Adhe ence Measu e
o assess medica ion adhe ence. Following he in e iew, a ained echnician collec ed a
blood sample o labo a o y analysis o as ing blood glucose and glyca ed hemoglobin
(HbA1c). Blood p essu e (BP) was measu ed using a calib a ed sphygmomanome e . T2D
was conside ed adequa ely con olled when BP alues we e below 140/90 mmHg, HbA1c
was below 7%, and as ing blood glucose anged be ween 70 and 110 mg/dL [25,26].
Pa icipan s we e asked o iden i y all p esc ibed and non-p esc ibed d ugs hey we e
aking a he ime and we e no explici ly ins uc ed o dis inguish diabe es medica ions
om o he medica ions o ch onic condi ions. Pha macological ea men was assessed
and classi ied acco ding o he Ana omical The apeu ic Chemical (ATC) Classi ica ion
Sys em, and he o al numbe o medica ions pe pa ien was eco ded. Polypha macy
Diabe ology 2025,6, 33 4 o 15
was de ined as he use o i e o mo e medica ions, while excessi e polypha macy was
conside ed o pa ien s aking en o mo e d ugs [27].
Among he a ailable ins umen s o assessing medica ion adhe ence, he Mo isky
Medica ion Adhe ence Scale (MMAS-4 and MMAS-8) is one o he mos widely used [
28
,
29
].
In Po ugal, he T ea men Adhe ence Measu e (TAM), based on he MMAS me hodol-
ogy [
30
], is equen ly applied and has been alida ed o pa ien s wi h ch onic diseases.
In con as , he MMAS-8 has only been alida ed in a hype ensi e popula ion [
31
]. Med-
ica ion adhe ence in his s udy was e alua ed using he TAM, which has demons a ed
supe io in e nal consis ency compa ed o he Eu opean Po uguese-adap ed e sion o he
MMAS-8 (C onbach’s alpha: 0.74 s. 0.60) [
30
,
31
]. Mo eo e , he TAM exhibi s g ea e speci-
ici y han he o iginal MMAS-8 (73% s. 53%). Howe e , he sensi i i y and speci ici y o
he adap ed MMAS-8 o Eu opean Po uguese we e no epo ed by i s au ho s, p ecluding
a di ec compa ison o hese indica o s be ween he TAM and he MMAS-8 [28–30].
The TAM di e s p ima ily by including se en ques ions, wi h esponses ollowing
a Like scale o ma , which enhances bo h he sensi i i y and speci ici y o he ins u-
men [
30
]. The apeu ic adhe ence was conside ed adequa e o pa icipan s sco ing abo e
70%, ollowing he me hod used in i s alida ion s udy o a Po uguese se ing [
30
]. In
addi ion o o e all adhe ence (ques ions 1 o 7), sco es o in en ional (ques ions 3, 4, and 5)
and unin en ional (ques ions 1, 2, and 6) non-adhe ence we e also calcula ed. Ques ion 7,
“Ha e you e e s opped aking you medica ion o you illness o a eason o he han
you doc o ’s ecommenda ion?”, was excluded om he classi ica ion o in en ional o
unin en ional non-adhe ence, as i s wo ding was conside ed oo ambiguous and could
pe ain o bo h ca ego ies.
2.3. S a is ical Analysis
S a is ical analysis was conduc ed using IBM SPSS S a is ics 28.0. Da a we e desc ibed
using ela i e and absolu e equencies, means, medians, s anda d de ia ions, and in-
e qua ile anges, as app op ia e. Gi en ha none o he a iables unde s udy ollowed
a no mal dis ibu ion, as de e mined by he Kolmogo o –Smi no es , non-pa ame ic
s a is ical es s we e applied. G oup compa isons we e pe o med using he K uskal–Wallis
es , chi-squa e es , and Mann–Whi ney U es . Spea man’s co ela ion coe icien and
Wilcoxon’s signed- ank es we e used o analyze ela ionships be ween a iables. When
he chi-squa e es esul s we e no alid due o low expec ed cell coun s, Fishe ’s exac es
was applied as an al e na i e.
The e we e no missing da a; he e o e, no s a is ical p ocedu es we e necessa y o
add ess hem.
Co ela ion s eng h was classi ied as weak o coe icien alues below 0.1, mode a e
o alues be ween 0.1 and 0.2, and s ong o alues abo e 0.3 [32].
3. Resul s
This s udy included 133 indi iduals diagnosed wi h ype 2 diabe es, wi h a mean age
o 71.7
±
5.72 yea s. The majo i y we e male (57.9%; n= 77), and 68.5% (n= 91) had up o
ou yea s o o mal educa ion (p ima y educa ion). Mos pa icipan s had a diagnosis o
dyslipidemia and/o hype ension. Blood p essu e con ol (
≤
140/90 mmHg) was achie ed
by only 26.3% (n= 35) o he sample, while as ing glycemia (70–110 mg/dL) was wi hin
he a ge ange in 8.5% (n= 10), and glyca ed hemoglobin (HbA1c) alues
≤
7% we e
obse ed in 13.6% (n= 16) (Table 1).
Diabe ology 2025,6, 33 5 o 15
Table 1. Sociodemog aphic and clinical a iable desc ip ions.
Cha ac e is ic n(%)
Gende
Male 77 (57.9%)
Female 56 (42.1%)
Educa ion le el
0 yea s 5 (3.8%)
1–4 yea s 86 (64.7%)
5–9 yea s 26 (19.5%)
10–12 yea s 9 (6.8%)
>12 yea s 7 (5.3%)
Como bidi ies
Hype ension 113 (85.0%)
Dyslipidemia 73 (54.9%)
Ca diome abolic pa ame e s
BP (mmHg)
≤140/90 (con olled) 35 (26.3%)
>140/90 (non-con olled) 98 (73.7%)
Glycemia (mg/dL)
70–110 (con olled) 10 (8.5%)
<70–>110 (non-con olled) 108 (91.5%)
HbA1c (%)
≤7% (con olled) 16 (13.6%)
>7% (non-con olled) 102 (76.7%)
Abb e ia ions: BP, blood p essu e; HbA1c, glyca ed hemoglobin.
Rega ding pha macological he apy, all pa icipan s used a leas one medica-
ion classi ied unde ATC g oup A (Alimen a y ac and me abolism). Addi ionally,
36.8% (n= 49) used a leas one d ug om ATC g oup B (Blood and blood- o ming o gans),
90.2% (n= 120) we e aking medica ions om ATC g oup C (Ca dio ascula sys em), and
59.4% (n= 79) used a leas one d ug om o he ATC classi ica ions.
The numbe o medica ions pe indi idual anged om 1 o 17, wi h a mean o
5.8
±
2.99 (Table 2). Polypha macy (
≥
5 medica ions) was obse ed in 63% (n= 84) o
he pa icipan s, wi h 51.1% (n= 68) aking be ween i e and nine medica ions and 12%
(n= 16) classi ied as excessi ely polymedica ed (≥10 medica ions).
Table 2. Sociodemog aphic and ca diome abolic a iable desc ip ions.
Cha ac e is ic nMean ±SD Median [Min–Max]
IQR
Age (yea s) 133 71.7 ±5.72 72.0 [60–86]
7
SBP (mmHg) 133 151.7 ±21.98 150.0 [99.0–232.0]
26
DBP (mmHg) 133 79.8 ±11.92 79.0 [50.0–121.0]
16
Glycemia (mg/dL) 118 163.2 ±45.22 157.6 [76.0–341.0]
50
HbA1c (%) 118 8.3 ±1.10 7.95 [6.0–13.0]
1
Numbe o
Medica ions 133 5.8 ±2.99 5.0 [1–17]
3
Abb e ia ions: DBP, dias olic blood p essu e; HbA1c, glyca ed hemoglobin; IQR, in e qua ile ange; SBP, sys olic
blood p essu e; SD, s anda d de ia ion.
Diabe ology 2025,6, 33 6 o 15
All pa icipan s comple ed he TAM ques ionnai e, wi h 97.7% (n= 130) classi ied as
adhe en o he apy (sco e > 4; adhe ence > 70%). Adhe ence a es anged om 61.9% o
100%, wi h a median o 92.86% and a mean o 92.3
±
7.08%. The mean alues o in en ional
and unin en ional adhe ence we e 96.1 ±7.91% and 87.3 ±9.97%, espec i ely.
Despi e he high adhe ence a es, o e 70% o pa icipan s did no achie e he ecom-
mended a ge s o key physiological pa ame e s, including blood p essu e, as ing glucose,
and glyca ed hemoglobin (HbA1c). The mean sys olic and dias olic blood p essu e alues
we e 152.0
±
22.20 mmHg and 78.6
±
11.30 mmHg, wi h medians o 149.0 mmHg and
79.0 mmHg, espec i ely. The mean as ing blood glucose le el was 163.2
±
45.22 mg/dL,
while he mean HbA1c was 8.3 ±1.10% (Table 2).
No s a is ically signi ican di e ences we e obse ed be ween adhe ence and sociode-
mog aphic o clinical a iables. Simila ly, no signi ican associa ions we e ound be ween
adhe ence o he apy and he o al numbe o medica ions consumed o he p esence o
polypha macy (p> 0.05), al hough a mode a e co ela ion was iden i ied be ween age and
he o al numbe o medica ions (p= 0.013; = 0.216).
Al hough blood p essu e and blood glucose alues we e simila ly dis ibu ed be ween
adhe en and non-adhe en indi iduals (p> 0.05), a s a is ically signi ican di e ence was
ound in he dis ibu ion o HbA1c alues (p= 0.010; U= 221.500). Howe e , no signi ican
co ela ion was de ec ed be ween adhe ence pe cen ages and HbA1c alues (p= 0.330;
=−0.090).
Rega ding in en ional and unin en ional non-adhe ence, no s a is ically signi ican
di e ences we e obse ed ac oss he analyzed a iables (p> 0.05). The co ela ion be ween
age and lowe he apy adhe ence sco es was also non-signi ican (p= 0.107; =
−
0.145), as
was he case o in en ional (p= 0.096; =
−
0.145) and unin en ional (p= 0.513; =
−
0.057)
adhe ence sco es.
Only 34.6% o pa ien s epo ed ne e o ge ing o ake hei medica ion, while 27.1%
s a ed ha hey we e ne e ca eless wi h hei dosing schedule. Conce ning beha io s
ela ed o in en ional non-adhe ence, 21.8% epo ed ha ing discon inued hei medica ion
because hey el be e , 11.3% had done so because hey el wo se, and only 0.5% epo ed
aking addi ional doses o hei medica ion due o eeling unwell.
The di e ences in pha macological adhe ence sco es we e no s a is ically signi ican
(p> 0.05) be ween g oups ca ego ized by blood glucose le els. Howe e , in en ional
and unin en ional adhe ence sco es we e lowe in pa ien s wi h uncon olled as ing
glycemia (5.8
±
0.48 and 5.3
±
0.55, espec i ely). Beha io s, such as o ge ing o ake
medica ion o inconsis en adminis a ion schedules, we e mo e equen in his g oup. In
con as , al hough no eaching s a is ical signi icance, adhe ence sco es we e highe among
diabe ic pa ien s wi h uncon olled HbA1c alues (5.8
±
0.45). In his g oup, o ge ulness
and ca elessness ega ding medica ion in ake we e associa ed wi h he lowes adhe ence
sco es (Table 3).
Table 3. Medica ion adhe ence and glycemic con ol.
Ques ion
Glycemia
(Mean ±SD)
p
HbA1c
(Mean ±SD)
p
Con olled
(n= 10)
Non-Con olled
(n= 108)
Con olled
(n= 8)
Non-Con olled
(n= 110)
1. Ha e you e e o go en o ake
you medicine o you illness? 5.2 ±0.92 5.0 ±0.89 0.584 4.8 ±1.17 5.1 ±0.86 0.411
2. Ha e you e e been ca eless wi h
he ime o aking medica ion o
you illness? 5.2 ±0.63 4.9 ±0.92 0.313 4.9 ±1.36 4.9 ±0.87 0.670
Diabe ology 2025,6, 33 7 o 15
Table 3. Con .
Ques ion
Glycemia
(Mean ±SD)
p
HbA1c
(Mean ±SD)
p
Con olled
(n= 10)
Non-Con olled
(n= 108)
Con olled
(n= 8)
Non-Con olled
(n= 110)
3. Ha e you e e s opped aking
you medica ion o you illness
because you el be e ? 5.8 ±0.63 5.6 ±0.93 0.420 5.3 ±1.49 5.6 ±0.86 0.582
4. Ha e you e e s opped aking he
medica ion o you illness, on you
own ini ia i e, a e ha ing
wo se eeling?
5.8 ±0.42 5.8 ±0.58 0.403 5.6 ±0.74 5.8 ±0.55 0.197
5. Ha e you e e aken one o mo e
pills o you illness, on you own
ini ia i e, a e
eel wo se?
5.9 ±0.32 5.9 ±0.34 0.469 6.0 ±0.00 5.9 ±0.35 0.500
6. Ha e you e e s opped he apy
o you illness because you le i
end you
medicines?
5.8 ±0.42 5.9 ±0.33 0.293 5.8 ±0.00 5.9 ±0.30 0.736
7. Ha e you e e s opped aking
medica ion o you illness o any
eason o he han
be he doc o ’s ecommenda ion?
6.0 ±0.00 5.8 ±0.63 0.317 6.0 ±0.00 5.8 ±0.63 0.375
Adhe ence
0.340 0.820
Sco e 5.7 ±0.34 5.6 ±0.39 5.5 ±0.71 5.6 ±0.36
% 94.5 ±5.62 92.7 ±6.53 91.1 ±11.85 93.0 ±5.94
In en ional
0.754 0.766
Sco e 5.8 ±0.36 5.8 ±0.48 5.6 ±0.74 5.8 ±0.45
% 97.2 ±6.00 96.2 ±8.03 93.8 ±12.40 96.5 ±7.48
Non-In en ional
0.534 0.978
Sco e 5.40 ±0.54 5.3 ±0.55 5.1 ±0.98 5.3 ±0.50
% 90.0 ±8.99 87.9 ±9.08 85.4 ±16.25 88.3 ±8.39
Abb e ia ions: SD, s anda d de ia ion.
4. Discussion
4.1. The apy Adhe ence
This s udy assessed medica ion adhe ence in 133 elde ly pa ien s wi h T2D and i s
associa ion wi h ca diome abolic con ol. App oxima ely 63% o he pa icipan s we e
polymedica ed and exhibi ed poo ca diome abolic con ol. Howe e , adhe ence o he apy
was no ably high (97.7%), wi h no s a is ically signi ican di e ences obse ed be ween
indi iduals wi h and wi hou adequa e ca diome abolic con ol. Despi e his, pa ien s wi h
uncon olled glucose le els ended o ha e poo e heal h pa ame e s.
Pa icipan s classi ied as non-adhe en had signi ican ly highe HbA1c alues com-
pa ed o adhe en pa icipan s (p= 0.010; U= 221.500). Howe e , his inding was no
suppo ed by he co ela ion analysis be ween he con inuous adhe ence pe cen age and
HbA1c alues (p= 0.330; =
−
0.090). This appa en disc epancy is likely due o he high
adhe ence a e obse ed and gene ally poo glycemic con ol ac oss he sample, wi h 76.7%
classi ied as uncon olled. Thus, while pa icipan s below he adhe ence h eshold (70%)
showed signi ican ly wo se glycemic con ol, he la ge g oup abo e his h eshold exhibi ed
simila ly ele a ed HbA1c alues, minimizing a iabili y and esul ing in a weak o e all
co ela ion. O e all, he indings sugges ha adhe en and non-adhe en indi iduals
p esen simila clinical ou comes.
Diabe es is a ch onic condi ion in which adhe ence o he apy is a c i ical de e minan
o disease con ol, playing a key ole in p e en ing se e e complica ions. Non-adhe ence o
Diabe ology 2025,6, 33 8 o 15
he apy is no me ely a pa ien - ela ed issue bu a he a mul i ace ed p oblem in luenced
by socioeconomic, medical, psychological, and pha maco he apeu ic ac o s [33–35].
The complexi y o pha maco he apeu ic egimens is a well-documen ed con ibu o
o non-adhe ence. Al hough no signi ican di e ences we e ound in his s udy be ween
adhe ence and polypha macy, ch onic pa ien s—who a e o en olde and polymedica ed—
gene ally expe ience g ea e di icul y in unde s anding hei medica ion and managing
hei ea men independen ly [
36
]. In T2D, bo h polypha macy and complex he apeu ic
egimens ha e been iden i ied as key ac o s in luencing adhe ence [
37
]. Addi ionally,
belie s abou medica ion, pe cei ed ea men e icacy, and ea o hypoglycemic episodes
ha e been epo ed as ba ie s o adhe ence [
17
]. These challenges, when combined
wi h ac o s, such as lack o amily suppo , poo communica ion be ween heal hca e
p o essionals and pa ien s, and insu icien in o ma ion abou pha maco he apy, u he
con ibu e o non-adhe ence [
38
,
39
]. Howe e , some s udies sugges ha g ea e egimen
complexi y and disease se e i y may, in ce ain cases, enhance pa ien engagemen and
mo i a ion, ul ima ely leading o imp o ed adhe ence [40].
The e a e mul iple s a egies o enhance adhe ence o he apy, many o which in ol e
simple ye e ec i e in e en ions. These include imp o ed medical and pha maceu ical
guidance, pa ien educa ion o os e a be e unde s anding o he disease and he bene i s
o ea men , and he use o adhe ence-suppo ing de ices ha help p e en o ge ul-
ness [
33
,
34
]. S udies ha e shown ha such in e en ions signi ican ly imp o e adhe ence
a es and, consequen ly, enhance disease con ol [
41
,
42
]. A sys ema ic e iew and me a-
analysis indica ed ha he complexi y o he he apeu ic egimen is no among he p ima y
ba ie s o non-adhe ence [43].
Conside ing pa ien s’ p e e ences ega ding hei ea men is ano he key ac o
in p omo ing adhe ence [
44
]. In T2D, pa ien -cen e ed pha maceu ical in e en ions—
whe he deli e ed in a communi y o clinical se ing by pha macis s alone o as pa o
mul idisciplina y eams—ha e been shown o be e ec i e in imp o ing sel -ca e, medica-
ion adhe ence, and ca diome abolic con ol, as p e iously demons a ed [
45
]. Tools, such
as he Mo isky Medica ion Adhe ence Scale, ha e been widely used o moni o adhe ence
and p o ide aluable insigh s in o pa ien beha io [46,47].
In he p esen s udy, adhe ence o he apy was assessed using a subjec i e ool ha
elies on pa ien s’ sel - epo ed beha io . While such me hods a e suscep ible o g ea e bias
compa ed o di ec measu es—such as se um d ug concen a ion analysis o biochemical
and physiological pa ame e s— hey o e p ac ical ad an ages [
48
]. Al hough biased, sub-
jec i e adhe ence measu es a e concise, easy o implemen in medical and pha maceu ical
consul a ions, and cos -e ec i e [
49
,
50
]. Mo eo e , hey p o ide immedia e eedback and
acili a e he iden i ica ion o unde lying causes o non-adhe ence, allowing o ailo ed
in e en ions based on indi idual pa ien conce ns and needs [51,52].
Despi e he exis ence o mul iple adhe ence assessmen me hods, no single app oach
is conside ed he gold s anda d. The e o e, i is ecommended ha adhe ence e alua-
ions combine mo e han one me hodology o ensu e a mo e accu a e assessmen [
53
]. A
sys ema ic e iew and me a-analysis e alua ing pa ien - epo ed ou come measu es o
medica ion adhe ence in indi iduals wi h ca dio ascula disease and/o ype 2 diabe es
concluded ha none o he assessed ins umen s, including he TAM, me he c i e ia o
being classi ied as bo h eliable and ecommended o hese popula ions [
54
]. None heless,
ools based on he Mo isky scale emain among he mos widely used in clinical p ac ice
and ha e demons a ed sa is ac o y esul s in moni o ing medica ion adhe ence [47].
In he p esen s udy, adhe ence alues g ea e han 70% we e used as he cu -o poin ,
wi h he adhe ence a e eaching 97.7%, a igu e conside ably highe han hose ypically
epo ed in he li e a u e. I is possible ha he pa icipan s’ low le el o educa ion in lu-
Diabe ology 2025,6, 33 9 o 15
enced hei comp ehension o he TAM ques ions, po en ially leading o an o e es ima ion
o adhe ence due o social desi abili y bias—one o he p ima y limi a ions o indi ec
adhe ence assessmen me hods [
55
]. In addi ion o low educa ional a ainmen , social
desi abili y bias has been iden i ied as a con ibu ing ac o o he o e epo ing o good
adhe ence, which may ha e in luenced he esul s in his s udy [56].
A Po uguese s udy ha also employed he TAM ool in diabe ic pa ien s epo ed an
adhe ence a e o 62.3% [
57
], whe eas a s udy conduc ed in B azil using he same me hod-
ology and s udy popula ion obse ed a highe adhe ence a e (84.4%) [
58
]. Al hough he
TAM is based on he Mo isky ques ionnai e, i was o iginally de eloped in Po uguese,
and i s adhe ence calcula ion me hod di e s om he o iginal ins umen . These disc ep-
ancies p esen a challenge in making di ec compa isons ac oss s udies. Fu he mo e, as
sys ema ic e iews ha e epo ed, adhe ence a es in diabe es a e highly a iable, anging
om 15% o 93% [19].
Al hough no signi ican di e ences we e ound be ween he mean alues o in en ional
and unin en ional non-adhe ence, he lowe sco es in i ems ela ed o unin en ional non-
adhe ence sugges ha ailu es in adhe ence a e o en due o o ge ulness o a lack o
medica ion a ailabili y. A subs an ial p opo ion o pa icipan s epo ed ha ing o go en
o ake hei medica ion (65.4%) o being ca eless wi h hei dosing schedule (72.9%). V ies
e al. [
59
] simila ly ound ha mo e han 80% o non-adhe en pa ien s o go o ake hei
diabe es medica ion. Indeed, o ge ulness is widely ecognized as one o he p ima y
easons o non-adhe ence o diabe es he apy [43].
Unin en ional non-adhe ence is pa icula ly p e alen among he elde ly, no only
due o age- ela ed cogni i e decline bu also because o ac o s, such as belie s abou medi-
ca ion and a diminished pe cep ion o he necessi y o pha macological ea men [
60
,
61
].
Mo eo e , a diagnosis o diabe es i sel has been iden i ied as a p edic o o unin en ional
non-adhe ence beha io s [
61
]. The e o e, in e en ions should no only ocus on add ess-
ing o ge ulness and adhe ence o medica ion schedules bu also a ge cogni i e ac o s,
medica ion- ela ed belie s, and sel -e icacy, as hese a e essen ial in imp o ing long- e m
adhe ence o he apy [62].
Al hough he as majo i y o s udy pa icipan s (97.7%) we e classi ied as adhe en
o medica ion, mo e han 90% p esen ed glyca ed hemoglobin (HbA1c) alues equal o o
abo e 7%. HbA1c is a well-es ablished ma ke o diabe es con ol, e lec ing he a e age
blood glucose le els o e he p eceding h ee mon hs, and i s ele a ed alue sugges s
a po en ial o e es ima ion o adhe ence a es [
63
]. Simila indings we e epo ed in an
I anian s udy, whe e mo e han 78% o pa ien s classi ied as highly adhe en o he apy
exhibi ed poo glycemic con ol [
64
]. Howe e , o he s udies ha e demons a ed a s a-
is ically signi ican associa ion be ween medica ion adhe ence and educ ions in HbA1c
le els [65,66].
The high p opo ion o he apy-adhe en pa ien s wi h inadequa e glycemic con ol
may be a ibu ed o esponse bias, as pa ien s may epo highe adhe ence due o social
desi abili y, a emp ing o demons a e compliance wi h pha maco he apeu ic egimens
and medical ecommenda ions o a oid emba assmen . Mo eo e , glycemic con ol is no
solely dependen on medica ion adhe ence. Adhe ence o a heal hy li es yle, including
egula physical ac i i y and an app op ia e die , plays a c ucial ole in ca diome abolic
egula ion [
67
–
69
]. Physical ac i i y adhe ence is pa icula ly low among elde ly indi idu-
als, wi h mobili y limi a ions and como bid condi ions causing pain being majo ba ie s o
exe cise engagemen [
70
–
72
]. In ype 2 diabe es, physical ac i i y is essen ial o imp o ing
ca diome abolic con ol, much like adhe ence o die a y ecommenda ions.
I should also be no ed ha , as p e iously s a ed, pa icipan s in his s udy we e no
speci ically ins uc ed o ocus solely on hei diabe es medica ions when esponding o he
