REVIEW ARTICLE (META-ANALYSIS)
Associa ion Be ween Pain Coping and Symp oms
o Anxie y and Dep ession, and Wo k
Absen eeism in People Wi h Uppe Limb
Musculoskele al Diso de s: A Sys ema ic Re iew
and Me a-analysis
Rod igo N
u~
nez-Co
es, MSc,
a,b,c
Ande Espin, MSc,
b,d,e
So
ıa P
e ez-Alenda, PhD,
a
Rub
en L
opez-Bueno, PhD,
b, ,g
Ca los C uz-Mon ecinos, PhD,
a,c,h
Ka ina Glies Vincen s-Seebe g, MSc,
b
Thomas A. P€
uschel, PhD,
i
Joaqu
ın Cala ayud, PhD,
b,g
La s Louis Ande sen, PhD
b
F om he
a
Depa men o Physio he apy, Physio he apy in Mo ion Mul ispeciali y Resea ch G oup (PTinMOTION), Uni e si y o Valencia,
Valencia, Spain;
b
Na ional Resea ch Cen e o he Wo king En i onmen , Copenhagen, Denma k;
c
Depa men o Physical The apy, Facul y o
Medicine, Uni e si y o Chile, San iago, Chile;
d
Ageing On Resea ch G oup, Depa men o Physiology, Uni e si y o he Basque Coun y (UPV/
EHU), Leioa, Spain;
e
Bioc uces Bizkaia Heal h Resea ch Ins i u e, Ba akaldo, Spain;
Depa men o Physical Medicine and Nu sing, Uni e si y
o Za agoza, Za agoza, Spain;
g
Exe cise In e en ion o Heal h Resea ch G oup (EXINH-RG), Depa men o Physio he apy, Uni e si y o
Valencia, Valencia, Spain;
h
Di ision o Resea ch, De elopmen and Inno a ion in Kinesiology, Kinesiology Uni , San Jos
e Hospi al, San iago,
Chile; and
i
Ecology and E olu iona y Biology Di ision, School o Biological Sciences, Uni e si y o Reading, Reading, Uni ed Kingdom.
Abs ac
Objec i e: To de e mine he p ospec i e associa ion o pain coping s a egies and symp oms o anxie y and dep ession wi h wo k absen eeism in
people wi h uppe limb musculoskele al diso de s.
Da a Sou ces: A sys ema ic sea ch o PubMed, Web o Science, Embase, Coch ane Lib a y, and Scopus da abases was conduc ed om incep ion
o Sep embe 23, 2022.
S udy Selec ion: P ospec i e obse a ional s udies o adul s wi h uppe limb musculoskele al diso de s we e included. Included s udies had o
p o ide da a on he associa ion o pain coping s a egies (ca as ophizing, kinesiophobia, sel -e icacy o ea a oidance) o symp oms o anxie y
and dep ession wi h wo k absen eeism.
Da a Ex ac ion: S udy selec ion, da a ex ac ion, and assessmen o me hodological quali y (Newcas le O awa Scale) we e pe o med by 2 inde-
penden au ho s. Random-e ec s models we e used o quan i a i e syn hesis.
Da a Syn hesis: Eigh een s udies (n=12,393 pa icipan s) we e included. Mos s udies (77.8%) epo ed a leas 1 signi ican associa ion be ween 1
o mo e exposu e ac o s (pain coping s a egies o symp oms o anxie y and dep ession) and wo k absen eeism. Me a-analyses showed a s a is i-
cally signi ican co ela ion be ween he exposu e ac o s o ca as ophizing ( =0.28, 95% con idence in e al [CI]: 0.15 o 0.40; P<.0001) and
symp oms o anxie y and dep ession ( =0.23, 95% CI: 0.10 o 0.34; P=.0003) wi h wo k absen eeism. The co ela ion be ween sel -e icacy and
wo k absen eeism was non-signi ican ( =0.24, 95% CI: -0.02 o 0.47; P=.0747).
Rod igo N
u~
nez-Co
es is suppo ed by he Na ional Resea ch and De elopmen Agency o Chile (ANID/2020-72210026). Rub
en L
opez-Bueno is suppo ed by Eu opean Union—Nex Gene a ion-EU.
The o he au ho s ha e no hing o disclose.
0003-9993/$36 - see on ma e Ó2023 by he Ame ican Cong ess o Rehabili a ion Medicine. Published by Else ie Inc. This is an open access a icle
unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/)
h ps://doi.o g/10.1016/j.apm .2023.07.003
A chi es o Physical Medicine and Rehabili a ion
jou nal homepage: www.a chi es-pm .o g
A chi es o Physical Medicine and Rehabili a ion 2024;105: 781−91
Conclusions: Rehabili a ion eams should conside assessing ca as ophizing and symp oms o anxie y and dep ession o iden i y pa ien s a isk
o wo k absen eeism. Add essing hese a iables may also be conside ed in e u n- o-wo k p og ams o indi iduals wi h uppe limb diso de s.
A chi es o Physical Medicine and Rehabili a ion 2024;105:781−91
Ó2023 by he Ame ican Cong ess o Rehabili a ion Medicine. Published by Else ie Inc. This is an open access a icle unde he CC BY license
(h p://c ea i ecommons.o g/licenses/by/4.0/)
Uppe limb musculoskele al symp oms a e common in he
gene al popula ion and accoun o a signi ican p opo ion o
wo k- ela ed p oblems, wi h a subs an ial e ec on physical
unc ion and heal h ca e u iliza ion.
1,2
The annual incidence o
wo k- ela ed uppe limb musculoskele al diso de s anges om
0.08% o 6.3%, and he p e alence om 0.14% o 14.9%.
3
Fu he mo e, a ecen me a-analysis ocusing on wo ke s in
seconda y indus ies (ie, indus ies esponsible o con e ing
aw ma e ials in o consume p oduc s) in Eu ope iden i ied
ha he mos equen musculoskele al symp oms in he uppe
limb a e loca ed in he shoulde and w is egion, wi h 12-
mon h mean p e alence alues o 50% and 42%, espec i ely.
1
Thus, his heal h condi ion places a g ea economic bu den on
he indi idual, he employe , and socie y because o wo k
absen eeism and a loss o p oduc i i y.
4
Addi ionally, he eco-
nomic bu den o p esen eeism (los p oduc i i y due o illness
and impai ed pe o mance a wo k) is conside ably la ge
among wo ke s wi h musculoskele al symp oms and is associ-
a ed wi h a highe isk o subsequen absen eeism.
5-7
Wo k absen eeism can be de ined in e ms o e u n o wo k
(RTW) o in e ms o du a ion o sick lea e. RTW p o ides eco-
nomic sel -es eem, psychological well-being, and social connec -
edness.
8
Howe e , as he du a ion o sick lea e inc eases, he
likelihood o RTW dec eases and he isk o long- e m disabili y
inc eases.
9
Bo h absen eeism and p esen eeism impose signi ican
indi idual cos s and bu dens, pa icula ly on ulne able popula-
ions. Fo example, wo ke s wi h low socio-economic s a us may
be mo e likely o su e heal h p oblems due o ad e se physical
and psychosocial wo king condi ions.
10
Simila ly, mig an wo k-
e s a e mo e likely o su e occupa ional inju ies han non-
mig an wo ke s.
11
On he o he hand, p e ious s udies ha e
epo ed ha young emale wo ke s may ha e highe le els o
uppe limb pain and musculoskele al diso de s han hei male
coun e pa s.
12
These di e ences may be due o biological di e -
ences (eg, lean muscle mass o endoc ine unc ion) o cul u al
gende s e eo ypes, whe e women end o pe o m mo e epe i i e
and mono onous wo k han men, which may inc ease hei isk o
inju y, pa icula ly in he uppe ex emi y.
13
Gi en ha ime o
wo k can be as long as 304 days o people wi h musculoskele al
condi ions in he uppe limb,
14
ea ly iden i ica ion o ac o s ha
may a ec imely RTW could be impo an in educing cos s and
disabili y, pa icula ly in popula ions a high isk o wo k- ela ed
diso de s.
Pe sonal and en i onmen al ac o s modi y he p og ess o
wo k absen eeism in di e en heal h and inju y condi ions.
15
Fac o s associa ed wi h posi i e e u n- o-wo k ou comes
include op imis ic expec a ions o eco e y and RTW and
socio-economic s a us, while olde age, emale gende , g ea e
pain o disabili y, and g ea e physical demands o wo k a e
associa ed wi h nega i e ou comes.
15
Recen esea ch has
highligh ed he impo ance o s udying pain coping s a egies
(ca as ophic hinking, kinesiophobia, sel -e icacy, ea a oid-
ance) and men al heal h ac o s (symp oms o anxie y and
dep ession) because o hei s ong associa ion wi h pos ope a-
i e pain, unc ion, and esponse o ea men in pa ien s wi h
ch onic pain.
16,17
Wi hin his pa icula amewo k, he ea -
a oidance model sugges s ha indi iduals who hold ca a-
s ophic belie s abou hei pain a e mo e likely o pe cei e
ce ain si ua ions as h ea ening, a oid ce ain ac i i ies and
ul ima ely expe ience educed engagemen , disabili y, and
dep essi e symp oms.
18
In his ega d, se e al sys ema ic
e iews ha e shown ha ea and ca as ophizing play a c u-
cial ole as p edic o s o he de elopmen o ch onic pain and
i s pe sis ence o e ime.
19-22
Howe e , sca ce in o ma ion
exis s abou he associa ion o hese a iables wi h wo k
absen eeism in people wi h uppe ex emi y diso de s. P e ious
e iews in people wi h uppe limb diso de s has shown ha
high sel -e icacy is posi i ely associa ed wi h an ea ly e u n
o wo k (E-RTW),
23
while dep ession is associa ed wi h a
lowe likelihood o e u ning o wo k.
14
On he o he hand, in
pa ien s wi h ca pal unnel synd ome, 2 p e ious sys ema ic
e iews iden i ied ca as ophizing and poo e men al heal h
s a us as p edic o s o poo e employmen ou come a e
su ge y.
24,25
Howe e , o ou knowledge, no p e ious me a-
analysis has examined he associa ion be ween pain coping
s a egies and symp oms o anxie y and dep ession wi h absen-
eeism in people wi h uppe limb diso de s.
Inc easing e idence suppo s ha pain coping s a egies and
men al heal h ac o s (symp oms o anxie y and dep ession) a e
modi iable h ough a ge ed in e en ion s a egies (eg, pain edu-
ca ion, exe cise, cogni i e beha io al he apy, and mind ul-
ness).
26-28
A be e unde s anding o he associa ion be ween hese
ac o s and absen eeism can help ehabili a ion eams (psycholo-
gis s, physio he apis s, occupa ional he apis s, and physicians) o
design s a egies o imp o e pa ien s’ physical and men al heal h,
op imize RTW and, indi ec ly, dec ease economic cos s. The e-
o e, his sys ema ic e iew and me a-analysis aims o de e mine
he associa ion o selec ed pain coping s a egies and symp oms o
anxie y and dep ession wi h wo k absen eeism in people wi h
uppe limb musculoskele al diso de s.
Lis o abb e ia ions:
CI con idence in e al
CES-D Cen e o Epidemiologic S udies Dep ession Scale
CSQ Coping S a egy Ques ionnai e
FABQ Fea A oidance Belie s Ques ionnai e
GSES Gene al Sel -E icacy Scale
E-RTW ea ly e u n o wo k
L-RTW la e e u n o wo k
PASS Pain Anxie y Symp om Scale
PCCL Pain Coping and Cogni ion Lis
PCS Pain Ca as ophizing Scale
PHQ-9 Pa ien Heal h Ques ionnai e
PSEQ Pain Sel -E icacy Ques ionnai e
RTW Re u n o Wo k
SF-36 MC 36-i em sho o m heal h su ey (Men al componen )
TSK Tampa scale o kinesiophobia
VRMCS Ve e ans RAND Men al Componen Sco e
782 R. N
u~
nez-Co
es e al
www.a chi es-pm .o g
Me hods
P o ocol and egis a ion
This sys ema ic e iew and me a-analysis was p ospec i ely egis-
e ed in PROSPERO ( egis a ion numbe CRD42022362385) and
epo ed ollowing he P e e ed Repo ing I ems o Sys ema ic
Re iews and Me a-Analyses (PRISMA 2020) and he Me a-analy-
sis o Obse a ional S udies in Epidemiology checklis .
29,30
Eligibili y c i e ia
Inclusion c i e ia we e based on he Popula ion, Exposu e, Com-
pa ison, Ou come and S udy Design (PECOS) me hodology. (1)
Popula ion: Adul s >18 yea s wi h any ype o musculoskele al
diso de o he uppe limb (ie, shoulde , a m, elbow, hand), includ-
ing g adual onse symp oms, acu e inju ies, o o hopedic su ge -
ies; (2) Exposu e: S udies in es iga ing pain coping s a egies
(ca as ophic hinking, kinesiophobia, ea -a oidance, o sel -e i-
cacy) o selec ed men al heal h ac o s (symp oms o anxie y o
dep ession) using alida ed ques ionnai es (supplemen al able
S1; a ailable online only a h p://www.a chi es-pm .o g/); (3)
Compa ison: Bo h low le els o exposu e and no exposu e o isk
ac o ; (4) Ou comes: Wo k absen eeism, e alua ed in e ms o
ime o RTW (eg, days, weeks, mon hs, yea s), du a ion o sick
lea e (eg, days o wo k missed) o absen eeism du ing ollow-up;
(5) S udy Design: P ospec i e obse a ional s udies. O iginal,
pee - e iewed a icles w i en in English o Spanish we e included
(Publica ion da e: Published om incep ion o Sep embe 23,
2022). All edi o ials, le e s, e iews and me a-analyses, e ospec-
i e, and c oss-sec ional s udies we e excluded.
Da a sou ces and sea ches
A sys ema ic sea ch o PubMed (including he da abase
“MEDLINE”), Web o Science Co e Collec ion, Embase, Coch ane
Lib a y, and Scopus da abases was pe o med om incep ion o Sep-
embe 23, 2022. Speci ic sea ch s a egies, using a combina ion o
MeSH (Medical Subjec Headings) and keywo d e ms, combined
wi h Boolean ope a o s, a e shown in supplemen al able S2 (a ail-
able online only a h p://www.a chi es-pm .o g/). The e e ence lis s
and bibliog aphies o he included s udies we e also sc eened.
S udy selec ion
All eco ds we e analyzed in he ee web e sion o Rayyan.
31,a
A e elimina ing duplica es, he i les and abs ac s o all eco ds
we e e iewed. Sc eening and selec ion we e pe o med by 2 inde-
penden e iewe s (R.NC. and A.E.) wi h a hi d au ho (J.C.) as e -
e ee in case o disag eemen . Subsequen ly, he same au ho s (R.NC.
and A.E.) independen ly e iewed he ull- ex a icles o eligibili y
acco ding o he p ede ined c i e ia. Any disc epancies we e esol ed
by consensus in consul a ion wi h a hi d au ho (J.C.).
Da a ex ac ion
Two au ho s (R.NC. and A.E.) independen ly ex ac ed da a using
a s anda dized ex ac ion o m. Co esponding au ho s we e con-
ac ed by e-mail i essen ial da a we e missing o unce ain ies
exis ed. The ollowing a iables we e collec ed o each s udy:
au ho , yea o publica ion, coun y, numbe o pa icipan s
en olled, sex (%), mean age, musculoskele al condi ion, ollow-up,
exposu e ac o , ou come, numbe o pa icipan s analyzed, com-
pa a i e measu e be ween g oups o measu e o associa ion (co e-
la ion coe icien , odds a io o haza d a io) and co esponding
measu e o dispe sion (s anda d e o , s anda d de ia ion, in e -
qua ile ange) o p ecision (95% con idence in e al [95% CI]).
Me hodological quali y assessmen
The Newcas le O awa scale was used o assess he quali y o he
included s udies.
32
Each s udy was independen ly assessed by 2
e iewe s (R.NC., A.E.) using a h ee-domain sco ing sys em: (1)
Selec ion (4 poin s); (2) Compa abili y (2 poin s); and (3) Exposu e/
ou come (3 poin s). I he e we e disc epancies o disag eemen s
be ween he e iewe s’ judgmen s, a hi d e iewe (J.C.) was con-
sul ed. The sum o poin s de e mined he me hodological quali y o
each s udy, anging om 0 (poo es quali y) o 9 (bes quali y) poin s.
Quan i a i e syn hesis
All analyses we e pe o med in R . 4.1.1.
b
Fo me a-analyses, o
a oid pe o ming a low-powe analysis, pooling o da a we e con-
side ed i he e we e a leas 3 o mo e s udies measu ing he same
p ognos ic ac o . In addi ion, s udies ha ope a ionalized he
exposu e ac o in a ma kedly di e en way om mos o he s ud-
ies we e excluded om he es ima ion. Fi s , he o iginal da a (eg,
odds a ios, co ela ions, eg ession coe icien s) we e con e ed
o Pea son’s using s anda d o mulas.
33
To main ain consis ency,
associa ions we e calcula ed in he same di ec ion. The da a we e
hen con e ed o Fishe ’s z using he scalc() unc ion o he R
package “me a o ” . 3.8-1.
34
In he nex s ep, using he ma()
unc ion o he same package, we apply a andom e ec s model o
syn hesize he quan i a i e esul s o he included s udies o each
o he co ela ional da a on exposu e ac o s (ca as ophic hink-
ing, kinesiophobia, ea -a oidance o sel -e icacy, symp oms o
anxie y o dep ession) and ou come (wo k absen eeism). This ype
o model was p e e ed because i akes in o accoun he he e oge-
nei y o he s udies and does no assume ha all s udies a e om a
single common popula ion ha we e es ed unde iden ical o
ai ly simila condi ions.
35
Fo he inal in e p e a ion, he esul o each me a-analysis was
again ans o med in o Pea son’s and he magni ude o he e ec
size was in e p e ed as small ( =0.1 o 0.29), mode a e ( =0.3 o
0.49) and la ge ( ≥0.5).
36
Following he ecommenda ions o he
Coch ane Handbook,
37
s a is ical he e ogenei y was classi ied as
negligible (I
2
=0%-40%), mode a e (I
2
=30%-60%), subs an ial
(I
2
=50%-90%) o conside able (I
2
=75%-100%). Finally, o es
plo s we e gene a ed o isualize he e ec size (and 95% CI) o
each included s udy and he calcula ed summa y e ec size. In
addi ion, a sensi i i y analysis was pe o med by including in he
me a-analyses only s udies o high quali y (ie, 7 o mo e poin s on
he Newcas le O awa scale) when he e we e a leas 3 s udies
ha me his condi ion o he a iable o in e es .
Resul s
S udy selec ion
A o al o 827 po en ially eligible s udies we e iden i ied by
sea ching da abases and e e ence lis s. A e elimina ing 328
Absen eeism in uppe ex emi y diso de s 783
www.a chi es-pm .o g
duplica e eco ds and 451 by sc eening publica ions by i le and
abs ac , 48 s udies we e po en ially eligible o inclusion and ull-
ex a icles we e e ie ed. Thi y epo s we e excluded based on
eligibili y c i e ia, ha is, w ong popula ion (n=12), w ong expo-
su e ac o (n=8), w ong ou come (n=4), w ong s udy design
(n=5), w ong publica ion ype (n= 1). Finally, 18 s udies we e
included in his sys ema ic e iew ( ig 1).
38-55
Cha ac e is ics o he s udies
The cha ac e is ics o he included s udies a e de ailed in
able 1.
38-55
The s udies we e conduc ed in he Uni ed S a es o
Ame ica (n=6), he Ne he lands (n=4), Sweden (n=2), Aus alia
(n=1), Ge many (n=1), Is ael (n=1), Taiwan (n=1), and in Mul i-
coun ies (n=1). The yea o publica ion anged om 1997
38
o
2020.
55
The sample size anged om 40
47
o 8587
55
en olled pa -
icipan s. Follow-up ime anged om 1 mon h
49
o 2 yea s.
43,51
Pa icipan s
In o al, 12,393 pa icipan s we e included among he included
s udies, wi h a mean age anging om 37.4§11.0 o 52.1§
8.8 yea s. O e all, he pe cen age o men pa icipan s anged om
26% o 100% (median=51%) and women pa icipan s anged om
0% o 74% (median=49%). Thi een s udies included pa icipan s
wi h musculoskele al condi ions in he hand/w is egion.
38-
40,42,43,45-50,53,55
Among hem, 6 s udies included pa ien s wi h
ca pal unnel synd ome.
38-40,42,46,50
5 s udies included pa ien s
wi h hand inju ies,
43,45,47,48,53
1 s udy included pa icipan s wi h
dis al adius ac u e,
55
and 1 s udy included pa icipan s wi h
Fig 1 S udy selec ion p ocess. Abb e ia ion: CENTRAL, Coch ane Cen al Regis e o Con olled T ials.
784 R. N
u~
nez-Co
es e al
www.a chi es-pm .o g
Table 1 Cha ac e is ics o he i ems included
Au ho Coun y N Sex (M/F) Age (y) MSDs (Condi ion) Follow-up P ognos ic Fac o Ou come
Ka z 1997
38
USA 135 M: 31%, F: 69% NR Ca pal unnel elease 6 mon hs Men al heal h Wo k absence a 6 mon hs
Ka z 1998
39
USA 220*M: 29%, F: 71% 43§11 Ca pal unnel synd ome 18 mon hs Men al heal h Wo k absence a 18 mon hs
Ka z 2005
40
USA 181 M: 42%, F: 58% 45.7§9.4 Ca pal unnel elease 12 mon hs Men al heal h
Sel -e icacy
Wo k absence a 12 mon hs
Kuijpe s 2006
41
The Ne he lands 350 M: 55%, F: 45% 45§11 Shoulde pain 6 mon hs Ca as ophizing
Kinesiophobia
Fea a oidance
Sick lea e
Tu ne 2007
42
USA 899 M: 38%, F: 62% 44.3§9.7 Ca pal unnel synd ome 1 yea Men al heal h
Ca as ophizing
Wo k disabili y (≥180)
Ops eegh 2009
43
The Ne he lands 91 M: 69%, F: 31% 43§11.5 Hand diso de s/inju ies 2 yea s Sel -e icacy RTW (≤10 weeks)
Ka els 2010
44
The Ne he lands 483 M: 33%, F: 67% 41.5§10.4 Uppe limb complain s 3 mon hs Ca as ophizing
Kinesiophobia
Sickness absence
Chen 2012
45
Taiwan 120 M: 77%, F: 23% 35.7 (17-57) Hand inju y NR Men al heal h Time o wo k
Cowan 2012
46
USA 66 M: 26%, F: 74% 49.7§11.3 Ca pal unnel elease 2-4 mon hs Dep essi e symp oms
Anxie y symp oms
Ca as ophizing
Time o RTW
Ramel 2013
47
Sweden 40 M: 80%, F: 20% 40§14.0 Hand inju y 12 mon hs Men al heal h RTW a 12 mon hs
Roesle 2013
48
Aus alia 192 M: 85%, F: 15% 35.1 (18-63) Hand inju y 12 weeks Sel -e icacy RTW wi hin 12 weeks
Bo 2014
49
USA 70 M: 64%, F: 36% 43§15 Finge ip inju ies 1 mon h Sel -e icacy dep essi e symp oms Days o wo k missed
Conzen 2016
50
Ge many 71 M: 37%, F: 63% 50.5 (40-60) Ca pal unnel elease 6 mon hs Dep essi e symp oms Du a ion o sick lea e
Feleus 2017
51
The Ne he lands 533 M: 47%, F: 53% 42 (18-64) Uppe limb complain s 2 yea s Ca as ophizing
Kinesiophobia
Sick lea e (>10 wo king days)
Gowd 2019
52
USA 89 M: 71%, F: 29% 52.1§8.8 Ro a o cu epai 12 mon hs Dep essi e symp oms Time o RTW
Ma om 2019
53
Is ael 178 M: 100%, F: 0% 37.4§11.0 Hand inju y 12 mon hs Sel -e icacy Time o RTW
Coggon 2020
54
Mul icoun y 8587 M: 34%, F: 66% 38.5§9.8 Shoulde pain 14 mon hs Fea a oidance Sickness absence
Egund 2020
55
Sweden 88 M: 100%, F: 0% 45 (21-64) Dis al adius ac u e 12 mon hs Men al heal h Weeks o sick lea e
NOTE. Da a a e p esen ed as mean §s anda d de ia ion o as median and ange (min-max).
Abb e ia ions: F, emale; M, male; MSDs, musculoskele al diso de s; RTW, e u n o wo k.
*Fo he na a i e and quan i a i e syn hesis, he non-ope a ed coho (n=64) was aken in o accoun .
www.a chi es-pm .o g
Absen eeism in uppe ex emi y diso de s 785
inge ip inju ies.
49
On he o he hand, 3 s udies included pa ici-
pan s wi h shoulde condi ions,
41,52,54
and 2 s udies included
pa ien s wi h complain s in di e en egions o he uppe
ex emi ies.
44,51
Me hodological quali y assessmen
O e all, he quali y o he included a icles was good, wi h a
median o 7 poin s ( ange: 6-9) on he Newcas le O awa scale
(supplemen al able S3; a ailable online only a h p://www.
a chi es-pm .o g/). In e ms o scope o selec ion, 16/18
(89%) o he s udies had ep esen a i e coho s, in 18/18
(100%) he unexposed coho was om he same communi y
as he exposed coho , in 18/18 (100%) exposu e ac o s we e
assessed wi h alid e idence, and 15/18 s udies (83%)
explici ly s a ed ha he ou come o in e es (ie, absence om
wo k) was no p esen a baseline. In e ms o compa abili y,
in 10/18 (56%) demog aphic ac o s we e con olled o and in
17/18 (94%) o he po en ial con ounde s we e con olled o .
In he exposu e/ou come domain, only 3 s udies (17%)
assessed ou come by eco d linkage, and he es did so by
sel - epo . In 17/18 (94%), he du a ion o ollow-up was ade-
qua e (ie, ≥3 mon hs) and in 12/18 (67%) he numbe o
losses o ollow-up was less han 20%.
Pain coping s a egies
The main esul s o he associa ion be ween pain coping s a e-
gies and ou comes a e p esen ed in able 2.Ele ens udies
e alua ed hese ac o s.
40-44,46,48,49,51,53,54
Mos s udies (9/11;
Table 2 Main esul s o he associa ion be ween each o he p ognos ic ac o s and wo k absen eeism
Au ho
P ognos ic Fac o
(n Analyzed) Resul s
Signi ican Resul
o Uni a ia e Analysis
Signi ican Resul o
Mul i a ia e Analysis
Newcas le
O awa Scale
Ka z 1997
38
SF-36 MC (n=135) OR=1.4 [95% CI: 1.1, 1.7] Yes Yes 7/9
Ka z 1998
39
SF-36 MC (n=64) OR=5.87 [95% CI 1.16, 29.83]*Yes - 7/9
Ka z 2005
40
SF-36 MC (n=143)
Sel -e icacy
y
(n=120)
P=.38
OR= 4.4 [95% CI: 1.4, 14.0]
No
Yes
No
Yes
8/9
Kuijpe s 2006
41
PCCL
z
(n=350)
TSK (n=350)
FABQ (n=350)
OR=1.6 [95% CI: 1.1, 2.2]
OR=1.7 [95% CI: 0.5, 5.3]
OR=1.1 [95% CI: 1.0, 1.1]
Yes
No
No
-
-
-
7/9
Tu ne 2007
42
SF-36 MC (n=899)
PCS (n=899)
OR=4.34 [95% CI: 2.69, 6.99]
OR=4.40 [95% CI: 2.55, 7.59]
Yes
Yes
Yes
Yes
8/9
Ops eegh 2009
43
GSES (n=91) E-RTW: 48.5 (IQR: 42.3-54.8) s
L-RTW: 48 (IQR 42-61)
No No 6/9
Ka els 2010
44
CSQ
x
(n=348)
TSK
ǁ
(n=348)
OR=2.8 [95% CI: 1.8, 4.5]
OR=2.1 [95% CI: 1.3.3.4]
Yes
Yes
- 7/9
Chen 2012
45
SF-36 MC (n=120) b=0.168, P<.05 - Yes 9/9
Cowan 2012
46
CES-D (n=66)
PASS (n=66)
PCS (n=34)
P=.480
P=.005; R
2
=0.03 (mul i a ia e)
P=.40; R
2
=0.15 (mul i a ia e)
No
Yes
Yes
No
Yes
Yes
6/9
Ramel 2013
47
SF-36 MC (n=40) RTW: 54 ( ange 19.9-58.6) s
No-RTW: 50 ( ange 19.5-63.2)
No - 6/9
Roesle 2013
48
GSES (n=192) E-RTW: 33.23§2.82 s
L-RTW: 33.24§3.84
No - 8/9
Bo 2014
49
PSEQ (n=56)
PHQ-9 (n=56)
=-0.52, P<.001
=0.54, P<.001
Yes
Yes
-
Yes
6/9
Conzen 2016
50
WHO-5 (n=42) HR=1.05 [95% CI: 0.85, 1.31] No - 6/9
Feleus 2017
51
CSQ
x
(n=533)
TSK
ǁ
(n=533)
OR=2.87 [95% CI: 1.48, 5.58]
OR=2.33 [95% CI: 1.22, 4.43]
Yes - 8/9
Gowd 2019
52
VRMCS (n=89) AUC=70.4% Yes - 6/9
Ma om 2019
53
Sel -e icacy
x
(n=178) HR=1.42 [95% CI:1.26, 1.66] Yes Yes 8/9
Coggon 2020
54
FAB (n=8386) OR=1.6 [95% CI: 1.1, 2.4] Yes - 6/9
Egund 2020
55
SF-36 (MC) (n=88) =0.03 [95% CI: -0.21, 0.27] No - 7/9
NOTE. Fo quan i a i e syn hesis, he o iginal da a (eg, odds a ios, co ela ions, eg ession coe icien s) we e con e ed o Pea son’s using s anda d
o mulae (Bo ens ein M, Hedges LV, Higgins JPT, Ro hs ein HR. In oduc ion o me a-analysis. 2021). Fo he s udies by Ramel 2013, Ops eegh 2009,
Roesle 2013, he con e sion was done om e ec size (dCohen).
Abb e ia ions: AUC, a ea unde he cu e; CES-D, Cen e o Epidemiologic S udies Dep ession Scale; CSQ, Coping S a egy Ques ionnai e; FABQ, Fea
A oidance Belie s Ques ionnai e; GSES, Gene al Sel -E icacy Scale; HR, haza d a io; IQR, in e qua ile ange; L-RTW, la e e u n o wo k; OR, odds a io;
PASS, Pain Anxie y Symp om Scale; PCCL, Pain Coping and Cogni ion Lis ; PCS, Pain Ca as ophizing Scale; PHQ-9; Pa ien Heal h Ques ionnai e; PSEQ,
Pain Sel -E icacy Ques ionnai e; SF-36 MC, 36-i em sho o m heal h su ey (Men al componen ); TSK, Tampa scale o kinesiophobia; VRMCS, Ve e ans
RAND Men al Componen Sco e; WHO-5, WHO-Fi e well-being index.
* alue es ima ed om he equency dis ibu ion (sco e ≤75) p esen ed o he non-ope a ed coho .
y
Assess using a 4-poin scale.
z
subdomains ca as ophizing, 1-6 poin s.
x
ca as ophizing: subscale 0-60.
ǁ
sho e sion wi hou he 4 e e sed i ems.
{
In he Ma oom 2019 s udy, a 5-poin like scale was used o assess sel -e icacy.
www.a chi es-pm .o g
786 R. N
u~
nez-Co
es e al
82%) epo ed a signi ican associa ion be ween a leas 1 o
hese ac o s and wo k absen eeism in uni a ia e o mul i a i-
a e analyses.
Ca as ophizing
Fi e s udies assessed ca as ophizing. Two s udies used he
Pain Ca as ophizing Scale,
42,46
2 s udies used he ca as ophiz-
ing subscale o he Coping S a egies Ques ionnai e,
44,51
and 1
s udy used he ca as ophizing subscale o he Pain Coping and
Cogni ion Lis .
41
Fou s udies p esen ed da a on he es ima es
be ween ca as ophizing and wo k absen eeism and we e
included in he me a-analysis (n=1631).
41,42,44,46
The o e all
esul o he andom-e ec s model was =0.28 (95% CI, 0.15
o 0.40, P<.0001) ( ig 2). He e ogenei y ac oss s udies was
conside able (I
2
=81.9%).
Sel -e icacy
Fi e s udies assessed sel -e icacy. Two s udies used he Gene al
Sel -E icacy Scale,
43,48
1 s udy used he Pain Sel -E icacy Ques-
ionnai e,
49
and 2 s udies used a Like - ype scale.
40,53
Fou s ud-
ies p esen ed da a on he es ima es be ween sel -e icacy and wo k
absen eeism and we e included in he me a-analysis (n=459). The
o e all esul o he andom-e ec s model was =0.24 (95% CI,
-0.02 o 0.47, P=.0747) ( ig 3). He e ogenei y ac oss s udies was
conside able (I
2
=86.9%).
Kinesiophobia
Th ee s udies assessed kinesiophobia using di e en e sions o
he Tampa Kinesiophobia Scale.
41,44,51
Two o he 3 s udies
showed a signi ican uni a ia e analysis esul o he associa ion
wi h wo k absen eeism.
44,51
Because Feleus e al
51
conduc ed a
Fig 2 Fo es plo o he associa ion o wo k absen eeism wi h ca as ophizing. Each s udy included in he me a-analysis co esponds o a poin
es ima e bounded by a 95% CI. The polygon a he bo om o he plo co esponds o he summa y e ec , and i s wid h ep esen s i s 95% CI.
S udies wi h la ge squa es ha e con ibu ed mo e o he summa y e ec size han o he s udies.
Fig 3 Fo es plo o he associa ion o wo k absen eeism wi h sel -e icacy. Each s udy included in he me a-analysis co esponds o a poin
es ima e bounded by a 95% CI. The polygon a he bo om o he plo co esponds o he summa y e ec , and i s wid h ep esen s i s 95% CI.
S udies wi h la ge squa es ha e con ibu ed mo e o he summa y e ec size han o he s udies.
www.a chi es-pm .o g
Absen eeism in uppe ex emi y diso de s 787
seconda y analysis o a coho p e iously included in he Ka els e
al s udy,
44
he e we e insu icien s udies on kinesiophobia o con-
duc a quan i a i e syn hesis.
Fea a oidance
Finally, 2 s udies p esen ed da a on es ima es be ween ea a oid-
ance (using he Fea A oidance Belie s Ques ionnai e).
41,54
One
o he 2 s udies showed a signi ican esul in he uni a ia e analy-
sis o he associa ion wi h wo k absen eeism.
54
The e we e insu -
icien s udies o his exposu e o pe o m a quan i a i e syn hesis.
Symp oms o anxie y and dep ession
The main esul s o he associa ion be ween men al heal h ac o s
and ou comes a e p esen ed in able 2. Ele en s udies e alua ed
he chosen men al heal h ac o s. Se en s udies used he men al
heal h componen o he SF-36 ques ionnai e,
38-40,42,45,47,55
while
1 s udy used he Ve e ans RAND Men al Componen Sco e.
52
On he o he hand, 2 s udies measu ed dep essi e symp oms
using he Pa ien Heal h Ques ionnai e and he WHO-Fi e well-
being index.
49,50
One s udy assessed bo h symp oms o anxie y
and dep ession wi h he Cen e o Epidemiologic S udies
Dep ession Scale and Pain Anxie y Symp om Scale, espec-
i ely.
46
Six y- h ee pe cen o he s udies (7/11) epo ed a sig-
ni ican associa ion wi h wo k absen eeism in uni a ia e o
mul i a ia e analysis. Fo he quan i a i e syn hesis, only s udies
ha assessed men al heal h using he same measu emen ins u-
men , ha is, he men al heal h componen o he SF-36 ques ion-
nai e, we e pooled.
38-40,42,45,47,55
The e we e insu icien s udies
speci ically assessing anxie y o dep essi e symp oms o pool
hese exposu es sepa a ely. Se en s udies p esen ed da a on he
es ima es be ween men al heal h and wo k absen eeism and we e
included in he me a-analysis (n=1748).
38-40,42,45,47,55
The o e -
all esul o he andom-e ec s model was =0.23 (95% CI, 0.10
o 0.34, P=.0003) ( ig 4). He e ogenei y ac oss s udies was sub-
s an ial (I
2
=74.1%).
Sensi i i y analysis
Sensi i i y analyses including only high-quali y s udies in me a-
analyses we e possible o he a iables o ca as ophizing and
symp oms o anxie y and dep ession. Bo h esul s showed consis-
ency wi h he main esul s. Fo ca as ophizing, he o e all esul
o he andom e ec s model was =0.27 (95% CI, 0.12 o 0.40,
P=.0005) (supplemen al ig S1; a ailable online only a h p://
www.a chi es-pm .o g/). Fo symp oms o anxie y and dep es-
sion, he o e all esul o he andom e ec s model was =0.25
(95% CI, 0.12 o 0.37, P=.0001) (supplemen al ig S2; a ailable
online only a h p://www.a chi es-pm .o g/).
Discussion
This sys ema ic e iew and me a-analysis o p ospec i e s udies
p o ides upda ed e idence on he associa ion o pain coping s a -
egies and symp oms o anxie y and dep ession wi h wo k absen-
eeism in people wi h uppe limb musculoskele al diso de s. Mos
s udies (14/18; 77.8%) epo ed a leas 1 signi ican associa ion
be ween pain coping s a egies o symp oms o anxie y and
dep ession and wo k absen eeism ( able 2). The me a-analyses
showed a small s a is ically signi ican co ela ion be ween ca a-
s ophizing and symp oms o anxie y and dep ession wi h wo k
absen eeism. The e o e, ou indings sugges ha ehabili a ion
eams could conside ca as ophizing and men al heal h ac o s
(symp oms o anxie y and dep ession) when designing p og ams
o acili a e E-RTW o people wi h uppe limb diso de s. How-
e e , hese esul s should be in e p e ed wi h cau ion, as he small
sample size, he small e ec sizes and he conside able he e oge-
nei y obse ed educe he le el o ce ain y o hese esul s.
Fig 4 Fo es plo o he associa ion o wo k absen eeism wi h symp oms o dep ession and anxie y. Each s udy included in he me a-analysis
co esponds o a poin es ima e bounded by a 95% CI. The polygon a he bo om o he plo co esponds o he summa y e ec , and i s wid h ep-
esen s i s 95% CI. S udies wi h la ge squa es ha e con ibu ed mo e o he summa y e ec size han o he s udies. NOTE. Ka z 1997
38
: Rec ui ed
July 1992-Oc obe 1993 (su gical coho , n=135); Ka z 1998
39
: Rec ui ed be ween July 1992 and Oc obe 1993 (non-su gical coho , n=64); Ka z
2005
40
: Rec ui ed be ween Ap il 1997 and Oc obe 1998.
www.a chi es-pm .o g
788 R. N
u~
nez-Co
es e al
Ou indings a e consis en wi h p e ious sys ema ic e iews
wi hou me a-analyses ha p o ide low ce ain y e idence ha
pain coping and men al heal h ac o s a e associa ed wi h E-RTW
in people wi h uppe ex emi y diso de s.
14,24,25
Howe e , ou
me a-analysis p o ides quan i a i e e idence o he e ec size o
hese p e iously na a i ely desc ibed associa ions, ha is, small
s a is ically signi ican co ela ion. Fo example, Bous ield e al
14
ecommended ha clinicians assess pa ien s’ psychological s a us
(eg, dep ession) o p edic RTW in people wi h elbow, w is , and
hand diso de s. Howe e , a ele an psychosocial ac o such as
ca as ophizing was no speci ically included in he sea ch s a egy
o hei s udy. The e o e, he au ho s we e unable o make ecom-
menda ions o his a iable. In ou me a-analysis, we ound a
small bu s a is ically signi ican associa ion be ween ca as ophiz-
ing and wo k absen eeism.
In con as o ou esul s, Black e al
23
in a e iew o he li e a-
u e, ound ha highe le els o sel -e icacy had a posi i e associ-
a ion wi h RTW in wo ke s wi h psychological o uppe -body
musculoskele al inju ies bu did no p o ide a quan i a i e syn he-
sis. Among he s udies included in ou me a-analysis, Bo e al
49
epo ed a mode a e co ela ion be ween days o wo k and sel -
e icacy as measu ed by he Pain Sel -E icacy Ques ionnai e.
Howe e , no associa ion was obse ed in he 2 s udies ha used
he Gene al Sel -E icacy Scale.
43,48
In his sense, a speci ic
assessmen o sel -e icacy as a coping s a egy in ela ion o pain
may be use ul o u u e compa isons. I is also wo h men ioning
ha , al hough no signi ican in he p esen me a-analysis, he size
o he co ela ion coe icien o sel -e icacy was qui e simila o
ha o he o he exposu e ac o s (ie, pain coping s a egy and anx-
ie y and dep essi e symp oms).
Wi h ega d o symp oms o anxie y and dep ession, i is
impo an o no e ha all s udies included in he me a-analysis
used he SF-36 Men al Heal h Componen , which has been used o
assess symp oms o anxie y and dep ession in a ious popula ions
and heal h condi ions.
56,57
A psychome ic s udy based on da a
om 35,908 ch onic pain pa ien s ound a high and signi ican co -
ela ion be ween he SF-36 men al heal h subscale and he Hospi-
al Anxie y and Dep ession Scale (HAD).
58
Simila ly, P oh e al
ound a s ong co ela ion be ween SF-36 men al heal h sco es
and HAD symp oms o dep ession and anxie y ( =-0.72 o -0.79)
in su i o s o acu e espi a o y ailu e in a c oss-sec ional analy-
sis o 1229 pa icipan s om he US, UK, and Aus alia.
59
In his
con ex , ou esul s p o ide a clea app oxima ion o he ela ion
be ween anxie y and dep essi e symp oms and wo k absence.
O he p e ious sys ema ic e iews o pa ien s wi h musculo-
skele al symp oms in o he egions o he body (eg, he spine) also
suppo ou indings. Fo example, Ziege e al ound ha symp-
oms o anxie y and dep ession had a signi ican e ec on pa ien s’
RTW a e disk su ge y.
60
We li e al also ound e idence ha
ea -a oidance belie s a e associa ed wi h poo ou comes in
pa ien s wi h low back pain, including RTW.
61
Acco ding o he
ea -a oidance model, people wi h nega i e coping s a egies
ela ed o pain (eg, ca as ophizing) a oid ce ain expe iences o
ac i i ies ha a e pe cei ed as h ea ening, de eloping disabili y,
and men al heal h p oblems,
18
which may also a ec pa icipa ion
in wo k.
62
The pain coping s a egies and men al heal h ac o s conside ed
may be modi iable by a ious psychosocial ea men app oaches,
such as pain educa ion, exe cise, cogni i e beha io al he apy, o
mind ulness.
26-28
The e o e, ea ly in e en ion s a egies ha ein-
o ce posi i e belie s, a i udes, and beha io s may be e ec i e in
modi ying he pain expe ience and hus acili a ing an E-RTW. Fo
example, educa ion and counseling on pain managemen and exe -
cise can educe he du a ion o wo k absen eeism in people wi h
ea -a oidance belie s and acu e low back pain.
63
Simila ly, mul i-
disciplina y in e en ions based on pa icipa o y e gonomics and
g aded ac i i ies based on cogni i e-beha io al p inciples can also
op imize sus ainable RTW in people wi h ch onic musculoskele al
pain.
64
Thus, add essing pain coping s a egies and men al heal h
in u u e s udies may be a c i ical in e en ion oppo uni y o
imp o e heal h ou comes in people wi h uppe limb musculoskele-
al symp oms and p e en long- e m wo k disabili y. In addi ion,
o people wi h ch onic pain o ha e a success ul and sa is ying
RTW, i is impo an o ocus on pain managemen , managing
wo k ela ions, and making wo kplace accommoda ions.
65
These
include using s a egies such as slowing down, aking con inuous
b eaks, wo king mo e slowly, and being awa e o wo kloads, as
well as changing wo king hou s and schedules, and inc easing
employe s’ unde s anding and awa eness o pain issues.
65
S eng hs and limi a ions
Among he s eng hs o his sys ema ic e iew, we highligh he
igo ous adhe ence o epo ing guidelines and he exhaus i e
sea ch o 5 da abases and addi ional sou ces o iden i y ele an
s udies. Mo eo e , o ou knowledge, ou me a-analysis is he i s
o examine he co ela ion be ween pain coping s a egies and
men al heal h and wo k absen eeism in people wi h uppe ex em-
i y diso de s in p ospec i e s udies. In con as , ou e iew has
se e al limi a ions ha should be conside ed o a cau ious in e -
p e a ion o he esul s: (1) Mos s udies assessed absen eeism by
sel - epo , and ew used eco d linkage, which se e ely comp o-
mised he me hodological quali y o he ou come measu es. To
add ess his poin , we conduc ed a sensi i i y analysis ha
included only s udies o high me hodological quali y in he me a-
analyses, which showed ha he di ec ion and magni ude o he
o e all e ec we e consis en wi h he main esul s; (2) G ay li e -
a u e (ie, s udies no indexed in he da abases e iewed) was no
sea ched. The e o e, publica ion bias should no be uled ou ; (3)
The po en ial o signi ican clinical he e ogenei y due o he ype
o musculoskele al condi ion and du a ion o ollow-up may esul
in some deg ee o measu emen bias. Fu he mo e, he s a is ical
he e ogenei y o he me a-analyses anged om subs an ial o con-
side able. Fo his eason, he esul s should be in e p e ed wi h
cau ion; (4) The lack o a ailable da a o some exposu e ac o s
o he use o associa ion measu es ha we e no possible o con-
e o Pea son’s using s anda dized o mulas (eg, haza d a io)
limi ed he possibili y o pe o ming a me a-analysis o o he
exposu e ac o s (eg, kinesiophobia, ea a oidance, and anxie y
symp oms o dep essi e symp oms sepa a ely). Consequen ly,
he e may be some deg ee o selec ion bias; (5) No speci ic ool
was used o assess isk o bias. Howe e , he Newcas le-O awa
scale allows he quali y o p ospec i e s udies o be assessed,
which is an impo an componen o a comp ehensi e me a-analy-
sis. In addi ion, he guidelines o epo ing Me a-analysis o
Obse a ional S udies in Epidemiology ecommend assessmen o
s udy quali y.
30
(6) P e ious s udies ha e used di e en de ini ions
o pain coping s a egies and men al heal h ac o s. This may in lu-
ence he esul s and also label pa ien s, which may no be condu-
ci e o he RTW p ocess. Finally, i was no possible o analyze
he co ela ions o men and women sepa a ely in o de o iden i y
sex o gende di e ences. P e ious esea ch has shown ha
women ha e lowe pain ole ance and highe pain in ensi y han
men when exposed o simila pain ul s imuli.
66,67
While he e is
www.a chi es-pm .o g
Absen eeism in uppe ex emi y diso de s 789
