Obse a ional S udy
Spa ial clus e s and empo al
ends o cu aneous malignan
melanoma mo ali y in Spain
Luc
ıa Cayuela
1
,
Jos
e-Juan Pe ey a-Rod iguez
2,3
,
Juan-Ca los He na
´ndez-Rod
ıguez
3
and
Au elio Cayuela
4
Abs ac
Objec i e: To analyse ime ends in mo ali y due o cu aneous malignan melanoma (CMM) in
Spain be ween 1980 and 2020, and o iden i y possible spa ial clus e s o p o inces wi h an excess
isk o CMM mo ali y du ing he pe iod 2011–2020.
Me hods: Joinpoin eg ession models we e used o assess empo al ends in da a om he
Spanish Na ional Ins i u e o S a is ics. C ude a es, s anda dized mo ali y a io (SMR), smoo hed
ela i e isk (RR), and pos e io p obabili ies (PP) o RR >1 du ing he pe iod 2011–2020 we e
calcula ed. The Global Mo an I index was used o assess global spa ial au oco ela ion.
Resul s: Two ime-pe iods we e de ec ed in women: a signi ican inc ease du ing 1980–1994
(annual pe cen change [APC], 6.5% pe yea ) and a e s abiliza ion du ing 1994–2020 (nonsig-
ni ican APC, 0.17%). A simila pa e n was obse ed in men, wi h h ee pe iods comp ising a
signi ican inc ease du ing 1980–1985 (APC, 16.59%), a slowing o he inc ease du ing 1985–1998
(APC, 4.40%), and s abiliza ion du ing 1998–2020 (nonsigni ican APC, 0.37%). Spa ial analysis
showed g ea e spa ial he e ogenei y wi h an eas -no h pa e n in men compa ed wi h he
pa e n in women, which ended o be concen a ed in no h-wes e n a eas.
Conclusion: Mo ali y a es associa ed wi h CMM in Spain ha e emained s able in ecen yea s.
The e we e p o incial clus e s ha exhibi ed an excess isk o mo ali y om CMM, wi h
di e en pa e ns acco ding o sex.
1
Depa men o In e nal Medicine, Hospi al Se e o
Ochoa, Legan
es, Spain
2
Depa men o Medicine, Uni e si y o Se ille, Se ille,
Spain
3
Depa men o De ma ology, Vi gen del Roc
ıo Uni e si y
Hospi al, Se ille, Spain
4
Uni o Public Heal h, P e en ion and Heal h P omo ion,
Sou h Se ille Heal h Managemen A ea, Se ille, Spain
Co esponding au ho :
Jos
e-Juan Pe ey a-Rod
ıguez, Depa men o
De ma ology, Vi gen del Roc
ıo Uni e si y Hospi al &
Depa men o Medicine, Uni e si y o Se ille, A da
Manuel Siu o s/n, Se ille 41013, Spain.
Email: jpe e[email p o ec ed]
Jou nal o In e na ional Medical Resea ch
2025, Vol. 53(2) 1–13
!The Au ho (s) 2025
A icle euse guidelines:
sagepub.com/jou nals-pe missions
DOI: 10.1177/03000605251319609
jou nals.sagepub.com/home/im
C ea i e Commons Non Comme cial CC BY-NC: This a icle is dis ibu ed unde he e ms o he C ea i e
Commons A ibu ion-NonComme cial 4.0 License (h ps://c ea i ecommons.o g/licenses/by-nc/4.0/) which pe mi s
non-comme cial use, ep oduc ion and dis ibu ion o he wo k wi hou u he pe mission p o ided he o iginal wo k is a ibu ed
as speci ied on he SAGE and Open Access pages (h ps://us.sagepub.com/en-us/nam/open-access-a -sage).
Keywo ds
Melanoma, epidemiology, mo ali y, ecological s udy, skin cance
Da e ecei ed: 4 July 2024; accep ed: 19 Decembe 2024
In oduc ion
In 2020, abou 325 000 people wo ldwide
(174 000 men and 151 000 women) we e
diagnosed wi h cu aneous malignan mela-
noma (CMM), and a ound 57 000 (32 000
men and 25 000 women) died om he dis-
ease.
1
I he a es a e main ained, he
bu den o CMM is expec ed o inc ease o
510 000 new cases (app oxima ely 50%
inc ease) and up o 96 000 dea hs (app ox-
ima ely 68% inc ease) by 2040.
1
O e all ends in CMM a e inc easing in
e ms o incidence and mo ali y,
2,3
bu
he e a e na ional and egional excep ions,
wi h g ea a iabili y be ween coun ies
(incidence a es a y six- old ac oss EU-27
coun ies, while mo ali y a es a y h ee-
old),
4
and also be ween sexes and age
g oups. In addi ion, he pe sis en gende
mo ali y gap pe sis s in mos coun ies
and has inc eased o e ime.
5
Since 2013, he mo ali y a e o CMM
has declined apidly in he USA,
6
Canada,
7
and Ge many,
8
wi h he in oduc ion o
new sys emic ea men s oge he wi h
immune checkpoin inhibi o s and a ge ed
he apies o me as a ic melanoma.
9
P e ious s udies ha e analysed he ends
o CMM mo ali y in Spain, a a na ional
le el and in some au onomous communi-
ies.
10–13
These s udies showed ha mo al-
i y a es sha ply inc eased om he ea ly
1970s o he ea ly 1990s, ollowed by a s a-
bilisa ion obse ed in he age-s anda dized
a es o all ages. Analysis o mo ali y
ends by age g oup and bi h coho
showed a slowe inc ease (men) o s abilisa-
ion (women) in age-adjus ed CMM
mo ali y a es, due o he dec ease in
mo ali y egis e ed in younge age g oups
and ecen coho s. I cu en a es pe sis ,
o e all a es a e expec ed o g ow in he
coming yea s, al hough a es among
young people will p og essi ely s abilise o
decline.
14
Despi e his, in o ma ion on he
geog aphical dis ibu ion o CMM mo al-
i y is sca ce in Spain.
15
Visualisa ion and ca og aphic analysis
o e en s du ing a ime-pe iod ha e been
conside ed as an essen ial ins umen o
disease moni o ing. The iden i ica ion o
clus e s is also impo an in CMM epidemi-
ological s udies.
16
When a s a is ically
signi ican excess o cases is obse ed, sub-
sequen epidemiological s udies can in es i-
ga e whe he he clus e is ela ed o he
en i onmen , heal h ca e con ac , o o he
cha ac e is ics o he esiden popula ion,
and may explain p e iously unknown ae i-
ologies. In addi ion, knowledge o he geo-
g aphical pa e ns o CMM in a popula ion
can be used o mo e e ec i ely a ge sc een-
ing and p e en ion e o s.
17
The e o e, he
aim o he p esen s udy was o analyse he
empo al ends in CMM mo ali y in Spain
be ween 1980 and 2020, and o iden i y pos-
sible spa ial clus e ing o CMM mo ali y a
he p o incial le el be ween 2011 and 2020.
Ma e ials and me hods
Da a sou ce
Fo his obse a ional s udy, da a on
dea hs due o CMM and he popula ion
equi ed o he calcula ions we e ob ained
om he Spanish Na ional Ins i u e o
S a is ics (INE).
18
All dea hs wi h CMM
codes 172 and C43 o he In e na ional
2Jou nal o In e na ional Medical Resea ch
Classi ica ion o Disease (ICD-9 h and
ICD-10 h e ision) we e analysed. The
s udy was no deemed o equi e pa ien
consen o e hics commi ee app o al, as
all da a ex ac ed om he Na ional
Ins i u e o S a is ics we e anonymized,
he s udy ollowed he p inciples o good
clinical p ac ice and he Decla a ion o
Helsinki (as e ised in 2013), no pa ici-
pan s we e iden i ied, and no pe sonal
in o ma ion was accessed. The s udy was
pe o med acco ding o he S eng hening
he Repo ing o Obse a ional S udies in
Epidemiology (STROBE) guidelines.
S a is ical analyses
To analyse ime ends, age-s anda dized
mo ali y a es (ASMR) we e calcula ed in
bo h sexes o he pe iod 1980–2020 acco d-
ing o he di ec me hod, aking he
Eu opean S anda d Popula ion as a e e -
ence.
19
Ra es a e p esen ed pe 100 000
inhabi an s and we e analysed by
Joinpoin eg ession wi h a maximum o
h ee possible poin s o change in he
end, using a segmen ed Poisson model
wi h Na ional Cance Ins i u e Joinpoin
eg ession so wa e, e sion 4.9.1.0.
20
To examine he spa ial pa e n o CMM
mo ali y in bo h men and women a he
p o incial le el in Peninsula Spain, he
c ude a es, s anda dized mo ali y a io
(SMR), smoo hed ela i e isk (RR), and
pos e io p obabili ies (PP) o RR >1
du ing he pe iod 2011–2020 we e calcula ed.
To calcula e he SMR, he expec ed cases
we e es ima ed o each p o ince, aking he
a e o Spain as a e e ence. Nex , he SMR
wi h he numbe o obse ed cases was
compu ed as he nume a o and he
numbe o expec ed cases o each p o ince
was compu ed as he denomina o . The
SMR measu es he RR o a p o ince
ega ding he e e ence egion.
Rega ding spa ial da a analysis, di e en
s a is ical es s we e used o assess
o e dispe sion, spa ial au oco ela ion and
o e all spa ial clus e ing. The s anda d
2
-
es and Po ho -Whi inghill es we e
used o es ima e he spa ial he e ogenei y
o he isk sco es and o es o o e all sig-
ni ican di e ences o obse ed and
expec ed cases. The exis ence o global spa-
ial au oco ela ion was assessed wi h he
Global Mo an I index, de e mining whe he
he analysis uni s ended o clus e , sca e ,
o andomise. The local indica o s o spa-
ial associa ion (LISA) s a is ics de ine he
spa ial pa e ns o ho spo s/clus e s o
CMM mo ali y a he p o incial le el.
The LISA e eal signi ican ho spo s
(high alues nex o high [HH]), cold spo s
(low alues nex o low [LL]) and spa ial
ou lie s (high among low [HL] o low
among high [LH]) o mo ali y a es.
Mo an’s index, wi h a signi icance le el o
0.05, was used o assess whe he he a es
showed dis ibu ions wi h a p opensi y o
clus e ing, dispe sion o andomisa ion.
The RRs we e es ima ed o each p o -
ince. S uc u ed (clus e ed) and uns uc-
u ed (he e ogeneous) spa ial dependence
we e modelled using he p oposals o
Besag, Yo k and Molli
e.
21
Spa ial depen-
dence was assumed whe eby isk es ima es
in a gi en a ea depended on neighbou ing
a eas. Since he e we e no neighbou ing
a eas, he island p o inces and he ci ies
o Ceu a and Melilla we e excluded. Fi s
o second o de ‘ andom walks’ (RW1,
RW2) we e conside ed o model he empo-
al e ec s. Models ha included one o he
possible spa io empo al in e ac ions p o-
posed by Kno -Held we e also e alua ed.
22
Model i ing and in e ence we e pe o med
by means o he app oxima e Bayesian
in e ence ‘In eg a ed Nes ed Laplace
App oxima ion’ (INLA) echnique. Model
selec ion echniques, such as De iance
In o ma ion C i e ion (DIC), we e used o
iden i y he mos sui able model o cap u -
ing he empo al end in mo ali y a es
ac oss Spanish p o inces.
23
I he analysis
Cayuela e al. 3
selec ed a model wi h an RW1 e m o he
ime e ec , he changes in mo ali y a es
we e andom luc ua ions om he p e ious
pe iod. Con e sely, i a model wi h an RW2
e m was chosen, he changes depended on
bo h he immedia e pas change and he
change om wo pe iods ea lie , po en ially
indica ing a mo e complex empo al end.
The model wi h he smalle DIC alue was
chosen. One s eng h o he INLA me hod
is ha i may be execu ed wi hin ee
R so wa e h ough he R package
‘R-INLA’,
24,25
which allows calcula ions
o : (i) accu a e es ima es o he spa io em-
po al pa e n o isk (RRs), based on da a
om neighbou ing a eas and adjacen
yea s; and (ii) he pos e io p obabili ies
ha he RRs we e >1 (PP).
To iden i y a eas wi h a highe - han-
expec ed mo ali y a e, decision ules
based on he es ima ed p obabili y ha
he speci ied RR is >1, and he ollowing
h esholds o 0.05, 0.20, 0.80 and 0.95,
26
we e applied. A eas wi h a alue >0.80
we e conside ed a isk, and excess mo al-
i y was signi ican i he p obabili y o
dea h was >0.95. P obabili ies be ween 0.2
and 0.8 we e conside ed o ep esen only
weak indica ions ha he pe cen age a io is
>1, so ha he speci ic mo ali y a e in
hese p o inces is analogous o he e e -
ence mo ali y a e. A eas <0.20 we e con-
side ed low- isk a eas, and a eas wi h a
p obabili y <0.05 we e conside ed o be
a eas wi h a speci ic mo ali y a e signi i-
can ly below he e e ence a e.
The maps we e c ea ed wi h he open-
sou ce so wa e ool o explo a o y spa ial
da a analysis GeoDa, e sion 1.20 o
Windows.
27
Resul s
Age-s anda dized melanoma mo ali y
a es and he ime end by sex o he gen-
e al popula ion in Spain a e shown in
Figu e 1. Al hough a signi ican inc ease
o 1.09% pe yea was obse ed in women
Figu e 1. Time end o age-adjus ed melanoma mo ali y a es in Spain be ween 1980 and 2020.
4Jou nal o In e na ional Medical Resea ch
du ing he o e all pe iod, Joinpoin analy-
sis de ec ed a end change wi h wo pe i-
ods: 1980–1994, ma ked by a signi ican
inc ease in melanoma mo ali y (APC,
6.5% pe yea ; P<0.05) and 1994–2020,
in which a es s abilized (APC, 0.17%;
P>0.05). A simila pa e n was obse ed
o men. Du ing he whole pe iod, he e
was a signi ican inc ease in melanoma
mo ali y (1.76% pe yea ), bu Joinpoin
analysis de ec ed wo poin s o change
ha de e mined h ee ime-pe iods: The
i s pe iod showed a signi ican inc ease
be ween 1980–1985 (APC, 16.59%;
P<0.05), ollowed by a slowe s a is ically
signi ican inc ease be ween 1985–1998
(APC, 4.40%; P<0.05) and inally a s abi-
liza ion o melanoma mo ali y a es
be ween 1998–2020 (APC, 0.37%; P>0.
05). Compa ed wi h he gene al popula ion
end, he age s anda dized melanoma mo -
ali y a e was highe in males o he whole
pe iod.
The spa ial dis ibu ion o c ude a es,
he SMRs, he pos e io median es ima es
(RRs), and he PPs in men and women,
espec i ely ( o mainland Spain 2011–
2020) a e shown in Figu es 2 and 3. Fo
bo h sexes, he lowes DIC alue co e-
sponded o he spa io empo al model wi h
a ype IV in e ac ion and an RW1 o he
ime e ec . Type IV in e ac ion conside s a
s uc u e in bo h space and ime and is
app op ia e when ime ends a y om
one p o ince o ano he bu a e simila o
ha o he adjacen p o inces.
The maps ob ained showed consis en
esul s when applying he di e en spa ial
analysis me hods. The pa e n in men
showed g ea e spa ial he e ogenei y wi h
Figu e 2. Spa ial dis ibu ion o c ude a es (pe 100 000 inhabi an s), s anda dized mo ali y a es (SMRs),
pos e io median es ima es ( ela i e isk [RR]), and pos e io exceedance p obabili ies (PP) o melanoma
mo ali y in women (mainland Spain 2011–2020).
Cayuela e al. 5
an eas -no h pa e n, compa ed wi h he
pa e n in women, which ended o be con-
cen a ed in no h-wes e n a eas. All spa ial
es s we e s a is ically signi ican (P<0.05)
o bo h males and emales.
In women, se en o he 47 s udied p o -
inces we e iden i ied o show an excess o
melanoma mo ali y, wi h PP 80%, all o
which we e loca ed in he no h on he
Can ab ian coas (A Co u~
na, Lugo,
As u ias, Can ab ia and Bizkaia) o close
o i (Palencia and Na a a). In men, 16 o
he 47 s udied p o inces displayed an excess
melanoma mo ali y, wi h PP 80%, mos
o which we e loca ed on he Can ab ian
coas (Gipuzkoa and Bizkaia) o in he
icini y (A aba and Na a a), on he
Medi e anean coas (Ca
´diz, Ma
´laga,
Mu cia, Alican e, Valencia and Ta agona)
o in he icini y (Za agoza, Huesca, Te uel,
Cuenca and Albace e). Addi ionally, Mad id
showed ele a ed melanoma mo ali y.
The es ima ed smoo h spa io empo al
RRs o he pe iod 2011–2020 o all p o -
inces in mainland Spain a e shown in
Figu e 4, displaying dis inc geog aphical
pa e ns o mo ali y isk. Fo women, a
pe sis en no h-sou h g adien was
obse ed, wi h he no he n p o inces con-
sis en ly showing a signi ican ly highe
excess isk compa ed wi h he na ional
a e age. Fo men, he pa e n ollowed an
eas -wes g adien , wi h he eas e nmos
p o inces showing signi ican ly inc eased
isks. In addi ion, he igu e e lec s changes
in mo ali y ends o e ime, consis en
wi h he esul s o he na ional Joinpoin
analysis.
Figu e 3. Spa ial dis ibu ion o c ude a es (pe 100 000 inhabi an s), s anda dized mo ali y a es (SMRs),
pos e io median es ima es ( ela i e isk [RR]), and pos e io exceedance p obabili ies (PP) o melanoma
mo ali y in men (mainland Spain 2011–2020).
6Jou nal o In e na ional Medical Resea ch
Mo an’s I was ound o be 0.58 o
women and 0.57 o men, and was s a is i-
cally signi ican in bo h cases (P<0.05).
When using he LISA me hod (Figu e 5),
spa ial g oups we e clus e ed in he no he n
and eas e n zones o he coun y o men
(all P<0.05), while he spa ial g oups we e
dis ibu ed o a g ea e ex en in he wes e n
and no he n/wes e n p o inces o women,
showing signi ican g oups wi h P<0.05.
The numbe o dea hs ha occu ed
in he peninsula p o inces in Spain,
Figu e 4. Geog aphical ep esen a ion o he spa io empo ally smoo hed ela i e isks (RRs) and pos e io
exceedance p obabili ies (PP) o melanoma mo ali y in all p o inces in Spain, o each yea be ween 2011
and 2020, o women and men, espec i ely.
Figu e 5. Melanoma mo ali y ho spo s/clus e s a he p o incial le el in Spain (2011–2020), calcula ed by
he Local Indica o o Spa ial Associa ion (LISA) me hod, showing: signi ican ho spo s (high alues nex o
high [HH]), cold spo s (low alues nex o low [LL]), and spa ial ou lie s (high among low [HL] o ice e sa
[LH]) o mo ali y a es.
Cayuela e al. 7
dis ibu ed by sex in each yea o s udy, a e
shown in Supplemen a y Table S1.
Discussion
The esul s o he p esen s udy showed a
s abiliza ion o melanoma mo ali y isk in
men and women, which concu s wi h a p e-
iously published s udy ha epo ed s an-
da dized a es in men showing a signi ican
inc ease du ing he pe iod 1992–2014, ol-
lowed by a s abiliza ion un il 2018. In his
same s udy, he melanoma mo ali y a es
in women emained s able du ing he pe iod
1994–2018.
11
Despi e imp o emen s in
diagnosis, ea ly de ec ion and new ea -
men s, he end in mo ali y a es has no
e e sed as obse ed in o he s udies,
7–9
e en o ad anced disease. The e o e, u -
he s udies a e needed o y o iden i y he
de e minan s o his si ua ion in Spain.
In e ms o he p esen geog aphical
analysis, some a eas p esen ed a highe
CMM mo ali y han ha expec ed o he
en i e coun y. Al hough wi h ce ain nuan-
ces, he geog aphical pa e ns obse ed in
men and women we e simila o hose
obse ed in he ‘A las o mo ali y om
cance and o he causes in Spain, 1978–
1992’.
17
He ein, i was ound ha , whils
no s a is ically signi ican geog aphical
agg ega ions we e de ec ed, he p o inces
wi h he highes a es we e concen a ed in
he eas e n and no he n coas al a eas in
men and women, espec i ely. Fu he mo e,
p o incial di e ences and he geog aphical
pa e n we e conside ably mo e ma ked
among men.
The ‘A las de mo alidad en Cas illa-La
Mancha 2003–2014’ showed ha he spa ial
pa e ns o mo ali y o men and women
we e di e en , simila o ha obse ed in
he p esen s udy, wi h excess mo ali y in
he p o ince o Albace e in men bu no
in women.
28
The p esen geog aphical pa e ns may
e lec an inc eased ulne abili y o UV
adia ion due o unheal hy ec ea ional
sun exposu e pa e ns, popula ion ageing
and/o ozone deple ion.
29
Some isk ac o s
o he de elopmen o CMM a e no mod-
i iable by he indi idual (e.g., Fi zpa ick
pho o ype I-III skin), bu o he s, such as
li es yle ac o s, could ha e an in luence.
As p e en ion measu es in his ega d, i is
essen ial o a oid pe iods o in ense bu
in e mi en sun exposu e, such as du ing
holidays in sunny loca ions, and o limi
p olonged exposu e, as expe ienced when
wo king ou doo s.
30
No ably, p e ious epo s in Canada and
he USA,
31,32
compa ing coas al and inland
a eas, documen ed a g ea e incidence o
CMM cases along he coas , a e adjus ing
o socioeconomic s a us, UV index and la -
i ude. The p esen g aphical esul s high-
ligh hese indings isually, as well as
p o iding speci ic de ails on he p o inces
a isk. Inhabi an s o hese p o inces a e
likely o enjoy he beach o ou doo s mo e
equen ly han in o he p o inces, exposing
hem o a highe isk o de eloping CMM.
In addi ion, wa e and sand ha e a highe
su ace e lec ance han g ass, possibly u -
he inc easing he isk o CMM.
Da a on inc eased isk o CMM and la -
i ude a e con adic o y. In Eu ope, he e is
e idence o a ela ionship be ween la i ude
and CMM isk,
33
and he p esen esul s in
women may co obo a e his, whils in men
his ela ionship is weake . The di e en
pa e ns by sex may be a ibu ed o di e -
en ial exposu e o isk ac o s by sex, medi-
a ed by li es yles, and could be due o
gende di e ences in exposu e o po en ial
isk ac o s, pa ly a ibu able o gende
di e ences in li es yles and beha iou al
habi s among men and women.
34
Se e al s udies ha e obse ed a ela ion-
ship be ween socioeconomic s a us (educa-
ional le el, income, and employmen
s a us) and he incidence, mo bidi y, and
mo ali y o CMM. High socioeconomic
s a us was associa ed wi h highe incidence
8Jou nal o In e na ional Medical Resea ch
and low socioeconomic s a us was associa -
ed wi h delayed diagnosis and poo p ogno-
sis in cases o CMM.
35
In gene al, in
coun ies wi h uni e sal heal h sys ems (as
is he case in Spain), he e is a ela ionship
be ween a high socioeconomic le el and a
highe incidence o CMM, while a low
socioeconomic le el is co ela ed wi h
ad anced s ages o he disease.
36
In addi-
ion, a eas wi h a la ge p opo ion o
u al esiden s a e signi ican ly associa ed
wi h highe incidence and highe mo ali y
a es o CMM.
37,38
Unlike incidence a es, which depend
p ima ily on exposu e o isk ac o s, mo -
ali y a es may also e lec di e ences in
medical ca e, as well as he e ec o p e en-
ion campaigns. Beyond disc epancies in
p e en ion, ea ly de ec ion, and egis a-
ion, he signi ican di e ences in CMM
ou comes in Spain may be oo ed in dispa -
i ies h oughou he con inuum o CMM
ca e. To al heal h expendi u e pe capi a
co ela es closely wi h es ima ed CMM su -
i al in Eu ope,
39
and his pa ame e
emains highly he e ogeneous in Spain.
40
Imp o emen s in access o heal h ca e
and ea ly de ec ion o CMM, wi h he sub-
sequen imp o emen in su i al, may
explain pa o he geog aphical di e ences
obse ed in Spain. Howe e , despi e he
cha ac e is ics o he Spanish Na ional
Heal h Sys em, wi h uni e sal co e age,
he e a e suspicions ha he e may be di -
e ences in access o heal hca e and diagno-
sis be ween egions.
41,42
In he USA, immuno he apy con ibu es
o imp o ing su i al imes in me as a ic
CMM, bu geog aphic access o clinical
ials o such he apies is di icul o a sig-
ni ican pa o he popula ion, which may
also be happening in Spain.
43
Al hough new
ea men s ha e played a signi ican ole,
o he ac o s, such as dispa i ies in access
o heal h ca e, egional di e ences in
heal h esou ces, and he e ec i eness
o public heal h campaigns, migh also
in luence he obse ed ou comes. These ac-
o s may a ec bo h ea ly de ec ion and
managemen o ad anced melanoma, con-
ibu ing o he geog aphic a ia ions iden-
i ied in he p esen s udy.
When in e p e ing he cu en esul s,
se e al ac o s should be conside ed.
Mo ali y is no he bes indica o o s udy
CMM dis ibu ion; howe e , i emains he
only comp ehensi e sou ce o in o ma ion
on CMM in Spain. Geog aphical di e ences
migh ha e esul ed in a ia ions in he qual-
i y o ce i ica ion o causes o dea h in
Spain.
44
Al hough wo ICDs we e used
du ing he p esen s udy pe iod (9 h and
10 h ICD e isions), i is unlikely o ha e
impac ed he mo ali y a es o CMM, as
his ca ego y is simila in he wo edi ions
used.
Ecological s udies ha e se e al sho -
comings. The s a is ical echniques
employed in he p esen s udy allowed he
spa ial pa e n o CMM mo ali y o be
desc ibed, bu canno explain he di e en-
ces be ween a eas. Since he da a a e agg e-
ga ed, he le el o exposu e o any isk
ac o o he deceased and non-deceased is
unknown. The e o e, a main limi a ion o
he s udy was ha he impo ance o ac o s
including gene ics, sunbu ns, and anning
bed use in CMM ae iology could no be
assessed. Mo eo e , since dea hs in Spain
a e egis e ed acco ding o he las known
add ess wi hou speci ying how long he
pe son had li ed he e, i is impossible o
know whe he indi iduals cu en ly esiding
in a p o ince ha e li ed he e mos o hei
li es and ha e been exposed o he associa -
ed en i onmen al isk ac o s. Conside ing
ha , any hypo hesis ha sugges s an associ-
a ion be ween he excess mo ali y obse ed
in any p o ince and social inequali ies, he
use o heal hca e se ices, o he indi idual’s
en i onmen al exposu es could all in o he
well-known ecological allacy.
45
Fu he
esea ch combining ecological da a wi h
indi idual-le el in o ma ion is needed o
Cayuela e al. 9
