sus ainabili y
A icle
Exe cise Addic ion and Sa is ac ion o Fi ness Cen e Use s as
P ecu so s o he In en ion o Con inuing o Engage in
Physical Ac i i y
An onio Fe nández-Ma ínez 1,* , Víc o Mu illo-Lo en e 2, Albe o Sa mien o 2, Ja ie Ál a ez-Medina 2and
Albe o Nu iala 1


Ci a ion: Fe nández-Ma ínez, A.;
Mu illo-Lo en e, V.; Sa mien o, A.;
Ál a ez-Medina, J.; Nu iala, A.
Exe cise Addic ion and Sa is ac ion o
Fi ness Cen e Use s as P ecu so s o
he In en ion o Con inuing o Engage
in Physical Ac i i y. Sus ainabili y
2021,13, 129. h ps://dx.doi.o g/
10.3390/su13010129
Recei ed: 5 Decembe 2020
Accep ed: 21 Decembe 2020
Published: 25 Decembe 2020
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 ilia ions.
Copy igh : © 2020 by he au ho s. Li-
censeeMDPI, Basel, Swi ze land. This
a icleisanopenaccessa icledis ibu ed
unde he e ms and condi ions o he
C ea i eCommonsA ibu ion(CCBY)
license(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
1Depa men o Spo s and Compu e Science, Pablo de Ola ide Uni e si y, 41013 Se ille, Spain;
[email p o ec ed]
2Facul y o Heal h and Spo Sciences, Uni e si y o Za agoza, 22003 Huesca, Spain;
mu illo@uniza .es (V.M.-L.); [email p o ec ed] (A.S.); ja ial @uniza .es (J.Á.-M.)
*Co espondence: [email p o ec ed]; Tel.: +34-954-977-602
Abs ac :
Exe cise addic ion occu s when a pe son engages in excessi e physical ac i i y un il hey
lose con ol. Among indi iduals a ending i ness cen e s, he isk o de eloping exe cise addic ion
has been es ima ed o be close o 10%. The objec i e o his s udy was o de e mine whe he exe cise
addic ion may be an an eceden o sa is ac ion wi h he se ice ecei ed and/o o he in en ion o
con inuing o pa ake in leisu e- ime physical ac i i y a i ness cen e s in a di ec o indi ec manne .
A o al o 361 indi iduals (29.05
±
11.40 yea s old) who we e physically ac i e a Spanish i ness
cen e s we e asked o espond o a ques ionnai e on exe cise addic ion, sa is ac ion wi h he spo s
se ice, and hei in en ion o con inue o engage in physical ac i i y. A mul i-g oup analysis was
conduc ed o check o di e ences in ela ionships by sex. The esul s show he ela ionships be ween
addic ion, sa is ac ion, and in en ion o con inue physical ac i i y. The s anda dized alues display
di e ences be ween men and women, wi h women exhibi ing he highes alues o he ela ionships
be ween addic ion and he es o he cons uc s. Sa is ac ion wi h he spo s se ice is a media o in
he ela ionship be ween exe cise addic ion and in en ion o engage in leisu e- ime physical ac i i y.
Keywo ds:
exe cise addic ion; i ness; spo s se ices; sa is ac ion; u u e in en ions; physical ac i i y
1. In oduc ion
Physical exe cise is one o he main ac i i ies ca ied ou in leisu e ime (leisu e- ime
physical ac i i y, LTPA) [
1
]. In Eu ope, 54% o he popula ion epo s playing spo s o
doing some kind o physical ac i i y, wi h 40% doing so egula ly o somewha egula ly [
2
].
The Eu opean Commission s udy inds ha h ee ou o en Eu opean ci izens a e membe s
o an ins i u ion whe e hey exe cise, wi h 11% belonging o a i ness cen e . In he same
ein, a s udy conduc ed in he 28 coun ies o he Eu opean Union, in addi ion o No way,
Russia, Swi ze land, Tu key, and Uk aine, s a es ha 62.2 million people epo being
membe s o a i ness club [3].
Many o he cus ome s a end i ness cen e s looking o imp o e hei heal h [
2
].
Howe e , when exe cise is aken in excess, i can become addic i e and cause physical and
psychological issues [
4
]. The concep o exe cise addic ion e e s o he need o signi ican
amoun s o exe cise o he ex en ha he indi idual loses con ol o e hei ac i i y [
5
].
Signs and symp oms consis en wi h his diso de include: excessi e p eoccupa ion wi h
exe cise; educ ion o cessa ion o o he daily ac i i ies; wi hd awal symp oms in he
absence o exe cise (anxie y, i i abili y, and ouble concen a ing o sleeping); and e en
ea ing diso de s, among o he s [6–8].
I has been es ima ed ha 3% o hose who p ac ice spo s a e a isk o becoming
addic ed o exe cise [
7
,
9
], al hough he a iabili y in he s udies conduc ed so a also shows
Sus ainabili y 2021,13, 129. h ps://dx.doi.o g/10.3390/su13010129 h ps://www.mdpi.com/jou nal/sus ainabili y
Sus ainabili y 2021,13, 129 2 o 12
a g ea s a is ical dispe sion [
4
]. Speci ic s udies in ol ing i ness cen e cus ome s show
highe , albei dispa a e, esul s. In a s udy wi h nea ly 300 subjec s, 42% we e ound o
mee he diagnos ic c i e ia o exe cise dependence [
10
]; in ano he s udy, wi h a sample o
162 Swedish and 269 Po uguese a hle es o bo h sexes, esea che s ound a isk o exe cise
dependence in 9.2% o Swedish and 5.2% o Po uguese a hle es [
11
]; in ano he s udy,
wi h a sample o 531 use s, 8.1% we e ound o be a isk [
12
]. In mo e ecen s udies [
13
], i
was ound ha 7.8% o a sample o 128 indi iduals we e a isk o de eloping dependence.
Finally, in a sys ema ic e iew o he spo s wi h he highes isk o gene a ing exe cise
addic ion [
14
], i was ound ha 8.2% o i ness cen e a endees we e a isk o becoming
addic ed o exe cise.
Wi h espec o he age a iable, mos s udies show an in e se end wi h inc easing
age [
15
,
16
]. The esul s o s udies on i ness use s in pa icula con inue his end, wi h
olde age g oups displaying lowe le els o exe cise dependence [
11
,
12
]. In he o he
land, he esul s ega ding gende a e inconclusi e [
12
,
17
–
19
]. A s udy ocusing speci i-
cally on i ness cen e use s aged be ween 18 and 40 yea s old concluded ha he e we e
no signi ican di e ences in exe cise addic ion le els [
20
]. Al hough ecen s udies sug-
ges ha women may be mo e p one o exe cise addic ion han men, u he esea ch is
needed [
21
,
22
]. Exe cise addic ion has also been linked o psychosocial a iables such
as mo i a ion [
23
]; sel -es eem and na cissism [
24
]; obsessi e passion and dedica ion o
spo s [
25
]; exe cise iden i y and social physique anxie y [
26
] o o ho exia [
27
]. Howe e ,
exe cise addic ion has no been linked o undamen al a iables o physical exe cise, such
as sa is ac ion wi h he spo s se ice ecei ed, he physical ac i i y pe o med [
28
–
30
], o
he in en ion o pa ake in LTPA [31,32], among i ness cen e cus ome s.
Sa is ac ion is he p oduc o a compa ison be ween expec a ions and he se ice
deli e ed [
33
], which may be in luenced by cogni i e [
34
] and/o emo ional aspec s [
35
].
A numbe o s udies ha e analyzed sa is ac ion among i ness cen e use s [
29
,
36
,
37
],
highligh ing i s ele ance in shaping he u u e in en ions o use s. In hese s udies, he
links be ween sa is ac ion and la en a iables such as quali y, pe cei ed alue, and loyal y
a e analyzed. Howe e , sa is ac ion among i ness cen e use s has no been s udied in
e ms o gende . S udies [
38
,
39
] ha e shown ha g ea e sa is ac ion wi h he spo s se ice
and he physical ac i i y pe o med co esponds o a g ea e equency o exe cise, while
dedica ion o spo s is a p edic o o exe cise addic ion [
25
]. Howe e , he e a e no s udies
ela ing his a iable o exe cise addic ion. Based on his e idence, he ollowing hypo hesis
can be o mula ed (H1): Addic ion o physical exe cise among i ness cen e use s is an
an eceden o sa is ac ion wi h he spo s se ice and wi h he physical ac i i y pe o med.
Rega ding he in en ion o engage in physical ac i i y, se e al s udies ha e shown
ha in en ion is one o he main de e minan s when i comes o a emp ing o engage
in physical exe cise in he u u e [
40
]. In en ion can hus p edic ac ual beha io [
41
]. I
has also been es ablished ha he de e minan o olun a y beha io is he in en ion o
pa icipa e in ha beha io [
42
]. I he in en ion o be physically ac i e is ela ed o he
le els o physical ac i i y pe o med [
43
–
45
] and dedica ion o spo s is a p edic o o
exe cise addic ion [
25
], hen he ollowing hypo hesis may be o mula ed (H2): Exe cise
addic ion is a p ecu so o he in en ion o pa ake in LTPA.
The li e a u e on se ices in gene al and spo s se ices in pa icula has highligh ed
he ela ionship be ween sa is ac ion wi h he spo s se ice ecei ed and loyal y, unde -
s ood as he in en ion o epu chase o euse he se ice [
46
,
47
]. Howe e , whe he o no
he e is a ela ionship be ween sa is ac ion wi h he spo s se ice and he spo s ac i i y
pe o med and he in en ion o con inue o pe o m physical ac i i y o e ime has ye o
be s udied. As a esul , he ollowing hypo hesis can be o mula ed (H3): Sa is ac ion wi h
he physical ac i i y pe o med and wi h he spo s se ice can be an an eceden o he
in en ion o pa ake in LTPA.
Sus ainabili y 2021,13, 129 3 o 12
Upon comple ion o he li e a u e e iew, he objec i e o his s udy was o de e mine
whe he exe cise addic ion can be an an eceden o sa is ac ion wi h he se ice ecei ed,
an an eceden o sa is ac ion wi h he physical ac i i y pe o med, and/o an an eceden
o he in en ion o con inue o pe o m LTPA a i ness cen e s ei he di ec ly o indi ec ly
h ough sa is ac ion wi h he se ice ecei ed (Figu e 1).
Sus ainabili y 2021, 13, x FOR PEER REVIEW 3 o 12
be s udied. As a esul , he ollowing hypo hesis can be o mula ed (H3): Sa is ac ion wi h
he physical ac i i y pe o med and wi h he spo s se ice can be an an eceden o he
in en ion o pa ake in LTPA.
Upon comple ion o he li e a u e e iew, he objec i e o his s udy was o de e mine
whe he exe cise addic ion can be an an eceden o sa is ac ion wi h he se ice ecei ed,
an an eceden o sa is ac ion wi h he physical ac i i y pe o med, and/o an an eceden
o he in en ion o con inue o pe o m LTPA a i ness cen e s ei he di ec ly o indi ec ly
h ough sa is ac ion wi h he se ice ecei ed (Figu e 1).
Figu e 1. Model o he ela ionships be ween exe cise addic ion, sa is ac ion wi h he i ness cen-
e , and in en ion o pa ake in leisu e- ime physical ac i i y (LTPA).
2. Ma e ials and Me hods
2.1. Pa icipan s
All use s a se e al i ness cen e s we e in i ed o pa icipa e in he s udy, p o ided
hey had been en olled a hei cen e o mo e han h ee mon hs (N = 4857). In he end,
361 use s comple ed he ques ionnai e success ully, esul ing in a ma gin o e o o 5%
and a 95% con idence le el. Thei ages anged om 18 o 64 yea s old, wi h he mean age
being 29.05 ± 11.40. A o al o 39.9% o he pa icipan s we e women. The majo i y o pa -
icipan s had comple ed seconda y educa ion (50.6%). Thei mean equency o physical
ac i i y was 3.63 ± 1.27 sessions pe week, while he mean ime pe session was 90.49 ±
27.38 min.
2.2. Measu es
Se e al ins umen s we e used o conduc his s udy. The i s one was he Spanish
e sion o he Exe cise Addic ion In en o y (EAI) [48]. This scale is made up o 6 i ems.
Once he da a had been collec ed, eliabili y, as measu ed wi h C onbach’s alpha, was
0.708. Highe sco es indica e g ea e symp oms o exe cise addic ion.
The second ins umen used was he Spo s O ganiza ions Pe cep ion Scale, e sion
2 (EPOD2) [49]. This is a scale o measu ing he sa is ac ion o use s o spo s se ices.
Reliabili y, as measu ed wi h C onbach’s alpha, was 0.943. Finally, he Spanish e sion o
he ques ionnai e In en ion o Pa ake in Leisu e-Time Physical Ac i i y [50] was used.
This is a 3-i em scale o measu ing in en ions o ca ying ou physical ac i i y du ing
leisu e ime. Reliabili y, as measu ed wi h C onbach’s alpha, was 0.969. In bo h cases,
highe sco es indica e g ea e sa is ac ion and g ea e in en ion o engage in LTPA.
Responden s we e ins uc ed o a e hei le el o ag eemen wi h each i em using a
i e-poin Like scale anging om “s ongly disag ee” (1) o “s ongly ag ee” (5). Addi-
ional ques ions we e added o he scales: age, gende , le el o educa ion, numbe o phys-
ical ac i i y sessions pe week, and mean ime pe session (Table 1).
Figu e 1.
Model o he ela ionships be ween exe cise addic ion, sa is ac ion wi h he i ness cen e ,
and in en ion o pa ake in leisu e- ime physical ac i i y (LTPA).
2. Ma e ials and Me hods
2.1. Pa icipan s
All use s a se e al i ness cen e s we e in i ed o pa icipa e in he s udy, p o ided
hey had been en olled a hei cen e o mo e han h ee mon hs (N= 4857). In he end,
361 use s comple ed he ques ionnai e success ully, esul ing in a ma gin o e o o 5% and
a 95% con idence le el. Thei ages anged om 18 o 64 yea s old, wi h he mean age being
29.05
±
11.40. A o al o 39.9% o he pa icipan s we e women. The majo i y o pa icipan s
had comple ed seconda y educa ion (50.6%). Thei mean equency o physical ac i i y
was 3.63
±
1.27 sessions pe week, while he mean ime pe session was
90.49 ±27.38 min.
2.2. Measu es
Se e al ins umen s we e used o conduc his s udy. The i s one was he Spanish
e sion o he Exe cise Addic ion In en o y (EAI) [
48
]. This scale is made up o 6 i ems.
Once he da a had been collec ed, eliabili y, as measu ed wi h C onbach’s alpha, was 0.708.
Highe sco es indica e g ea e symp oms o exe cise addic ion.
The second ins umen used was he Spo s O ganiza ions Pe cep ion Scale, e sion
2 (EPOD2) [
49
]. This is a scale o measu ing he sa is ac ion o use s o spo s se ices.
Reliabili y, as measu ed wi h C onbach’s alpha, was 0.943. Finally, he Spanish e sion
o he ques ionnai e In en ion o Pa ake in Leisu e-Time Physical Ac i i y [
50
] was used.
This is a 3-i em scale o measu ing in en ions o ca ying ou physical ac i i y du ing
leisu e ime. Reliabili y, as measu ed wi h C onbach’s alpha, was 0.969. In bo h cases,
highe sco es indica e g ea e sa is ac ion and g ea e in en ion o engage in LTPA.
Responden s we e ins uc ed o a e hei le el o ag eemen wi h each i em using
a i e-poin Like scale anging om “s ongly disag ee” (1) o “s ongly ag ee” (5).
Addi ional ques ions we e added o he scales: age, gende , le el o educa ion, numbe o
physical ac i i y sessions pe week, and mean ime pe session (Table 1).
Sus ainabili y 2021,13, 129 4 o 12
Table 1. Desc ip i e s a is ics.
Ins umen I ems Mean SD Skewness Ku osis
Sa is ac ion (EPOD2)
Joining his ins i u ion was a good decision. 4.22 0.79 −0.728 −0.025
I am glad I joined his club. 4.31 0.75 −0.876 0.400
I was a good decision o engage in spo s ac i i ies in
his club. 4.20 0.78 −0.690 0.032
I am pleased o be en olled in his ac i i y. 4.32 0.75 −0.906 0.418
Exe cise Addic ion In en o y
(EAI)
Exe cise is he mos impo an hing in my li e. 2.93 1.13 −0.062 −0.575
Con lic s ha e a isen be ween me and my amily
and/o my pa ne abou he amoun o exe cise I do.
1.68 1.10 1.456 1.298
I use exe cise as a way o changing my mood. 3.58 1.21 −0.681 −0.188
O e ime I ha e inc eased he amoun o exe cise I do
in a day. 3.54 1.09 −0.471 −0.326
I I ha e o miss an exe cise session, I eel moody
and i i able. 2.42 1.30 0.473 −0.827
I I cu down he amoun o exe cise I do, and hen
s a again, I always end up exe cising as o en as I
did be o e.
2.99 1.24 .005 −0.936
In en ion o pa ake in
leisu e- ime physical ac i i y
I in end o do physical exe cise a leas h ee imes a
week nex mon h. 4.62 0.77 −1.417 1.172
I am planning o do physical exe cise a leas h ee
imes a week nex mon h. 4.54 0.78 −1.070 1.645
I ha e decided o do physical exe cise a leas h ee
imes a week nex mon h. 4.58 0.78 −1.192 1.013
No e: SD: S anda d De ia ion.
2.3. P ocedu e
The p esen s udy me he es ablished sa e y and e hical s anda ds and was app o ed
by he Resea ch E hics Commi ee o he Andalusian Regional Go e nmen (Andalusia,
Spain). The heads o he pa icipa ing o ganiza ions we e in o med abou he objec i es
and pu pose o he s udy, which was conduc ed once hei app o al was ob ained. The
con iden iali y o he da a and he anonymi y o he pa icipan s we e p ese ed in compli-
ance wi h he Spanish O ganic Law 3/2018 on Pe sonal Da a P o ec ion and Gua an ee
o Digi al Righ s. The undamen al p inciples es ablished in he Decla a ion o Helsinki
(2013, B azilian e ision) we e also obse ed a all imes. All pa icipan s ga e hei in-
o med consen p io o da a collec ion, which was ca ied ou using a sel -adminis e ed
ques ionnai e in he p esence o a esea che .
2.4. Da a Analysis
Ini ially, using he so wa e S a is ical Package o he Social Sciences, e sion 22.0
(IBM, A monk, NY, USA), desc ip i e s a is ics (means and s anda d de ia ions), skew-
ness, ku osis, he co ela ions be ween he la en a iables unde s udy, A e age Va iance
Ex ac ed (AVE), Composi e Reliabili y (CR), and C onbach’s alpha we e calcula ed. Sub-
sequen ly, he model ela ing exe cise addic ion, sa is ac ion wi h he se ice ecei ed,
and in en ion o pa ake in LTPA in he o e all sample (model 0) was hen es ed using
a Con i ma o y Fac o Analysis (CFA). The analysis was pe o med using he maximum
likelihood me hod, ollowing Thompson’s ecommenda ion (2004). The adjus men o he
model was assessed by examining a ious indices: he Compa a i e Fi Index (CFI), he
Goodness o Fi Index (GFI), he Roo Mean Squa e E o o App oxima ion (RMSEA), he
Roo Mean Residual (RMR), and he Expec ed C oss-Valida ion Index (ECVI). The
χ
2 alue
(CMIN) and he χ2 alue/deg ees o eedom (CMIN/DF) we e also used.
A mul i-g oup analysis was ca ied ou o e i y he in a iance o he ac o ial s uc-
u e [
51
]. The g oup was di ided by gende . The aim was o check ha he e we e no
signi ican di e ences, in each o he a iables s udied, be ween a model wi hou in a i-
ance and di e en models wi h es ic ions in some pa ame e s. The i indices used we e
Sus ainabili y 2021,13, 129 5 o 12
he same as hose used o he CFA o he model. Finally, he s anda dized eg ession
coe icien s we e calcula ed o he ela ionships in he model based on he gende o i ness
cen e use s.
3. Resul s
The in en ion o pa ake in LTPA yielded a mean alue highe han ha o sa is ac ion
wi h he spo s and i ness se ice ecei ed (Table 2). Exe cise addic ion sco ed 17 poin s
on a e age, wi h he esul ha his popula ion may be conside ed o be symp oma ic.
Di e ences be ween men and women we e obse ed only in men’s in en ion o engage in
physical ac i i y.
Table 2.
Means, s anda d de ia ions, ANOVA, and signi icance le els. Co ela ions, C onbach’s alpha on he diagonal,
A e age Va iance Ex ac ed (AVE), and Composi e Reliabili y (CR).
Mean Men Women 1 2 3 AVE CR
1. Sa is ac ion 4.26 ±0.71 4.22 4.33 (0.943) 0.321 ** 0.282 ** 0.855 0.959
2. In en ion-LTPA 4.58 ±0.76 4.68 * 4.42 * (0.963) 0.359 ** 0.942 0.980
3. Exe cise addic ion 17.17 ±4.53 17.46 16.74 (0.708) 0.643 0.804
No e: ** p< 0.01; * p< 0.05.
In e nal consis ency was assessed o each esea ch cons uc . The esul s o he CR
es e ealed ha all alues we e g ea e han 0.6. The calcula ion o he AVE sugges s
con e gen alidi y, since he esul ing AVE alues o he esea ch cons uc s we e close o
o abo e he ecommended alue o 0.5. The disc iminan alidi y o he da a was e i ied
by calcula ing he co ela ion ma ix be ween he ac o s. As can be obse ed, he e was a
signi ican posi i e co ela ion be ween he ac o s ha make up he s udy.
The alidi y o he ac o s uc u e o he model ela ing he EAI, sa is ac ion, and
in en ion-LTPA was checked using a se o goodness-o - i indices, yielding adequa e alues
(Table 3; model 0). The model e ealed ela ionships be ween he EAI and sa is ac ion,
be ween sa is ac ion and in en ion-LTPA, as well as be ween he EAI and in en ion-LTPA
(Table 5).
Table 3. Assessmen o he measu emen in a iance o he model.
Model CMIN DF CMIN/DF CFI RMSEA RMR GFI ECVI
0 68.809 50 1.376 0.996 0.032 0.033 0.970 0.347
1 184.920 122 1.516 0.986 0.038 0.049 0.930 0.849
2 199.458 132 1.511 0.985 0.038 0.061 0.926 0.834
3 201.177 135 1.490 0.985 0.037 0.065 0.925 0.822
4 209.550 138 1.518 0.984 0.038 0.075 0.922 0.829
5 277.566 152 1.826 0.972 0.048 0.074 0.900 0.940
No e: Model 0 indica es i indices o he o e all sample; model 1 indica es no pa ame e s cons ained o be equal
ac oss g oups; model 2, ac o loadings cons ained o be equal; model 3, s uc u al weigh s and ac o loadings
cons ained o be equal; model 4, s uc u al esiduals, s uc u al weigh s, and ac o loadings cons ained o be
equal; model 5, measu emen esiduals, s uc u al esiduals, s uc u al weigh s, and ac o loadings cons ained
o be equal.
A e he ac o s uc u e o he model was e i ied, ac o in a iance es s we e
pe o med in o de o compa e he model based on he use s’ gende . When aking in o
accoun he di e ences in CMIN/DF be ween he model wi hou es ic ions (model 1)
and he es o he models wi h es ic ions (models 2–4), no di e ences we e obse ed.
Howe e , di e ences we e ound be ween model 1 and model 5 (Table 4). Using model 2
as co ec , Table 4shows he p esence o di e ences wi h model 5. Model 3 di e s om
model 4 and model 5. Al hough he di e ences in
χ
2 do no suppo he hypo hesis o
in a iance, he es o he indices con adic his conclusion. The CFI alues in he models
we e ound o be simila , wi h a di e ence be ween hem o
−
0.01, excep o he alue in

Sus ainabili y 2021,13, 129 6 o 12
model 5. The RMSEA alues in he models we e also ound o be simila excep o he
alue in model 5, which was 0.01. Simila ly, when obse ing he ECVI index, di e ences
in adjus men s a e minimal, excep o model 5, which indica es ha he di e en models
ha e e y simila alues. All o his poin s o he ac o ial in a iance o he model (Table 3).
Table 4. Compa ison be ween models using measu emen in a iance p ocedu es.
Model Di . DF Di . CMIN p Di . CFI Di . RMSEA
Assuming model 1
o be co ec
2 10 14.538 0.150 −0.001 0.000
3 13 16.257 0.236 −0.001 −0.001
4 16 24.630 0.077 −0.002 0.000
5 30 92.646 0.000 −0.014 0.01
Assuming model
2 o be co ec
3 3 1.718 0.633 0.000 −0.001
4 6 10.092 0.121 −0.001 0.000
5 20 78.108 0.000 −0.013 0.01
Assuming model
3 o be co ec
4 3 8.374 0.039 −0.001 0.001
5 17 76.390 0.000 −0.013 0.012
Assuming model
4 o be co ec 5 14 68.016 0.000 −0.012 0.01
No e: Model 1 indica es no pa ame e s cons ained o be equal ac oss g oups; model 2, ac o loadings cons ained
o be equal; model 3, s uc u al weigh s and ac o loadings cons ained o be equal; model 4, s uc u al esiduals,
s uc u al weigh s, and ac o loadings cons ained o be equal; model 5, measu emen esiduals, s uc u al
esiduals, s uc u al weigh s, and ac o loadings cons ained o be equal. Di . CMIN = di e ence be ween model
and he o he models; Di . DF = di e ence be ween model and he o he models; p = signi icance le el be ween
models; Di . CFI = di e ence be ween model and he o he models; Di . RMSEA = di e ence be ween model and
he o he models.
Model 1 ob ained adequa e alues in all indices. Mo eo e , i s alues we e all posi i e
and we e e y simila o he es o he models. The e o e, his model was conside ed o be
sui able o compa ing he wo g oups based on gende .
Values in Table 5show ha exe cise addic ion was di ec ly and signi ican ly ela ed o
sa is ac ion in he o e all sample. Rega ding gende , he
β
- alue was highe in he emale
g oup han in he male g oup. Exe cise addic ion was an an eceden o he in en ion o
engage in LTPA among i ness cen e use s. Women showed a sligh ly highe
β
- alue.
Simila ly, sa is ac ion is an an eceden o in en ions o engage in LTPA in he o e all sample
and ac oss gende s. The β- alues we e sligh ly highe in men han in women.
Table 5.
S anda dized eg ession weigh s. Di ec and indi ec e ec s o he ini ial model and o he models acco ding o
gende . Hypo hesis es ing.
O e all Sample Male Female
Di ec
E ec s
Indi ec
E ec s
Di ec
E ec s
Indi ec
E ec s
Di ec
E ec s
Indi ec
E ec s
Hypo hesis ββββββ
H. 1 SATIS ←EAI 0.325 ** – 0.271 ** – 0.418 ** –
H. 2 IPLTPA ←EAI 0.378 ** 0.069 ** 0.344 ** 0.067 ** 0.374 ** 0.087 **
H. 3 IPLTPA ←SATIS 0.211 ** – 0.247 ** – 0.209 * –
No e: ** p < 0.01; * p < 0.05. EAI = Exe cise Addic ion In en o y; SATIS = Sa is ac ion; IPLTPA = In en ion o pa ake in leisu e- ime physical
ac i i y.
4. Discussion
The objec i e o his s udy was o de e mine whe he exe cise addic ion can be an
an eceden o sa is ac ion wi h he se ice ecei ed, an an eceden o sa is ac ion wi h
he physical ac i i y pe o med, and/o an an eceden o he in en ion o con inue o
pe o m LTPA a i ness cen e s. Th ee hypo heses we e o mula ed: H1, addic ion o
Sus ainabili y 2021,13, 129 7 o 12
physical exe cise among i ness cen e use s is an an eceden o sa is ac ion wi h he spo s
se ice and wi h he physical ac i i y pe o med; H2, exe cise addic ion is a p ecu so
o he in en ion o pa ake in LTPA; H3, sa is ac ion wi h he physical ac i i y pe o med
and wi h he spo s se ice can be an an eceden o he in en ion o pa ake in LTPA. The
s udy con i med ha he h ee la en a iables a e ela ed, wi h exe cise addic ion being
an an eceden o bo h sa is ac ion and he in en ion o pe o m physical ac i i y among
i ness cen e use s. These esul s a e ele an bo h o hose esponsible o i ness cen e
ma ke ing s a egies and o p o essionals wo king on o esea ching exe cise addic ion.
In o de o e alua e he sui abili y o he model being es ed, a g oup o indices was
join ly assessed. The CFI and GFI alues we e 0.996 and 0.970, espec i ely, ep esen ing
an excellen esul [
52
]. The RMSEA alue, 0.032, indica ed a good model i [
52
]. The
RMR alue was 0.033, which is deemed o be accep able [
53
]. The a io be ween
χ
2 and he
numbe o deg ees o eedom, which was 1.376, may be conside ed accep able [
52
,
54
,
55
].
The o iginal model’s se o i indices may be conside ed o be accep able, so he model is
deemed sui able o his popula ion.
A mul i-g oup analysis was ca ied ou o e i y he in a iance o he ac o ial s uc-
u e [
51
]. The g oup was di ided by gende . The aim was o check ha he e we e no
signi ican di e ences, in each o he a iables s udied, be ween a model wi hou in a iance
and di e en models wi h es ic ions in some pa ame e s.
When aking in o accoun he di e ences in CMIN/DF be ween he model wi hou e-
s ic ions (model 1) and he es o he models wi h es ic ions (models 2–4), no s a is ically
signi ican di e ences we e ound. No signi ican di e ences in
χ
2 be ween he un es ic ed
model (model 1) and he es o he models (2–4) we e ound. Howe e , di e ences we e
ound be ween model 1 and model 5 and be ween model 2 and model 5. These esul s
could ep esen g ounds o ejec ing he in a iance o he model. Ne e heless, because
he CMIN/DF coe icien is sensi i e o sample size, he c i e ion es ablished by [
56
] was
also used, whe eby
∆
CFI alues lowe han o equal o 0.01 indica e ha he null hypo hesis
o in a iance canno be ejec ed. The
∆
CFI alues ound in his s udy when compa ing he
un es ic ed model wi h he es o he models sugges he in a iance o he ac o s uc u e
o he scale. In addi ion, he se o indices o each o he models is adequa e, p omp ing
he in a iance o he model based on gende o be con i med.
Once he in a iance o he model had been e i ied, i was necessa y o analyze he
mean alues ob ained in he cons uc s unde s udy. Fi s ly, sa is ac ion wi h he se ice
ecei ed shows an accep able alue, wi h no di e ences be ween men and women. This
esul is consonan wi h se e al s udies ha do no epo di e ences in sa is ac ion by
gende in use s o a mean age simila o ha o he p esen pape [
49
,
57
]. By con as ,
such di e ences ha e been ound in s udies wi h mean ages abo e 30 yea s old [
58
,
59
].
Sociodemog aphic da a should be ega ded as an impo an a iable in u u e s udies ha
include he sa is ac ion a iable.
Secondly, in his s udy, di e ences eme ged be ween men and women in e ms o
he in en ion o engage in LTPA a i ness cen e s, wi h men showing a g ea e in en ion
o engage in physical ac i i y. Simila esul s ha e been epo ed in o he s udies [
60
],
al hough hese ha e no co e ed i ness cen e s despi e hei high exe cise addic ion a es.
The gende di e ences obse ed may be due o women’s dissa is ac ion wi h hei body
image [
61
], hei conce n abou weigh con ol [
62
], o he belie ha physical ac i i y
has no heal h bene i s [
63
]. Fu he s udies a e hus necessa y o con inue o explo e his
ela ionship in di e en con ex s.
Finally, no di e ences we e obse ed in exe cise addic ion by gende , in line wi h o he
esea ch [
20
]. Howe e , he e seems o be e idence ha exe cise addic ion in emale i ness
cen e use s in luences hei subsequen assessmen s and beha io s, which ein o ces he
need o u he s udies [22].
The pu pose o his pape was o explo e he ela ionships be ween he cons uc s
o addic ion, sa is ac ion, and in en ion. The esul s o he s uc u al equa ion model
show ha exe cise addic ion, o he gene al popula ion, is a p ecu so o sa is ac ion wi h
Sus ainabili y 2021,13, 129 8 o 12
he se ice and, o a g ea e ex en , o he in en ion o pa ake in LTPA. This esul was
expec ed, since he e y concep o addic ion implies he willingness o engage in mo e
physical ac i i y, as poin ed ou by o he s udies [
5
]. The ole o he spo s se ice, in his
case assessed h ough sa is ac ion, may be cha ac e ized as a mode a o , since i limi s
o educes he in en ion o do mo e LTPA. This mode a ion o he ela ionship be ween
addic ion and in en ion, h ough sa is ac ion wi h he se ice, may be le e aged bo h by
heal hca e p o essionals and by hose in cha ge o i ness cen e s as a wo king ool o hei
di e en objec i es.
When analyzing di e ences in ela ionships by gende , i should i s be no ed ha he
ways in which men and women unde s and and expe ience physical ac i i y and spo s
a e comple ely di e gen . In he case o women, hey a e in luenced by he aes he ics and
heal h dyad, while in he case o men, he ocus is on a ques o imp o ed pe o mance
and compe i i eness, al hough aes he ic alues also play a ole [
62
,
64
–
66
]. The concep o
aes he ics in women is ela ed o sel -pe cep ion o body image and accep ance o body
weigh [
67
–
69
], wo e y powe ul a iables in luencing women’s beha io [
70
]. The
p esen s udy shows
β
- alues ha a e much highe among women han among men in
he ela ionship be ween exe cise addic ion and sa is ac ion wi h he se ice ecei ed.
Al hough we ailed o ind simila s udies including he sa is ac ion a iable, we did ind
one s udy whe e emale use s o low-cos cen e s displayed a highe alue in pe cei ed
quali y, wi h pe cei ed quali y being an an eceden o sa is ac ion [71].
When conside ing he ela ionship be ween addic ion and in en ion o pa ake in
LTPA, he
β
- alue was also highe in women, al hough he di e ence be ween men and
women was much smalle han in he p e ious ela ionship. These esul s show ha ,
when i comes o physical ac i i y among emale i ness cen e use s, addic ion is a mo e
impo an a iable han in men. This may be explained by he e y pu pose o i ness
cen e s, which e lec women’s expec a ions o a g ea e ex en wi h hei ocus on body
image [72].
The e o e, his s udy shows ha he addic ion ac o in women has a g ea e e ec
on sa is ac ion wi h he se ice ecei ed and ha women a e mo e willing o con inue
o engage in physical ac i i y han men. Howe e , when he addic ion a iable is no
conside ed and sa is ac ion wi h he spo s se ice and he in en ion o engaging in LTPA
a e ela ed, he
β
- alue ob ained is highe in men han in women [
73
,
74
]. This could
be explained by sa is ac ion being an an eceden o loyal y, unde s ood as, among o he
hings, he in en ion o epea o epu chase a se ice, which is in luenced by he use ’s
expe ience [
75
,
76
]. Men a e he mos expe ienced use s in he ield o spo s se ices, gi en
ha s udies con i m ha women ha e lowe le els o physical ac i i y and leisu e- ime
exe cise han men [
77
]. E en he in en ion o be physically ac i e is lowe in women han
in men [78], which could explain a highe β- alue among men.
I is impo an o emphasize ha hese esul s should no only be ound in women
addic ed o exe cise, who a e a om he majo i y o women a ending i ness cen e s [
20
].
Spo s cen e s should a emp o ensu e ha women who a e no addic ed o exe cise also
display high le els o hese a iables. They could achie e his by using hei acili ies,
special p omo ions, ac i i ies, sales, e c. o imp o e cus ome sa is ac ion [79].
One majo limi a ion o his s udy is he size o he pa icipan popula ion. The model
should be es ed in la ge sample g oups, which would make i possible o con i m he
ela ionships be ween he cons uc s. Ano he limi a ion o his s udy lies in he lack o
knowledge a ailable abou he mo i a ional ac o s shaping indi iduals’ engagemen in
i ness ac i i ies, so i is also impo an o unde s and how di e en deg ees o engagemen
can al e he model. All ma e ial and human ac o s a ec ing sa is ac ion and he way in
which spo s se ices a e deli e ed should be aken in o conside a ion.
Sus ainabili y 2021,13, 129 9 o 12
5. Conclusions
Exe cise addic ion is an an eceden o sa is ac ion wi h he se ice ecei ed and o he
in en ion o engage in physical ac i i y among i ness cen e use s, displaying signi ican
posi i e co ela ions be ween hese la en a iables.
The esul s ega ding he ela ionships be ween he a iables o addic ion, sa is ac ion,
and in en ion show signi ican di e ences be ween men and women. The alues ob ained
o he ela ionships be ween addic ion and sa is ac ion and be ween addic ion and in en-
ion a e highe in women. Howe e , when i comes o he ela ionship be ween sa is ac ion
and in en ion, men ob ain highe alues han women. Addic ion o physical exe cise is a
mo e signi ican a iable in women han in men.
These esul s can be used by i ness cen e p o essionals o achie e highe le els
o sa is ac ion among use s; sa is ac ion can become a po en ially signi ican media ing
a iable in he ela ionship be ween addic ion and in en ion o pa ake in LTPA.
Au ho Con ibu ions:
Concep ualiza ion, V.M.-L. and A.S.; me hodology, A.N. and A.F.-M.; so -
wa e, J.Á.-M.; alida ion, V.M.-L. and A.S.; o mal analysis, A.N.; in es iga ion, A.N.; w i ing—
o iginal d a p epa a ion, A.F.-M.; V.M.-L. and A.N.; w i ing— e iew and edi ing, A.F.-M. and
A.N.; isualiza ion, J.Á.-M.; supe ision, A.F.-M. and A.N. All au ho s ha e ead and ag eed o he
published e sion o he manusc ip .
Funding: This esea ch ecei ed no ex e nal unding.
Ins i u ional Re iew Boa d S a emen :
The s udy was conduc ed acco ding o he guidelines o he
Decla a ion o Helsinki, and app o ed by he Resea ch E hics Commi ee o he Andalusian Regional
Go e nmen (Andalusia, Spain)
In o med Consen S a emen :
In o med consen was ob ained om all subjec s in ol ed in he
s udy.
Con lic s o In e es : The au ho s decla e no con lic o in e es .
Abb e ia ions
LTPA Leisu e- ime physical ac i i y
EAI Exe cise Addic ion In en o y
EPOD2 Spo s O ganiza ions Pe cep ion Scale: e sion 2
AVE A e age Va iance Ex ac ed
CR Composi e Reliabili y
CFA Con i ma o y Fac o Analysis
CFI Compa a i e Fix Index
GFI Goodness o Fi Index
RMSEA Roo Mean Squa e E o o App oxima ion
RMR Roo Mean Residual
ECVI Expec ed C oss Valida ion Index
CMIN chi-squa ed
CMIN/DF chi-squa ed/deg ees o eedom
SD S anda d De ia ions
ANOVA Analysis o Va iance
Re e ences
1.
C uz-Sánchez, E.D.L.; Mo eno-Con e as, M.I.; Pino-O ega, J.; Ma ínez-San os, R. Ac i idad ísica du an e el iempo lib e y su
elación con algunos indicado es de salud men al en España. Salud Men al 2011,34, 45–52.
2.
Eu opean Commission. Special Eu oba ome e 472. Spo and Physical Ac i i y. 2018. A ailable online: h ps://da a.eu opa.eu/
euodp/en/da a/da ase /S2164_88_4_472_ENG (accessed on 10 Oc obe 2020).
3.
Eu opeAc i e. Eu opeAc i e Eu opean Heal h & Fi ness Ma ke Repo (EHFMR). 2019. A ailable online: h ps://www2.deloi e.
com/con en /dam/Deloi e/de/Documen s/consume -business/Eu opeAc i e_Deloi e_EHFMR%202019_DE.pd (accessed on
12 Oc obe 2020).
4.
Má quez, S.; de la Vega, R. Exe cise addic ion: An eme gen beha io al diso de . Nu . Hosp.
2015
,31, 2384–2391. [C ossRe ]
[PubMed]
5.
Be czik, K.; G i i hs, M.D.; Szabo, A.; Ku imay, T.; Kokonyei, G.; U ban, R.; Deme o ics, Z. Exe cise addic ion—The eme gence
o a new diso de . Aus alas. Epidemiol. 2014,21, 36–40.