In e na ional Jou nal o
En i onmen al Resea ch
and Public Heal h
A icle
Alcohol Consump ion among F eshman College S uden s in
Spain: Indi idual and Pooled Analyses o Th ee
C oss-Sec ional Su eys (2005, 2012 and 2016)
Alicia Bus o Mi amon es 1, Lucía Mou e-Rod íguez 1,2,*, Na meen Mallah 1,2,3, Aina a Díaz-Geada 1,3 ,
Mon se a Co al 2,4 , Fe nando Cada ei a 2,4 and F ancisco Caamaño-Iso na 1,2,3
Ci a ion: Bus o Mi amon es, A.;
Mou e-Rod íguez, L.; Mallah, N.;
Díaz-Geada, A.; Co al, M.;
Cada ei a, F.; Caamaño-Iso na, F.
Alcohol Consump ion among
F eshman College S uden s in Spain:
Indi idual and Pooled Analyses o
Th ee C oss-Sec ional Su eys (2005,
2012 and 2016). In . J. En i on. Res.
Public Heal h 2021,18, 2548. h ps://
doi.o g/10.3390/ije ph18052548
Academic Edi o : Paul Tchounwou
Recei ed: 31 Janua y 2021
Accep ed: 26 Feb ua y 2021
Published: 4 Ma ch 2021
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2021 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/).
1Depa men o Public Heal h, Uni e si y o San iago de Compos ela, 15782 San iago de Compos ela, Spain;
[email p o ec ed] (A.B.M.); [email p o ec ed] (N.M.); [email p o ec ed] (A.D.-G.);
[email p o ec ed] (F.C.-I.)
2Heal h Resea ch Ins i u e o San iago de Compos ela (IDIS), 15706 San iago de Compos ela, Spain;
[email p o ec ed] (M.C.); [email p o ec ed] (F.C.)
3
Epidemiology and Public Heal h Ne wo king Biomedical Resea ch Cen e (CIBERESP), 28029 Mad id, Spain
4Depa men o Clinical Psychology and Psychobiology, Uni e si y o San iago de Compos ela,
15782 San iago de Compos ela, Spain
*Co espondence: lucia.mou e. [email p o ec ed]
Abs ac :
Objec i e: We aimed o e alua e changes in he p e alence o Hea y Episodic D inking
(HED) and Risky Consump ion (RC) in eshman college s uden s be ween 2005, 2012 and 2016;
and o iden i y he explana o y a iables o hese pa e ns o consump ion using indi idual and
pooled analyses. Me hods: A c oss-sec ional s udy in ol ing 5260 s uden s was ca ied ou in Spain
in 2005, 2012 and 2016. HED and RC we e de e mined using he Alcohol Use Diso de s Iden i ica ion
Tes . Ano he ques ionnai e was used o measu e pa en al educa ion le el and alcohol use, alcohol-
ela ed p oblems, age o onse o alcohol use and alcohol- ela ed expec ancies. Adjus ed Odds Ra ios
(ORs) o RC and HED and hei 95% Con idence In e als we e es ima ed using logis ic eg ession.
Resul s: An inc ease in he p e alence a es o HED and RC was obse ed among women du ing he
h ee-s udy pe iods, none heless he e was no s a is ically signi ican di e ence in he p e alence
a es among men. High ma e nal educa ional le el, li ing away om pa en al home, ini ia ing
d inking be o e he age o 15 and ha ing posi i e expec ancies abou d inking a e associa ed wi h
highe p e alence o RC in bo h gende s. High posi i e expec ancies and ea ly onse o alcohol use
a e associa ed wi h highe a es o HED among men and women. S uden s ec ui ed in 2012 and
2016 a e p o ec ed agains RC in compa ison o hose ec ui ed in 2005. Conclusions: The age o
alcohol consump ion onse is he mos in luencing ac o on HED and RC o bo h gende s in he
h ee-s udy pe iods. Alcohol p e en ion campaigns a ge ing you h a ea ly ages can educe isky
d inking beha io s.
Keywo ds: hea y episodic d inking; isky alcohol consump ion; eshmen
1. In oduc ion
Alcohol consump ion is a majo public heal h p oblem ha has eached epidemic
le els and a ec s young people in pa icula [
1
,
2
]. The mos common pa e ns o alcohol
consump ion among young people a e: Hea y Episodic D inking (HED) de ined as a way
o d inking ha causes a ise in he Blood Alcohol Concen a ion (BAC) o a leas 0.08%
du ing a single d inking session (NIAAA, 2004) [
3
], and Risky Alcohol Consump ion (RC)
de ined as d inking a le els ha cause nega i e heal h consequences o he d inke o
anyone else when his pa e n o consump ion pe sis s o e ime [4].
Ea ly adul hood ep esen s a sensi i e pe iod o alcohol and d ug use because impo -
an physical, in ellec ual, and social changes ha shape he ma u i y and he de elopmen
In . J. En i on. Res. Public Heal h 2021,18, 2548. h ps://doi.o g/10.3390/ije ph18052548 h ps://www.mdpi.com/jou nal/ije ph
In . J. En i on. Res. Public Heal h 2021,18, 2548 2 o 13
o an indi idual occu du ing his s age [
5
]. Alcohol misuse is associa ed wi h neu ocog-
ni i e al e a ions in he de eloping b ain, whe e changes in he b ain con inues o ake
place un il la e hi ies [
6
,
7
]. Hea y alcohol consump ion du ing you h is also associa ed
wi h majo social p oblems such as a ic acciden s [
8
,
9
], unsa e sex [
10
], a emp ed sui-
cides [
11
], poo academic achie emen [
12
], and non-medical use o p esc ip ion d ugs
(NMUPD) [13].
Recen s udies e ealed ha he age o en ance o he college is a de e minan ac o
o alcohol- ela ed p oblems because his s age is undamen al o he de elopmen o
he indi idual and in ol es impo an social changes [
14
–
17
]. Fu he mo e, uni e si y
s uden s ep esen a highly ulne able popula ion due o academic s ess, lea ing he
pa en al home, as well as he need o independence and social app o al [14,15].
Ea ly onse o alcohol consump ion [
18
], p io d inking expe iences [
19
], pa en al
educa ional le el [
20
], li ing away om pa en al home [
21
], posi i e alcohol expec an-
cies [
22
], pee d inking [
23
] and p egaming [
24
] (i.e., d inking be o e going ou ) ha e
been ecognized as isk ac o s o he excessi e use o alcohol and he misuse o o he
d ugs among s uden s [
21
–
24
]. Pa en al alcohol consump ion and pa en al pe missibili y
o alcohol use a e also associa ed wi h s uden alcohol consump ion [
25
–
27
]. These ac o s
a e e y impo an because hey do no only a ec he cu en consump ion beha io o
he young eshman s uden s bu also in luence hei beha io in he u u e [
21
,
28
–
30
].
Mo eo e , he ini ial expe iences o college d inking can p edic u u e ajec o ies o alco-
hol consump ion and a ec s uden s’ success in college and subsequen adul oles and
independence [21–30].
Gende di e ences in he equency o d inking pa e ns and he explana o y ole o
o he isk ac o s o d inking among emales and males a e s ill unclea . Social e olu ion
like he inc eased p e alence o women s udying a he uni e si y in oduced changes
in women’s ole and consequen ly a ec ed hei pa e n o alcohol consump ion [
31
].
Mo eo e , he pe cep ion o he isk associa ed wi h d inking also a ies by gende [32].
Se e al s udies, mainly hose ca ied ou in USA, epo ed a educ ion in he equency
o alcohol d inking pa e ns among he new gene a ions o college s uden s [
33
]. To he
bes o ou knowledge, simila s udies measu ing changes in isky alcohol consump ion
be ween gene a ions o college s uden s in Spain a e no a ailable [
34
,
35
]. ESTUDES
(Su ey on d ug use in Seconda y Schools in Spain) [
36
], a s udy ca ied ou in you hs
om 14 o 18, ound a dec ease in he p e alence o alcohol pa e ns om 43.4% o 36% in
2018, bu his inding canno be ex apola ed o college s uden s who a e o a legal d inking
age. Economic s abili y, alcohol egula ions, law implemen a ions, and uni e si y con ex
di e conside ably among coun ies [
37
]. In addi ion, college alcohol consump ion and
i s associa ed ac o s can also a y be ween gene a ions. Acco dingly, he e is a need o
ela ed s udies in di e en sociocul u al con ex s.
Ou esea ch g oup has been moni o ing alcohol consump ion by uni e si y s uden s
in Galicia-Spain since 2005. We e alua ed he pa e ns o consump ion and he associa ed
ac o s and de e mined he social consequences o d inking. So a , we ha e unde aken
h ee coho s udies in eshman s uden s in 2005, 2012 and 2016.
In he p esen s udy, we aimed: (1) o assess changes in he p e alence o HED and
RC in college s uden s be ween 2005, 2012 and 2016, and (2) o iden i y he explana o y
a iables o hese pa e ns o consump ion using indi idual and pooled analyses.
2. Ma e ials and Me hods
2.1. S udy Design, and Popula ion
A c oss-sec ional s udy was ca ied ou in s uden s o he Uni e si y o San iago de
Compos ela, Galicia, Spain. The da abase o he p esen s udy was cons uc ed om he
baseline da a o h ee coho s udies ha we e unde aken in eshmen s uden s o he
same uni e si y in 2005, 2012, and 2016. The da a was collec ed in he i s semes e o
he co esponding academic yea (be ween Sep embe and Feb ua y). Pa icipan s we e
ec ui ed using clus e sampling so ha a leas one o he i s -yea classes was andomly
In . J. En i on. Res. Public Heal h 2021,18, 2548 3 o 13
selec ed om each acul y/school o he uni e si y. In each acul y/school, he numbe
o selec ed classes was p opo ional o ha o he s uden s. All eshman s uden s who
we e p esen in he class on he day o he su ey we e in i ed o pa icipa e in he
s udy (N
2005
= 1382, N
2012
= 1328 and N
2016
= 2550). This s udy was app o ed by he
Bioe hics Commi ee o he Uni e si y o San iago de Compos ela. The ques ionnai e was
anonymous. Pa icipan s o he s udies we e in o med bo h e bally and w i en, as pa
o he ques ionnai e, ha he pa icipa ion is olun a y, and ha hey can d op-ou om
he s udy a any ime.
2.2. Da a Collec ion
Each i s -yea class oom was isi ed by he esea che s and all s uden s a ending
he class we e in i ed o pa icipa e in he s udy. In he h ee occasions, alcohol use was
measu ed using he Galician alida ed e sion o he Alcohol Use Diso de s Iden i ica ion
Tes (AUDIT) [38,39].
In addi ion o he AUDIT, we used ano he ques ionnai e o de e mine ac o s po en-
ially associa ed wi h alcohol use such as pa en al educa ional le el and hei alcohol use,
alcohol- ela ed p oblems, and age o onse o alcohol use.
This ques ionnai e also included an i em ha was speci ically designed o measu e
alcohol- ela ed expec ancies. This ques ion was gene a ed using i ems om ano he ques-
ionnai e ha was p e iously used in a popula ion o young Spanish adul s [
40
]. The esh-
man s uden s o 2005 and 2012 we e asked o ank 14 expec ancies abou he e ec s o
alcohol. Subsequen ly, we compu ed he expec ancies placed by he s uden s in he op
7 posi ions by adding 1 poin o each posi i e expec a ion and es ing ano he poin o
each nega i e one as ollows: i adds un (+1), i helps me o socialize (+1), o eel mo e
elaxed (+1), o o ge abou p oblems (+1), o endu e p oblems (+1), i causes i i abili y
(
−
1), anxie y (
−
1), dep ession (
−
1), con usion (
−
1), sleep- ela ed p oblems (
−
1), ne -
ousness (
−
1), agg ession (
−
1), loss o con ol (
−
1), hea iness/d owsiness (
−
1). As o
pa icipan s in he 2016 s udy, expec ancies we e measu ed using he Alcohol Expec ancy
Ques ionnai e—Adolescen , B ie (AEQ—AB) [
41
]. Mo e de ails abou da a collec ion a e
a ailable in e e ence numbe 21 [21].
2.3. De ini ion o Va iables
2.3.1. Independen Va iables
Age o onse o alcohol consump ion: pa icipan s we e g ouped in o ou ca ego ies
acco ding o he age o onse o alcohol use (a e 16 yea s old, a 16, a 15, and be o e he
age o 15).
Socio-demog aphic a iables: gende , place o esidence (in he pa en al home/away
om he pa en al home), and ma e nal educa ional le el (p ima y school/high school/
uni e si y).
Sco e o alcohol expec ancies: Fo he baseline da a collec ed in 2005 and 2012, he num-
be o posi i e and nega i e expec ancies we e used as desc ibed ea lie o gene a e a sco e
ha anged be ween
−
7 and 5, he maximum numbe s o nega i e and posi i e expec an-
cies, espec i ely. Fo he s udy o 2016, a sco e was gene a ed om he Alcohol Expec ancy
Ques ionnai e—Adolescen , B ie (AEQ—AB) ques ionnai e [
41
]. Fo he analysis, all sco es
we e di ided in o e iles.
2.3.2. Dependen Va iable
Risky consump ion: i was asce ained using a sco e gene a ed om AUDIT [
38
,
39
].
Acco ding o he ecommenda ions o he Galician alida ed e sion o he AUDIT [
39
],
a gende -speci ic cu -o alue was es ablished. This a iable was dicho omous and he
adop ed cu -o alues o he sco e we e: ≥5 o women and ≥6 o men.
Hea y d inking: I was de e mined using ques ion no. 3 o he AUDIT: [
38
,
39
] “How
o en do you ha e six o mo e alcoholic d inks on a single occasion?” This ques ion was
answe ed by selec ing one o he ollowing op ions: Ne e , less han once a mon h, a leas
In . J. En i on. Res. Public Heal h 2021,18, 2548 4 o 13
once a mon h, a leas once a week, daily o almos daily. The answe s “a leas once a
mon h, a leas once a week, and daily o almos daily” we e hen g ouped oge he o
o m “mo e equen ly” ca ego y.
2.4. S a is ical Analysis
The baseline da a o he h ee p e ious coho s udies we e pooled in he same
da abase and analyzed using mul ile el logis ic eg ession models o ob ain adjus ed Odds
Ra ios (ORs) o isky consump ion and hea y episodic d inking. Mul ile el logis ic eg es-
sion models we e chosen o his analysis because hey a e mo e lexible han adi ional
models and hus pe mi s analyzing co ela ed da a. The s udy-pe iod was in oduced in o
he model as a andom a iable.
Maximal models we e gene a ed a i s by including all heo e ical independen
a iables. The a iables we e excluded successi ely om he models i hei elimina ion
did no change he es ima ed OR by mo e han 10%. Da a we e analyzed using Gene alized
Linea Mixed Models in IBM SPSS S a is ics o Windows, Ve sion 20.0. A monk, NY:
IBM Co p.
3. Resul s
The esponse a e in he h ee s udies eached 99.0% o s uden s who we e p esen
in he class on he day o he su ey. The samples collec ed in 2005, 2012 and 2016 a e
desc ibed in Table 1and ca ego ized by sex.
Table 1. Desc ip ion o samples collec ed in 2005, 2012 and 2016.
Pe cen age o Mean (95%CI)
Females Males
2005
n= 992
(99.8%)
2012
n= 836
(97.0%)
2016
n= 1497
(90.1%)
2005
n= 371
(99.9%)
2012
n= 449
(95.1%)
2016
n= 864
(94.85%)
Ma e nal
educa ional le el
P ima y school
41.8 (38.4–45.3) 26.2 (22.5–30.1)
31 (28.2–33.9) 32 [ 26.5–37.8) 23.7 (18.8–29) 23.8 (20.4–27.5)
High school
33.6 (30.2–37.1) 34.9 (31.2–38.8) 26.4 (23.6–29.3) 27.6 (22.1–33.3)
30 (25.1–35.3) 23.5 (20–27.2)
Uni e si y
24.6 (21.2–28.1) 38.9 (35.2–42.7) 42.5 (39.7–45.4) 40.3 (34.8–46.1) 46.3 (41.4–51.6)
52.7 (49.2–56.4)
Residence
In pa en al home
24.7 (22.1–27.5) 20.4 (17.7–23.2) 20.1 (18.1–22.1) 29.7 (25.1–34.5) 24.5 (20.6–28.7)
23.9 (21.1 – 26.8)
away om
pa en al home 75.3 (72.6 – 78)
79.6 (76.9–82.4)
79.9 (77.9–82)
70.3 (65.7–75.1) 75.5 (71.5–79.6)
76.1 (73.3–79)
Age o alcohol onse
A e 16 yea s old 16.5 (13–20.1) 12.8 (9.1–16.7)
14.1 (11.5–16.8) 23.4 (17.8–29.4) 15.2 (10.2–20.5)
15.2 (11.8–18.7)
A 16 yea s old
25.6 (22.2–29.3) 31.5 (27.8–35.4)
61.3 (58.7 –64)
21.6 (15.9–27.5)
26.6 (21.6 – 32) 61.9 (58.5–65.4)
A 15 yea s old
38.9 (35.4–42.5) 38.3 (34.6–42.2)
19.6 (17–22.3)
36.9 (31.2–42.8) 41.3 (36.3–46.6)
16.9 (13.5–20.4)
Be o e age o 15
yea s
19.0 (15.5–22.7) 17.4 (13.6–21.3)
4.9 (2.4–7.6)
18.1 (12.5–24.1) 16.9 (11.9–22.3)
6.0 (2.6–9.5)
AUDIT asco e
(mean)
5.42 (5.16–5.68) 5.19 (4.91–5.47) 5.78 (5.54–6.02) 7.65 (7.07–8.23)
6.78 (6.3–7.26) 6.75 (6.4–7.1)
Ques 3 AUDIT a,b
Ne e
61.2 (58.2–64.3) 55.5 (52.1–59.1) 54.4 (51.9–57.1) 39.1 (33.7–44.5) 38.2 (33.5–43.3)
38.1 (34.6–41.7)
Less han once
a mon h
20.9 (17.8–23.9) 23.3 (19.9–26.8) 23.1 (20.5–25.7) 25.3 (19.9–30.7) 23.6 (18.9–28.7)
24.7 [(21.3–28.4)
Mon hly 9.8 (6.8–12.9)
14.2 (10.8–17.7)
14.5 (12–17.2) 12.7 (7.3–18.1)
20.0 (15.3–25.1)
19.3 (15.8–22.9)
Weekly 8.1 (5–11.1) 6.8 (3.4–10.4) 7.9 (5.4–10.6)
22.9 (17.5–28.3) 18.0 (13.3–23.1)
17.8 (14.3–21.4)
Mo e equen ly 0.1 (0–3.2) 0.2 (0–3.8) 0.1 (0–2.7) 0.1 (0–2.7) 0.2 (0–5.3) 0.1 (0–3.8)
Risky Consump ion
51.5 (48.3–54.8) 49.8 (46.3–53.4) 54.8 (52.2–57.5) 58.0 (52.8–63.2) 53.9 (49.2–58.8)
54.3 (50.9–57.8)
Hea y episodic
d inking (%)
17.9 (15.6–20.3) 21.2 (18.5–24.1) 22.5 (20.4–24.7) 35.6 (30.7–40.7)
38.2 (33.7–43) 37.2 (33.9–40.6)
aAlcohol Use Diso de s Iden i ica ion Tes .; bHow o en do you ha e 6 o mo e alcoholic d inks on a single occasion?
In . J. En i on. Res. Public Heal h 2021,18, 2548 5 o 13
In ela ion o socio-demog aphic a iables, he ma e nal educa ional le el o male
and emale s uden s had imp o ed om 42.5% in 2005 o 52.7% in 2016. The dis ibu ion
o s uden s’ esidence (in o ou side pa en al home) was no modi ied h oughou he
nine-yea s s udy pe iod and nea ly h ee qua e s o s uden s li ed away om pa en s’
home du ing hei i s yea o uni e si y. The age o onse o alcohol consump ion was
delayed a leas one yea , o bo h women and men, om 15 o 16 yea s old.
Nea ly 50% o s uden s om bo h gende s main ained a high- isk use o alcohol.
Bo h pa e ns o consump ion (HED and RC) emained s able in he h ee ime pe iods o
bo h sexes.
The ela ionship be ween he p e alence a es o RC and HED and he es o socio-
demog aphic a iables a e shown in Table 2. Ha ing a mo he wi h uni e si y educa ion,
li ing away om pa en s’ home, s a ing d inking be o e 15 and ha ing posi i e alcohol
expec ancies we e associa ed wi h highe p e alence a es o RC among males and emales
in he h ee s udied s udy pe iods. Howe e , only ha ing high posi i e expec ancies and
s a ing alcohol consump ion a ea ly ages we e ela ed o highe a es o HED among men
and women.
Table 2.
P e alence a es o isky consump ion (RC) and hea y episodic d inking (HED) by ma e nal educa ional le el,
esidence, and age o onse o use o alcohol, and posi i e expec ancies.
Pe cen age o Subjec s (%)
Risky Consump ion Hea y Episodic D inking
Females Males Females Males
2005 2012 2016 2005 2012 2016 2005 2012 2016 2005 2012 2016
Ma e nal
educa ional le el
P ima y school
High school
Uni e si y
47.2
53.5
57.3 *
46.2
54.9
47.5
52.3
57.8
54.7ˆ
47.4
60
65.8 *
50
53.5
56.3
52.3
57.1
53.4ˆ
17.1
18.5
18.7
22.9
20.2
20.6
20.5
22.8
23.3
29.3
38
39
36.6
40.2
37.9
37.4
37.2
35.8
Residence
In pa en al home
Away om he
pa en al home
42.2
54.9 * 47.6
50.5 47.8
56.3 *ˆ 51.4
60.5 43.4
57.1 * 46.3
56.3 * 13.9
19.4 * 18.8
21.8 21.5
22.3 34.9
35.7 35.2
38.7 32
38
Age o onse o
alcohol use
A e 16 yea s old
A 16
A 15
Be o e he age o 15
30.5
57.3
69.2
78.9 *
26.2
47.2
71.2
78.1 *
10.4ˆ
39.8ˆ
65.6
80.6 *
32.8
60.2
73.9
89.3 *
36.8
53
76.6
75.4 *
15.2ˆ
36.4ˆ
66.5
79.3 *
7.9
13.4
28.5
38.7 *
7.8
15.1
32.8
47.9 *
6
10.2
26.9
39.8 *
17.2
31.4
46.4
66.7 *
26.5
32.3
57.9
62.7 *
4.3ˆ
19.4ˆ
43.9ˆ
68.1 *
Posi i e
expec ancies
1º e ile
2º e ile
3º e ile
27.5
55.7
70.8 *
26.9
53.5
63.6 *
42.9ˆ
55
68.1 *
35.4
64.5
76.2 *
31.7
60.4
66.4 *
44.8ˆ
52.3ˆ
68.4 *
6.9
18.9
24.3 *
8.7
17.1
36.7 *
16.7ˆ
22.8
28.3
*ˆ
19.2
39.7
48.8 *
22.3
35.7
52.8 *
28.7
33.6
52.5 *
To al 51.5 49.8 54.8 ˆ 58 53.9 54.3 17.9 21.2 22.5 ˆ 35.6 38.2 37.2
*p< 0.05 among ca ego ies; ˆ p< 0.05 among s udy pe iods.
The associa ion o socio-demog aphic a iables and posi i e expec ancies on RC and
HED o he h ee s udy pe iods a e shown in Tables 3–5.
3.1. Ma e nal Educa ional Le el
The associa ion o ma e nal educa ional le el wi h pa e ns o alcohol consump ion
a ied h oughou he h ee-s udy pe iods. In 2005, ma e nal uni e si y educa ion le el
was associa ed wi h highe isk o RC among male s uden s [OR: 2.31 (95%CI: 1.14–4.75)]
(Table 3). On he con a y, in 2012 and 2016, a lowe ma e nal educa ional le el (High
school) was ound o be associa ed wi h a wo- old inc eased isk o HED among men
[OR: 1.99 (CI: 1.01–3.97)] (Table 4) and wi h highe odds o RC among women [OR:1.44 (CI:
1.05–2.00)] (Table 5).
In . J. En i on. Res. Public Heal h 2021,18, 2548 6 o 13
Table 3.
In luence o explana o y ac o s on Risky consump ion and Hea y episodic d inking among male and emale
s uden s in 2005. Resul s a e p esen ed as adjus ed odds a io (95%CI) a.
Females Males
Risky
Consump ion
Hea y Episodic
D inking
Risky
Consump ion
Hea y Episodic
D inking
Ma e nal educa ional le el
P ima y school 1(-) 1(-) 1(-) 1(-)
High school 1 (0.68–1.48) 0.95 (0.6–1.5) 1.60 (0.76–3.42) 1.09 (0.52–2.30)
Uni e si y 1.38 (0.90–2.12) 1.01 (0.62–1.65) 2.31 (1.14–4.75) 1.64 (0.84–3.27)
Residence
In pa en al home 1(–) 1(–) 1(–) 1(–)
Away om he pa en al home 2.34 (1.59–3.45) 1.93 (1.20–3.22) 1.5 (0.78–2.88) 0.72 (0.39–1.35)
Age o onse o use o alcohol
A e 16 yea s old 1(–) 1(–) 1(–) 1(–)
A 16 2.86 (1.8–4.59) 1.46 (0.73–3.13) 3.19 (1.45–7.21) 1.80 (0.75–4.7)
A 15 5.16 (3.09–8.75) 4.04 (2.06–8.57) 5.98 (2.42–15.62) 3.76 (1.47–10.45)
Be o e he age o 15 8.26 (4.6–15.21) 6.03 (2.99–13.07) 10.97 (4.11–32.18) 9.7 (3.75–27.53)
Posi i e expec ancies
1º e ile 1(–) 1(–) 1(–) 1(–)
2º e ile 3.31 (2.15–5.15) 2.97 (1.64–5.55) 2.11 (1.07–4.19) 2.06 (1.01–4.35)
3º e ile 6.14 (4.10–9.29) 3.47 (2.04–6.17) 5.97 (2.65–14.19) 4.01 (1.85–9.04)
aResul s o mul i a ia e logis ic eg ession adjus ed o all a iables included in he able.
Table 4.
In luence o explana o y ac o s on Risky consump ion and Hea y episodic d inking among male and emale
s uden s in 2012. Resul s a e p esen ed as adjus ed odds a io (95%CI) a.
Females Males
Risky
Consump ion
Hea y Episodic
D inking
Risky
Consump ion
Hea y Episodic
D inking
Ma e nal educa ional le el
P ima y school 1(-) 1(-) 1(-) 1(-)
High school 1.32 (0.84–2.08) 0.66 (0.38–1.14) 1.73 (0.88–3.45) 1.99 (1.01–3.97)
Uni e si y 0.95 (0.61–1.48) 0.79 (0.47–1.34) 1.52 (0.82–2.82) 1.11 (0.6–2.08)
Residence
In pa en al home 1(-) 1(-) 1(-) 1(-)
Away om he pa en al home 1.23 (0.79–1.92) 0.91 (0.54–1.59) 2.6 (1.45–4.71) 1.41 (0.79–2.56)
Age o onse o use o alcohol
A e 16 yea s old 1(-) 1(-) 1(-) 1(-)
A 16 2.59 (1.54–4.47) 2.55 (1.1–7.01) 2.11 (1.06–4.32) 1.33 (0.63–2.91)
A 15 6.92 (3.98–12.36) 6.4 (2.81–17.32) 6.66 (2.99–15.54) 4.17 (1.89–9.65)
Be o e he age o 15 8.69 (4.25–18.61) 13.83 (5.55–39.97) 7.94 (3.08–22.14) 5.66 (2.27–14.92)
Posi i e expec ancies
1º e ile 1(-) 1(-) 1(-) 1(-)
2º e ile 2.57 (1.66–4.01) 1.72 (0.94–3.24) 2.83 (1.5–5.42) 1.5 (0.78–2.97)
3º e ile 3.11 (1.96–4.97) 4.04 (2.26–7.54) 3.68 (1.85–7.49) 3.8 (1.89–7.91)
aResul s o mul i a ia e logis ic eg ession adjus ed o all a iables included in he able.
Table 5.
In luence o explana o y ac o s on Risky consump ion and Hea y episodic d inking among male and emale
s uden s in 2016. Resul s a e p esen ed as adjus ed odds a io (95%CI) a.
Females Males
Risky
Consump ion
Hea y Episodic
D inking
Risky
Consump ion
Hea y Episodic
D inking
Ma e nal educa ional le el
P ima y school 1(-) 1(-) 1(-) 1(-)
High school 1.44 (1.05–2.00) 1.23 (0.86–1.75) 1.3 (0.81–2.10) 1.03 (0.64–1.65)
Uni e si y 1.23 (0.92–1.63) 1.27 (0.93–1.75) 1.14 (0.76–1.72) 1.02 (0.68–1.54)
Residence
In pa en al home 1(-) 1(-) 1(-) 1(-)
Away om he pa en al home 1.67 (1.24–2.26) 1.16 (0.84–1.64) 1.68 (1.14–2.48) 1.36 (0.92–2.04)
In . J. En i on. Res. Public Heal h 2021,18, 2548 7 o 13
Table 5. Con .
Females Males
Risky
Consump ion
Hea y Episodic
D inking
Risky
Consump ion
Hea y Episodic
D inking
Age o onse o use o alcohol
A e 16 yea s old 1(-) 1(-) 1(-) 1(-)
A 16 7.74 (3.26–22.89) 3.21 (0.92–20.34) 2.3 (0.96–6.15) 3.71 (1.01–24.02)
A 15 21.28 (9.21–61.87) 10.57 (3.26–64.86) 8.55 (3.86–21.72) 13.31 (3.97–82.82)
Be o e he age o 15 52.03 (20.87–159.39) 18.34 (5.47–114.2) 14.65 (5.92–40.75) 30.46 (8.53–195.3)
Posi i e expec ancies
1º e ile 1(-) 1(-) 1(-) 1(-)
2º e ile 1.51 (1.13–2.01) 1.36 (0.97–1.9) 1.01 (0.68–1.49) 0.98 (0.66–1.47)
3º e ile 2.40 (1.77–3.27) 1.57 (1.13–2.21) 2.22 (1.43–3.48) 2.27 (1.49–3.50)
aResul s o mul i a ia e logis ic eg ession adjus ed o all a iables included in he able.
3.2. Place o Residence
Li ing away om pa en s’ home du ing he i s yea o uni e si y is associa ed wi h
inc eased isk o RC, hough he associa ion a ied be ween males and emales du ing he
s udy pe iods. In 2005, emale s uden s who we e esiding away om pa en al home had
double he isk o RC [OR: 2.34 (CI: 1.59–3.45)] and HED [OR:1.9 (CI: 1.2–3.2)], whe eas
in 2012, he place o esidence was only associa ed wi h RC among males [OR: 2.6 (CI:
1.45–4.71)]. In 2016, bo h male and emale s uden s li ing ou side pa en s’ home we e a
inc eased isk o RC [OR: 1.7 (CI: 1.1–2.5)] and [OR: 1.7 (CI: 1.2–2.3)], espec i ely.
3.3. Age o Onse o Alcohol D inking Onse
The ea lie he onse o alcohol consump ion, (<15 yea s old), he highe he isk
o RC and HED in bo h gende s. This inding was obse ed in he h ee-s udy pe iods.
The highes ORs we e ob ained o emales ec ui ed in 2016 who s a ed d inking be o e
he age o 15. These s uden s we e a mo e han 50 imes highe isk o ha ing isky
consump ion [OR: 52.03 (CI: 20.87–159.39)]. Male s uden s who s a ed d inking a 15 we e
also a 30 imes highe isk o HED [OR: 30.46 (CI: 8.53–195.3)] in compa ison wi h hose
who s a ed d inking a e he age o 16.
3.4. Alcohol Expec a ions
S uden s ha ing posi i e alcohol d inking expec ancies a e a highe isk o RC and
HED. Globally, he magni ude o he associa ion had been dec easing om 2005 o 2016,
whe e OR es ima es o RC in women and men espec i ely declined om 6.14 and 5.97
in 2005, o 3.11 and 3.68 in 2012, and o 2.4 and 2.22 in 2016. OR o HED also d opped o
women and men om 4.04 and 3.8 in 2012 o 1.57 and 2.27 in 2016, espec i ely.
3.5. Pe iod-Based Analysis o he Pooled Da a o 2005, 2012 and 2016
We examined he in luence o s udy pe iod (yea ) on bo h pa e ns o alcohol con-
sump ion (Table 6). S uden s belonging o s udies ca ied ou in 2012 and 2016 we e a
lowe isk o RC and HED in compa ison o s uden s ec ui ed in 2005. Ea ly onse o
alcohol consump ion and posi i e expec ancies o alcohol we e con i med as s able isk
ac o s o bo h pa e ns o consump ion in men and women. Ha ing a mo he wi h a
seconda y educa ion le el and li ing away om he pa en al home also emained as a isk
ac o o RC among males and emales, al hough he place o esidence did no ha e an
in luence on HED among males.
In . J. En i on. Res. Public Heal h 2021,18, 2548 8 o 13
Table 6.
In luence o explana o y ac o s on Risky consump ion and Hea y episodic d inking among all male and emale
s uden s ec ui ed in 2005, 2012 and 2016. Resul s a e p esen ed as adjus ed odds a io (95%CI) a.
Females Males
Risky
Consump ion
Hea y Episodic
D inking
Risky
Consump ion
Hea y Episodic
D inking
Ma e nal educa ional le el
P ima y school 1(-) 1(-) 1(-) 1(-)
High school 1.28 (1.03–1.58) 1.02 (0.79–1.3) 1.44 (1.02–2.02) 1.21 (0.87–1.70)
Uni e si y 1.15 (0.93–1.41) 1.11 (0.88–1.4) 1.36 (1.01–1.84) 1.11 (0.82–1.50)
Residence
In pa en al home 1(-) 1(-) 1(-) 1(-)
Away om he pa en al home 1.70 (1.38–2.08) 1.29 (1.01–1.65) 1.81 (1.36–2.41) 1.20 (0.90–1.60)
Age o onse o use o alcohol
A e 16 yea s old 1(-) 1(-) 1(-) 1(-)
A 16 3.00 (2.2–4.1) 1.97 (1.18–3.31) 2.43 (1.57–3.78) 1.7 (1.01–2.85)
A 15 7.12 (5.19–9.77) 5.4 (3.26–8.94) 7.05 (4.47–11.11) 4.67 (2.80–7.81)
Be o e he age o 15 12.58 (8.61–18.38) 9 (5.33–15.19) 11.2 (6.57–19.08) 9.89 (5.68–17.23)
Posi i e expec a ions
1º e ile 1(-) 1(-) 1(-) 1(-)
2º e ile 2.05 (1.67–2.53) 1.65 (1.27–2.15) 1.47 (1.09–1.97) 1.22 (0.9–1.66)
3º e ile 3.28 (2.65–4.05) 2.28 (1.77–2.93) 2.9 (2.08–4.04) 2.78 (2.01–3.86)
S udy pe iods
2005 1(-) 1(-) 1(-) 1(-)
2012 0.71 (0.56–0.91) 1.25 (0.94–1.67) 0.71 (0.56–0.91) 1.25 (0.94–1.67)
2016 0.66 (0.53–0.83) 0.97 (0.76–1.25) 0.66 (0.53–0.83) 0.97 (0.76–1.25)
aAdjus ed by he all a iables included in he column and andom e ec was conside ed among coho s.
4. Discussion
The p e alence o HED and RC ollowed a s able end in bo h gende s. The e ec o
explana o y ac o s o alcohol consump ion also p esen ed some a ia ions be ween 2005
and 2016. We ound ha ea ly onse o alcohol consump ion emained o s ongly a ec
alcohol consump ion pa e ns in 2016, while opposi e indings we e obse ed o ha ing
posi i e alcohol expec a ions. Ma e nal educa ional le el and place o esidence in luenced
se e al pa e ns o consump ion among he h ee s udy pe iods. When pa icipan s o
he h ee- ime pe iods we e analyzed oge he , we obse ed a educed isk o RC o
pa icipan s ec ui ed in 2012 and 2016 in compa ison wi h hose o 2005. Li ing away
om home, ea ly alcohol consump ion, ha ing a mo he wi h high school educa ion le el
and posi i e alcohol expec a ions inc eased he isk o HED and RC in all male and emale
s uden s o he h ee s udies. The pa icipa ion a e was a ound 98.0% in he h ee yea s,
ensu ing he ep esen a i eness o he s udied popula ion.
One o he mos impo an di e ences be ween he h ee yea s is ela ed o ma e nal
educa ional le el, ha imp o ed o e ime (42.5% in 2005 e sus 52.7% in 2016 o mo he s
comple ing uni e si y educa ion). This inding is in acco dance wi h he inc eased a io
o women wi h college deg ees in he gene al popula ion [
42
]. The lowe equency o
g adua e women as compa ed o ha o 2016, may be also ela ed o he Spanish economic
c isis in 2008 which made i di icul o you h people wi h low pu chasing powe o access
he uni e si y. In p e ious s udies, we conside ed his a iable as a good indica o o
economic s a us [34].
The age o onse o alcohol consump ion delayed wi h yea s as e lec ed by he indings
ob ained om 2012 and 2016 s udies (Table 1). This end di e s om he one epo ed by
ESTUDES 2020 [36], in which he age o alcohol onse emained s able.
HED and RC eached high p e alence among eshman s uden s. Nea ly hal o you hs
p ac iced RC wi h simila a ios be ween gende s. None heless, hough HED p e alence
was high in bo h gende s, i was highe o male (one- hi d) han o emale s uden s
(one- i h). On he one hand, alcohol consump ion in younge gene a ions ends o be
equal be ween gende s, e lec ing social changes ha eme ged in he las decade [
31
].
On he o he hand, gende di e ences pe sis wi h espec o HED wi h highe p e alence
In . J. En i on. Res. Public Heal h 2021,18, 2548 9 o 13
among men han women. This inding could be explained by adi ions and lowe social
pe missibili y o women o d ink alcohol han men [
43
]. Ano he ac ha could explain
his gende di e ence, is ha HED could be unde es ima ed when measu ed using he
hi d ques ion o AUDIT es because ha ques ion does no allow di e en ia ing be ween
gende [
38
]. Ou indings di e om ha o ESTUDES 2020 which epo ed a dec easing
end o RC and HED pa e ns. These di e gences can be explained by he speci ici y o
ou s udied popula ion (uni e si y s uden s).
Acco ding o he li e a u e, uni e si y s uden s, who a e o legal d inking age, a e g ea e
consume s o alcohol han hei non-college pee s [
44
]. Acco dingly, i could be in e es ing
o de elop p e en ion campaigns o his subg oup o popula ion.
Ou esul s also do no coincide wi h ha o s udies om o he coun ies which e-
po ed a dec easing end o alcohol consump ion pa e ns [
33
,
36
]. In ou s udy, he p e a-
lence a es o RC and HED emained s able o e he yea s among males and emales.
These di e ences could be ela ed o he du a ion o s udies. In ou s udy, we had a nine-
yea pe iod ( om 2005 o 2016), while in o he s udies he majo changes in he ends o
consump ion we e no iced in s udies compa ing da a o 20-yea in e als [
33
,
45
–
47
]. In his
line, Wechsle e al., in a s udy ha in ol ed 119 colleges, ound a negligible inc ease in he
p e alence o hea y d inking pa e n be ween 1993 (43.9%) and 2001 (44.4%) [45].
Conce ning he place o esidence, i seems ha li ing away om pa en al home con-
e s a highe isk o RC, bu no o HED. Also, he in luence o his a iable a ied acco ding
o he ime-pe iod whe e i a ec ed only emales in 2005, only males in 2012, and bo h
gende s in 2016. Li ing ou side pa en s’ home, was ela ed o RC in he Uni ed S a es.
In speci ic, li ing in uni e si y campus inc eases he isk o alcohol
consump ion [47–49].
These beha io s could be pa ly explained by a less pa e nal supe ision and a s onge
pee in luence among s uden s who do no li e in he amily home [
46
–
48
]. S uden s li -
ing ou o pa en al home migh ha e a s able and p obably daily alcohol consump ion
pa e n compa ed o hose li ing wi h hei pa en s. A ac ha could explain he lack o
associa ion be ween li ing ou side pa en s’ home and HED is he cul u al accep ance o
alcohol consump ion in Spain [
50
], e en unde g ea e pa e nal supe ision, he e ec o
his a iable is likely o be lowe han ha in o he coun ies. The compa ison be ween he
h ee- ime pe iods in ou s udy showed ha he place o esidence in luences bo h pa e ns
o alcohol consump ion in bo h gende s, e en hough he e ec is highe o RC.
I is no iceable he delay in he age o onse o alcohol use om he age o 15 o
16 in 2016. Conside ing ha an ea lie age o onse has been ound as a isk ac o o
RC and HED pa e ns [
51
,
52
], his delay should p o ec s uden s agains isky d inking
beha io s. The e ec o he age o alcohol onse was high in he h ee s udy pe iods, bu he
g ea es e ec was obse ed in 2016. The ac ha only 24.5% o women and 22.9% o men
who pa icipa ed in 2016 s udy had s a ed d inking a he age o 15 o be o e (hal he
pe cen age ound in 2005 and 2012), demons a es ha he age o onse depic s a subg oup
o ex eme pe sonali ies o beha io s.
Fo he h ee yea s, he age o onse was main ained as he mos in luencing a iable
on HED and RC among males and emales. This inding ein o ces ha alcohol p e en ion
campaigns a ea ly ages can imp o e isky beha io s h oughou you h. Ne e heless,
ou indings mus be in e p e ed wi h cau ion because o he wide ange o he co espond-
ing con idence in e als.
In line wi h o he s udies and as expec ed, ou indings e eal ha posi i e alcohol
expec ancies ep esen a isk ac o o alcohol consump ion by males and emales [
22
,
34
].
Ne e heless, he magni ude o he e ec o ha ing posi i e expec ancies on RC and HED
pa e ns o consump ion is dec easing wi h ime.
The lack o signi ican di e ences in he endency o posi i e alcohol expec a ions
be ween men and women e lec s a educ ion in gende inequali y [31,53,54].
RC and HED ends emained s able o e ime. Al hough, when pa icipan s o he
h ee- ime pe iods we e analyzed oge he and a e adjus ing o all con ounding a iables,
RC seems o ollow a downwa d end in bo h gende s as e lec ed by OR es ima es (2005).
