G och d eis, Thomas e al.
A icle — Published Ve sion
Psychome ic e alua ion o he Ge man e sion o he
Reco e ing Quali y o Li e (ReQoL) measu es in pa ien s
wi h a ec i e diso de s
The Eu opean Jou nal o Heal h Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: G och d eis, Thomas e al. (2022) : Psychome ic e alua ion o he Ge man
e sion o he Reco e ing Quali y o Li e (ReQoL) measu es in pa ien s wi h a ec i e diso de s, The
Eu opean Jou nal o Heal h Economics, ISSN 1618-7601, Sp inge , Be lin, Heidelbe g, Vol. 24, Iss. 4,
pp. 499-512,
h ps://doi.o g/10.1007/s10198-022-01489-z
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/312826
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
1 3
The Eu opean Jou nal o Heal h Economics (2023) 24:499–512
h ps://doi.o g/10.1007/s10198-022-01489-z
ORIGINAL PAPER
Psychome ic e alua ion o  heGe man e sion o  heReco e ing
Quali y o Li e (ReQoL) measu es inpa ien s wi ha ec i e diso de s
ThomasG och d eis1 · Hans‑Helmu König1· AnjuDe ianeeKee ha u h2· Jü genGallina 3·
Alexande Konnopka1· Holge Schulz4· Ma inLambe 3· AnneKa ow3· Judi hDams1
Recei ed: 3 Ma ch 2022 / Accep ed: 7 June 2022 / Published online: 1 July 2022
© The Au ho (s) 2022
Abs ac
Backg ound The gene ic sel - epo ed Reco e ing Quali y o Li e (ReQoL) measu es we e de eloped o measu ing eco -
e y- ocused heal h- ela ed quali y o li e (H QoL) in pe sons wi h men al heal h condi ions. The aim o his s udy was o
assess he psychome ic p ope ies o he Ge man e sion o he ReQoL measu es in pa ien s wi h a ec i e diso de s in
Ge many.
Me hods Da a om a pa ien sub-sample in a andomized con olled ial ha e been used (N = 393). The in e nal consis ency
and he es – e es eliabili y o he ReQoL measu es we e assessed using C onbach’s Alpha and he in a-class co ela ion
coe icien (ICC). The concu en alidi y and he known-g oup alidi y o he ReQoL measu es we e assessed using Pea -
son’s Co ela ion coe icien and Cohen’s d. The esponsi eness was assessed using Glass' Δ and he s anda dized esponse
mean (SRM).
Resul s The eliabili y among he i ems o he ReQoL-20 was o e all excellen . The ICC o he ReQoL-20 was = 0.70,
indica ing mode a e es – e es eliabili y. The concu en alidi y o he ReQoL-20 wi h he clinical measu e PHQ-9 was
s ong wi h a co ela ion coe icien o = − 0.76. The known-g oup alidi y o he ReQoL-20 using PHQ-9 cu -o poin s was
la ge wi h an e ec size o d = 1.63. The ReQoL measu es we e sensi i e o ea men  esponse and emissiono symp oms
measu ed by he PHQ-9 wi h la ge e ec sizes/SRM.
Discussion The psychome ic p ope ies o he ReQoL measu es o he assessmen o pa ien s wi h a ec i e diso de s we e
o e all good. Wi h he ReQoL, alid and eliable measu es o he assessmen o eco e y- ocused H QoL o pe sons wi h
a ec i e diso de s a e a ailable in Ge man language.
Keywo ds Reliabili y· Validi y· Responsi eness· EQ-5D· SF6D· Heal h- ela ed quali y o li e· Mood diso de s·
Ge many
JEL Classi ica ion I10· I30
In oduc ion
The concep o heal h- ela ed quali y o li e (H QoL)
includes he aspec s o sel -pe cei ed well-being in ela ion
o disease o ea men [1]. Cu en ly, in economic e alua-
ions o men al heal h ca e o pa ien s wi h a ec i e diso -
de s in Ge many, H QoL is egula ly assessed using gene ic
pa ien epo ed-ou come measu es (PROMs; e.g., [2–4]),
wi h he EQ-5D [5, 6] and he Sho Fo m heal h su ey
(SF-12) being mos popula [7]. Gene ic PROMs ha e he
ad an age ha hey can be adminis e ed o a b oad ange
o diseases and ea men s, as hey co e many aspec s o
sel -pe cei ed well-being [1].
* Thomas G och d eis
.g och d [email protected]
1 Depa men o Heal h Economics andHeal h Se ices
Resea ch, Hambu g Cen e o Heal h Economics, Uni e si y
Medical Cen e Hambu g-Eppendo , Hambu g, Ge many
2 School o Heal h andRela ed Resea ch, The Uni e si y
o She ield, She ield, UK
3 Depa men o Psychia y andPsycho he apy, Uni e si y
Medical Cen e Hambu g-Eppendo , Hambu g, Ge many
4 Depa men o Medical Psychology, Uni e si y Medical
Cen e Hambu g-Eppendo , Hambu g, Ge many
500 T.G och d eis e al.
1 3
Gene ic H QoL measu es, howe e , ha e he disad-
an age ha hey a e no su icien ly sui able o ep esen
he eco e y p ocess o men al heal h condi ions, such as
a ec i e diso de s [8]. In cos -e ec i eness analyses in
Ge many, i was egula ly ecognized ha gene ic H QoL
measu es may no be esponsi e o changes in symp om
se e i y o sel -e icacy (e.g., [2, 9, 10]). In he ea men
o pe sons wi h men al heal h condi ions, i is also inc eas-
ingly ecognized ha hese pe sons wan o li e a ull li e
e en when symp oms a e p esen , and ha his can con-
ibu e o he men al eco e y p ocess [11, 12]. Thus, o
measu e he eco e y p ocess, an ins umen is needed
ha measu es he impac on quali y o li e o pe sons
wi h men al heal h condi ions a he han symp oms [11,
13, 14]. Fo ha eason, he Reco e ing Quali y o Li e
(ReQoL) measu es – ReQoL-10 and ReQoL-20 – ha e
been de eloped o assess H QoL o pe sons wi h men al
heal h condi ions and o cap u e he eco e y p ocess wi h
espec o leading a meaning ul li e, e en i symp oms o
he men al heal h condi ion a e p esen [11].
The ReQoL measu es we e de eloped on a b oad heo-
e ical basis and wi h he in ol emen o academics/psy-
chome icians, policymake s, clinicians and pa ien s [15].
The inal 10/20 i ems o he measu es we e selec ed a e
ac o analysis o da a om mo e han 6500 pa ien s and
i em esponse heo y models employed o in o m i em
selec ion. Recen ly, he ReQoL measu es o iginally de el-
oped in English ha e been ansla ed and linguis ically
alida ed in Ge man using es ablished me hodology [16].
ReQoL measu es a e also a ailable in o he languages,
such as Du ch [17], se en common Indian languages
[18–24] and adi ional Chinese [25].
The inal ReQoL measu es ep esen hose componen s
o eco e y by he se en domains: ac i i y, belonging and
ela ionships, choice, con ol and au onomy, hope, sel -
pe cep ion, well-being and physical heal h [15, 26]. Fu -
he mo e, he i ems o he ReQoL measu es a e o mu-
la ed nega i ely and posi i ely and he eby a e o decisi e
impo ance o measu ing bo h imp o ed and wo sened
H QoL as a esul o men al heal h p oblems [15, 26, 27].
Posi i ely and nega i ely wo ded i ems a e sco ed om
ze o o ou and ou o ze o, espec i ely [15]. By sum-
ming up he sco es o he i ems, an o e all sco e o he
ReQoL-10/ReQoL-20 can be calcula ed wi h ze o ep-
esen ing he poo es H QoL and 40/80 ep esen ing he
highes H QoL.
Fu he mo e, i is possible o calcula e quali y-adjus ed
li e-yea s om he ReQoL measu es o use in cos -u ili y
analyses. Fo his pu pose, p e e ence weigh s om he
gene al popula ion in he Uni ed Kingdom we e es ima ed
and he ReQoL-U ili y Index (UI) has been de eloped [28].
Un o una ely, p e e ence weigh s om he gene al popula-
ion in Ge many a e no a ailable a his ime.
So a , he psychome ic p ope ies o he ReQoL meas-
u es ha e been assessed in pa ien s wi h anxie y and dep es-
sion in he Uni ed Kingdom [29], in pa ien s wi h psychosis
in he Ne he lands [17] and Singapo e [30], in he gene al
popula ion in he Uni ed Kingdom [15] and Hong Kong [25]
as well as in a con enience sample in he Ne he lands [17].
Fo he pa ien popula ions, he ReQoL measu es showed
good in e nal consis ency and be e esponsi eness and
cons uc alidi y compa ed wi h he EQ-5D-5L in pa ien s
wi h dep ession, bu no in pa ien s wi h anxie y [29, 30].
Fu he mo e, he ReQoL-10 was eliable in a sample wi h
pa ien s wi h psychosis and he ReQoL measu es showed
good con e gen and known-g oup alidi y [17]. In a i s -
episode psychosis popula ion in Singapo e, ReQoL-10 was
ound o ha e good in e nal consis ency and adequa e con-
s uc alidi y [30].
To he bes o ou knowledge, he psychome ic p ope -
ies o he ReQoL measu es ha e no ye been assessed o
pa ien popula ions in Ge many. The e o e, he p ima y aim
o his s udy was o assess he psychome ic p ope ies o
he ReQoL-10 and ReQoL-20 o he assessmen o pa ien s
wi h bipola a ec i e diso de , majo dep ession and dys-
hymia in Ge many. The seconda y aim o his s udy was
o assess he alidi y o he ReQoL-10 and ReQoL-20 by
compa ison wi h clinical measu es and measu es o H QoL.
Ma e ials andme hods
Sample
Da a used o his s udy we e collec ed wi hin a andomized
con olled ial e alua ing an e idence-based, s epped
and coo dina ed ca e se ice model o men al diso de s
(RECOVER) [31]. Pa ien s we e ec ui ed in he egula
psychia ic ca e o he Uni e si y Medical Cen e Hambu g-
Eppendo , Ge many, om beginning o 2018 un il he end
o 2019. Pa ien s we e eligible o pa icipa ion i hey we e
a leas a he age o 16yea s and i hey we e diagnosed
wi h a leas one ele an men al diso de (among o he s,
e.g. schizoph enic spec um diso de s, bipola a ec i e dis-
o de , majo dep ession, anxie y diso de o pos - auma ic
s essdiso de ) acco ding o he In e na ional S a is ical
Classi ica ion o Diseases and Rela ed Heal h P oblems—
10 h Re ision, Ge man Modi ica ion (ICD-10) [31, 32].
Pa ien s we e excluded om he s udy i hey ul illed
he c i e ia o o ganic men al diso de s, addic ion diso de s
as main diagnosis, and/o mode a e o se e e men al e a -
da ion. Fu he mo e, pa ien s we e excluded i hey lacked
co ec able hea ing and/o ision impai men and/o i hey
we e wi h insu icien knowledge o Ge man [31].
Fo he cu en s udy, he pa ien sample o he
RECOVER ial was educed o pe sons wi h bipola
501
Psychome ic e alua ion o  heGe man e sion o  heReco e ing Quali y o Li e (ReQoL) measu es…
1 3
a ec i e diso de (ICD-10: F31), pe sons wi h majo dep es-
sion (ICD-10: F32.2) and pe sons wi h dys hymia (ICD-10:
F34.1). Da a was collec ed a baseline (T0), 6mon hs (T1)
and 12mon hs a e baseline (T2). The sample was u he
es ic ed o pe sons wi hou missing in o ma ion in he
ReQoL measu e a T0.
The ial was egis e ed p ospec i ely (NCT03459664),
e hics app o al was ob ained om he e hics commi ee o
he Hambu g Medical Associa ion (PV5672) and all pa -
icipan s o he ial p o ided w i en in o med consen . A
de ailed desc ip ion o he RECOVER ial can be ound
elsewhe e [31].
Measu es
The Ge man 20-i em e sion o he ReQoL measu es
(ReQoL-20) has been used o measu ing eco e y- ocused
H QoL [16]. The i s 10 i ems o he ReQoL-20 cons i u e
he ReQoL-10. The 20 i ems a e sco ed on a scale wi h i e
le els anging om ‘none o he ime’ o ‘mos o he ime’.
In due conside a ion o he posi i e and nega i e wo ding
o he i ems, he i ems sco es we e summed up o a o al
sco e anging om 0 o 80, wi h highe sco es indica ing a
highe eco e y- ocused H QoL [15]. The ReQoL-10 was
calcula ed om he i s 10 i ems o he ReQoL-20 wi h a
o al sco e anging om 0 o 40. While he ReQoL con ains
a physical heal h i em, his is no included in he sco e [28].
Symp om se e i y was assessed by he Ge man e sions
o he 9-ques ion dep ession scale o he Pa ien Heal h
Ques ionnai e (PHQ-9) o he measu emen o dep essi e
symp oms [33, 34] and he Al man Sel -Ra ing Mania Scale
(ASRM) o he assessmen o he p esence and se e i y
o manic o hypomanic symp oms [35]. The social, occu-
pa ional, and psychological unc ioning was assessed by
he Global Assessmen o Func ioning (GAF) scale [36],
and he se e i y o illness was assessed using he Clinical
Global Imp ession – Se e i y scale (CGI-S) [37]. H QoL
was measu ed using he indexes o he Ge man e sions o
he EQ-5D-5L [5, 6] based on Ge man p e e ence weigh s
[38] and he SF-12 (SF-6D) [7, 39] based on p e e ence
weigh s om he Uni ed Kingdom [40] as well as he isual
analogue scale o he EQ-5D-5L (EQ-VAS) [6]. Fu he -
mo e, he men al componen summa y sco e (MCS) and
physical componen summa y sco e (PCS) we e calcula ed
om he espec i e men al and physical dimensions o he
SF-12 [39]. Addi ional in o ma ion on he cons uc s and
sco es o he measu es used o he assessmen o symp om
se e i y and H QoL a e gi en in he Online Resou ce.
The sociodemog aphic a iables sel - epo ed age, sex,
ma i al s a us, mig a ion backg ound, school-lea ing quali-
ica ion, and educa ion we e used o desc ip ion o he sam-
ple. Fu he mo e, he numbe o como bid DSM-IV diag-
noses was collec ed and ca ego ized in o como bid clinical
diso de s (axis I diagnoses) and pe sonali y diso de s (axis
II diagnoses) [41].
S a is ical analysis
Obse a ions wi h missing in o ma ion we e emo ed om
he analyses by casewise dele ion. Sociodemog aphic cha -
ac e is ics o he sample we e analyzed using desc ip i e s a-
is ics, and he dis ibu ion o he ReQoL-10/ReQoL-20 was
assessed by his og ams o he indi idual i ems o he o al
s udy sample. No mali y o he dis ibu ion o he indi idual
i ems o he ReQoL measu es was analyzed using he Shap-
i o–Wilk es o no mal da a [42]. Reliabili y, alidi y, and
esponsi eness o he ReQoL measu es we e assessed. T0,
T1 and T2 da a was used o assess he es – e es eliabili y
and he sensi i i y o change o he ReQoL-10/ReQoL-20.
All o he analyses we e based on T0 da a.
The s uc u al alidi y o he ReQoL measu es was
assessed using con i ma o y ac o analysis (CFA) o con-
i m a co ela ed ai s model s uc u e comp ising o wo
dis inc elemen s o posi i ely and nega i ely wo ded i ems
[26]. Goodness o i was indica ed by he oo mean squa e
e o o app oxima ion (RMSEA), he compa a i e i index
(CFI), and he Tucke -Lewis index (TLI) wi h cu -o al-
ues o he RMSEA and he CFI/TLI o ≤ 0.08 and > 0.95,
espec i ely [26, 43, 44]. In e nal consis ency, he ex en o
which sco es o one measu e a e he same o epea ed meas-
u emen using di e en se s o i ems, o he ReQoL meas-
u es was assessed be ween all hal es o he ques ionnai es.
Fu he mo e, es – e es eliabili y, he ex en o which
sco es o one measu e a e he same o epea ed measu e-
men o e ime, o he ReQoL measu es was assessed o he
same measu es applied o e measu emen ime poin s. Fo
he assessmen o he in e nal consis ency, he ques ionnai es
we e spli in all possible hal es and he a e age co ela ions
o all hal es we e calcula ed using C onbach’s Alpha [45].
C onbach’s alpha coe icien s abo e α ≥ 0.7 we e de ined as
accep able, abo e α ≥ 0.8 as good and abo e α ≥ 0.9 as excel-
len [46]. Fo he assessmen o he es – e es eliabili y,
only hose pe sons wi hou any imp o emen o wo sening
o symp oms be ween measu emen poin s we e selec ed
om he sample. Unchanged symp oms be ween measu e-
men poin s we e de ined as PHQ-9 (ASRM) di e ence <
|5|
(
|4|
) be ween measu emen poin s [35, 47]. Tes – e es eli-
abili y was calcula ed using he in a-class co ela ion coe -
icien (ICC) [48]. ICC below = 0.50 we e de ined as poo ,
be ween = 0.50 and = 0.75 as mode a e, be ween = 0.75
and = 0.90 as good, and abo e = 0.90 as excellen [49].
The concu en alidi y, he ex en o how well one meas-
u e compa es o o he measu es, o he ReQoL measu es
was assessed by compa ison wi h clinical measu es (PHQ-9
and ASRM) and measu es o H QoL (EQ-5D-5L, EQ-VAS,
SF-6D, MCS and PCS). Fu he mo e, he known-g oup
502 T.G och d eis e al.
1 3
alidi y, he ex en o how well one measu e can demon-
s a e di e en sco es o di e en g oups, was assessed by
compa ison o sco es o g oups wi h less o mo e se e e
symp oms, and wi h good o poo global unc ioning and
lowe o highe se e i y o illness. The concu en alidi y
was assessed using Pea son’s Co ela ion Coe icien (PCC),
and by sca e plo s and LOWESS cu es o he espec i e
sco es. PCC below =
|0.30|
we e de ined as negligible,
be ween =
|0.30|
and =
|0.50|
as low, be ween =
|0.50|
and =
|0.70|
as mode a e, be ween =
|0.70|
and =
|0.90|
as high and abo e =
|0.90|
as e y high [50]. Fo
he assessmen o he known-g oup alidi y, he sample was
spli in o g oups wi h good and poo heal h using gene ic
measu es (CGI-S > 4, GAF ≤ 50 [36, 37]) and wi h less o
mo e se e e symp oms using clinical measu es (PHQ-9 > 4
and ASRM ≤ 50 [33–35]). The known-g oup alidi y was
assessed, in acco dance wi h he psychome ic e alua ion o
he o iginal ReQoL measu es [15], using he e ec size (ES)
Cohen’s d. ES o d = 0.20 we e de ined as small, o d = 0.50
as medium and o d = 0.80 as la ge [51, 52].
Responsi eness, he abili y o one measu e o de ec a
signi ican change assessed using a gold s anda d, o he
ReQoL measu es was assessed by sensi i i y o ea men
esponse ( educ ion o he PHQ-9 by ≥ 5 poin s) and emis-
sion o symp oms (PHQ-9 < 5) be ween ime poin s. Sensi-
i i y o change was assessed using he ES Glass' Δ and he
s anda dized esponse mean (SRM). Fu he mo e, ecei e
ope a ing cha ac e is ics (ROC) cu es we e cons uc ed and
disc imina i e abili ies we e assessed using he a ea unde
he cu e (AUC), wi h an AUC o 1.0 de ined as pe ec
disc imina i e abili ies and an AUC o 0.5 de ined as an-
dom chance. Op imal cu -o alues we e de e mined by he
dis ance o he op-le co ne om poin s on he ROC cu e.
Dis ance was de ined as d2 = (1 − sensi i i y)2 + (1 − speci ic-
i y)2. The eby, he poin wi h he lowes dis ance was de ined
as cu -o poin [53].
Da a analyses we e pe o med using S a a/MP 17.0
(S a aCo p, TX, USA). All s a is ics we e wo-sided wi h a
signi icance le el o p < 0.05.
Resul s
Sample cha ac e is ics
The mean age o he o al sample o pe sons wi h mood
diso de s (n = 393) was 39yea s (Table1). Mo e han
hal o he sample was emale (56%). The majo i y o he
sample was single (51%), had an uppe seconda y school
ce i ica e (56%) and had a oca ional aining deg ee
(45%). The mean numbe o como bid DSM-IV axis I
diagnoses and axis II diagnoses o he sample was 0.74
and 0.11, espec i ely. The comple e sociodemog aphic
cha ac e is ics o he o al sample and he sub-samples a e
shown in Table1.
Dis ibu ion o sco es
The mean sco es o he ReQoL-20 and ReQoL-10 in he
o al sample we e 33.07 (SD 13.31) and 16.61 (SD 6.98) a
T0, wi h a ange om 8 o 76 and 3 o 39, espec i ely. The
sco es o bo h he ReQoL-20 and ReQoL-10 we e sligh ly
igh skewed and lep oku ic wi h no appa en loo and ceil-
ing e ec s (da a no shown). A T1, he mean sco es we e
44.17 (SD 16.75) and 22.35 (SD 8.82) and a T2, he mean
sco es we e 45.52 (SD 16.36) and 23.01 (SD 8.71). In he
sub-samples o pe sons wi h bipola a ec i e diso de and
o pe sons wi h majo dep ession and dys hymia, he mean
sco es o he ReQoL-20 we e 42.08 (SD 16.64) and 32.19
(SD 12.61) a T0, 48.00 (SD 15.77) and 43.71 (SD 16.86) a
T1, and 48.93 (SD16.12) and 45.13 (SD 16.37) a T2, espec-
i ely. The espec i e mean sco es o he ReQoL-10 we e
21.44 (8.63) and 16.13 (6.61) a T0, 24.40 (8.17) and 22.11
(8.88) a T1, and 25.20 (SD 8.23) and 22.77 (SD 8.74) a T2.
The majo i y o he indi idual i ems o he ReQoL meas-
u es a T0 we e no no mally dis ibu ed (all wi h p < 0.05),
wi h he excep ion o he i ems 2 (‘I el able o us o he s’),
3 (‘I el unable o cope’), 6 (‘I hough my li e was no wo h
li ing’), 10 (‘I el con iden in mysel ’) and 13 (‘I el i i-
a ed’; Fig.1). The h ee i ems o he ReQoL measu es wi h
he mos posi i ely skewed dis ibu ion we e he i ems 5 (‘I
el happy’), 15 (‘I el in con ol o my li e’) and 18 (‘I had
p oblems wi h my sleep’). Goodness o i o he co ela ed
ai s model s uc u e o he ReQoL-20 (ReQoL-10) was
con i med wi h a RMSEA o 0.08 (0.09), ye wi h a CFI o
0.87 (0.92) and a TLI o 0.85 (0.89).
Reliabili y
C onbach’s’ alpha o he ReQoL-20 (ReQoL-10) in he o al
sample wi h mood diso de s was α = 0.91 (α = 0.83), indica -
ing excellen (good) eliabili y among he i ems. C onbach’s
alpha o he ReQoL-20 (ReQoL-10) in he sub-samples o
pe sons wi h bipola a ec i e diso de , ando pe sons wi h
majo dep ession and dys hymia was α = 0.94 (α = 0.87) and
α = 0.89 (α = 0.81), espec i ely, indica ing excellen (good)
and good eliabili y among he i ems.
The ICC o he ReQoL-20 (ReQoL-10) in he o al sam-
ple wi h mood diso de s wi hou imp o emen o wo sening
o symp oms om T0 o T1 and T1 o T2 measu ed by he
PHQ-9 was = 0.70 ( = 0.68) and = 0.76 ( = 0.75), indi-
ca ing mode a e and good es – e es eliabili y, espec i ely
(Table2).

503
Psychome ic e alua ion o  heGe man e sion o  heReco e ing Quali y o Li e (ReQoL) measu es…
1 3
Validi y
The co ela ion coe icien be ween he ReQoL-20 and he
ReQoL-10 was = 0.94. In he sub-samples o pe sons wi h
bipola a ec i e diso de and o pe sons wi h majo dep es-
sion and dys hymia, he co ela ion coe icien was = 0.96
and = 0.93, espec i ely.
The concu en alidi y o he ReQoL-20 and ReQoL-10
wi h he clinical measu e PHQ-9 was s ong and mode -
a e, indica ed by a co ela ion coe icien o = − 0.76 and
= − 0.69, espec i ely (Table3, Figu esS1 and S2 in he
Online Resou ce). In he sub-samples o pe sons wi h bipo-
la a ec i e diso de and o pe sons wi h majo dep ession
and dys hymia, he co ela ion coe icien s be ween he
ReQoL-20 (ReQoL-10) and he PHQ-9 we e = − 0.80
( = − 0.70) and = − 0.75 ( = − 0.67), indica ing an o e all
s ong nega i e linea ela ionship.
The concu en alidi y o he ReQoL-20 and ReQoL-
10 wi h measu es o H QoL was o e all mode a e, wi h
co ela ion coe icien s anging om = 0.55 o = 0.63
(Table3, Figu esS3 and S4 in he Online Resou ce).
The only excep ion was he concu en alidi y o he
ReQoL-20 and ReQoL-10 wi h he PCS, which was
only weak wi h a co ela ion coe icien o = 0.29. All
obse ed co ela ions we e in he expec ed di ec ions.
The co ela ion coe icien s be ween u he measu es o
H QoL and clinical measu es a e gi en in TableS4 in he
Online Resou ce.
The mean sco es o he ReQoL-20 (ReQoL-10) in he
sample wi h minimal o mode a e dep ession and mode -
a ely se e e dep ession o se e e dep ession measu ed by
he PHQ-9 we e 42.67 (21.16) and 25.70 (13.13), espec-
i ely (TableS2 in he Online Resou ce). The co espond-
ing known-g oup alidi y o he ReQoL-20 and ReQoL-10
using PHQ-9 cu -o poin s was la ge wi h ES o d = 1.63
and d = 1.39, espec i ely (Table4).
The mean sco es o he ReQoL-20 (ReQoL-10) in he
sample wi h good o poo global unc ioning dicho omized
by he GAF we e 35.10 (17.68) and 30.35 (15.16), espec-
i ely (TableS2 in he Online Resou ce). When he se e i y
o illness was dicho omized in o lowe o highe by he CGI-
S, he mean sco es o he ReQoL-20 (ReQoL-10) we e 34.78
(17.57) and 30.07 (14.92). The co esponding known-g oup
alidi y o he ReQoL-20 and ReQoL-10 using CGI-S and
Table 1 Sociodemog aphic cha ac e is ics o pe sons wi h mood diso de s (F30-F39; n = 393)
SD s anda d de ia ion
a Pe son had an own mig a ion expe ience o was bo n o a leas one pa en wi h own mig a ion expe ience
b One pe son had an unspeci ied mood diso de (F39)
Sociodemog aphic cha ac e is ic All Pe sons wi h mood
diso de s (F30-F39;
n = 393)b
Pe sons wi h bipola
a ec i e diso de (F31;
n = 36)
Pe sons wi h majo
dep ession (F32.2;
n = 350)
Pe sons wi h
dys hymia (F34.1;
n = 6)
Age: Mean (SD) 39.18 (14.93) 40.03 (13.95) 39.07 (15.00) 38.83 (19.92)
Female sex: n (%) 220 (55.98) 22 (61.11) 194 (55.43) 3 (50.00)
Wi h mig a ion backg ounda: n (%) 119 (30.43) 15 (41.67) 101 (29.02) 2 (33.33)
Ma i al s a us: n (%)
Single 200 (50.89) 23 (63.89) 172 (49.14) 5 (83.33)
Ma ied 60 (15.27) 6 (16.67) 54 (15.43) –
In pa ne ship 100 (25.45) 6 (16.67) 93 (36.57) –
Di o ced/widowed 33 (8.40) 1 (2.78) 31 (8.86) 1 (16.67)
School-lea ing quali ica ion: n (%)
No school-lea ing quali ica ion 9 (2.29) – 9 (2.57) –
Lowe seconda y school ce i ica e 160 (40.71) 5 (13.89) 150 (42.86) 4 (66.67)
Uppe seconda y school ce i ica e 219 (55.73) 31 (86.11) 186 (53.14) 2 (33.33)
O he school-lea ing quali ica ion 5 (1.27) – 5 (1.43) –
Educa ion: n (%)
No comple ed educa ion 111 (28.46) 11 (30.56) 97 (27.95) 2 (33.33)
Voca ional aining deg ee 175 (44.87) 9 (25.00) 163 (46.97) 3 (50.00)
Uni e si y deg ee/mas e c a sman's
examina ion
104 (26.67) 19 (44.44) 87 (25.07) 1 (16.67)
Numbe o como bid DSM-IV diagnoses:
mean (SD)
Axis I diagnoses 0.74 (0.95) 0.67 (0.96) 0.74 (0.95) 0.17 (0.52)
Axis II diagnoses 0.11 (0.37) 0.14 (0.36) 0.11 (0.37) 0.33 (0.41)
504 T.G och d eis e al.
1 3
GAF cu -o poin s was small wi h ES anging om d = 0.36
o d = 0.38.
Responsi eness
The ReQoL measu es we e sensi i e o ea men esponse
measu ed by he PHQ-9. The ES/SRM o he ReQoL-20
and ReQoL-10 anged be ween 1.20/1.40 and 2.02/1.73
be ween all measu emen poin s, indica ing high esponsi e-
ness (Table5, TableS3 in he Online Resou ce). The mean
di e ence o he ReQoL-20 (ReQoL-10) in he sample wi h
ea men esponse was 22.11 (10.70) be ween T0 and T1,
and 16.58 (8.35) be ween T1 and T2. Mean di e ences and
ES/SRM o measu es o H QoL based on ea men esponse
measu ed by he PHQ-9 a e gi en in TableS4 in he Online
Resou ce.
ROC analyses showed AUC o ReQoL-20 and ReQoL-
10 di e ences and ea men esponse measu ed by he
PHQ-9 anging om 0.87 o 0.89 and om 0.84 o 0.85,
espec i ely (Fig.2, Figu e S3 and TableS5 in he Online
Resou ce). The op imal cu -o alue o ea men esponse
was a ReQoL-20 (ReQoL-10) di e ence ≥ 12 (≥ 6) wi h a
sensi i i y o 78.03% (75.76%) and a speci ici y o 85.26%
(80.77%).
Fo emission o symp oms measu ed by he PHQ-9, he
ReQoL measu es we e also sensi i e wi h la ge ES/SRM o
he ReQoL-20 and ReQoL-10 be ween T0 and T1 (1.64/1.49
and 1.64/1.51) and wi h mode a e ES/SRM be ween T1 and
T2 (0.45/0.56 and 0.47/0.53; Table6, TableS3 in he Online
Resou ce). The mean di e ence o he ReQoL-20 (ReQoL-
10) in he sample wi h emission o symp oms was 24.65
(12.13) be ween T0 and T1, and 5.80 (3.12) be ween T1 and
T2. Mean di e ences and ES/SRM o measu es o H QoL
Fig. 1 Dis ibu ion o he indi idual i ems o he ReQoL measu es a
T0 (n = 393). 1 None o he ime, 2 only occasionally, 3 some imes, 4
o en, 5 mos o all o he ime, i em 1 I ound i di icul o ge s a ed
wi h e e yday asks, i em 2 I el able o us o he s, i em 3 I el una-
ble o cope, i em 4 I could do he hings I wan ed o do, i em 5 I el
happy, i em 6 I hough my li e was no wo h li ing, i em 7 I enjoyed
wha I did, i em 8 I el hope ul abou my u u e, i em 9 I el lonely,
i em 10 I el con iden in mysel , i em 11 I did hings I ound ewa d-
ing, i em 12 I a oided hings I needed o do, i em 13 I el i i a ed,
i em 14 I el like a ailu e, i em 15 I el in con ol o my li e, i em 16
I el e i ied, i em 17 I el anxious, i em 18 I had p oblems wi h my
sleep, i em 19 I el calm, i em 20 I ound i ha d o concen a e
505
Psychome ic e alua ion o  heGe man e sion o  heReco e ing Quali y o Li e (ReQoL) measu es…
1 3
Table 2 Tes –Re es Reliabili y o he ReQoL-10 and ReQoL-20 o pe sons wi hou imp o emen o wo sening o symp oms be ween measu emen poin s o pe sons wi h mood diso de s
ICC In a-class co ela ion coe icien
a PHQ-9: missing imp o emen o wo sening o symp oms be ween measu emen poin s was de ined as − 5 < PHQ-9 di e ence < 5
b ASRM: missing imp o emen o wo sening o symp oms be ween measu emen poin s was de ined as − 4 < ASRM di e ence < 4
c One pe son had an unspeci ied mood diso de (F39)
All pe sons wi h mood diso de s (F30-
F39)c
Pe sons wi h bipola a ec i e diso de
(F31)
Pe sons wi h majo dep es-
sion and dys hymia (F32.2
and F34.1)
N (%) ICC N (%) ICC N (%) ICC
T0 o T1 ReQoL-10 PHQ-9a (n = 282) 139 (49.29) 0.68 13 (48.15) 0.59 126 (49.61) 0.68
ASRMb (n = 31) – 19 (61.29) 0.45 – –
ReQoL-20 PHQ-9a (n = 282) 139 (49.29) 0.75 13 (48.15) 0.72 126 (49.61) 0.75
ASRMb (n = 31) – – 19 (61.28) 0.43 – –
T1 o T2 ReQoL-10 PHQ-9a (n = 252) 191 (75.79) 0.84 22 (78.57) 0.78 169 (75.45) 0.85
ASRMb (n = 28) – – 18 (64.29) 0.81 – –
ReQoL-20 PHQ-9a (n = 252) 191 (75.79) 0.86 22 (78.57) 0.78 169 (75.45) 0.87
ASRMb (n = 28) – – 18 (64.29) 0.78 – –
T0 o T2ReQoL-10 PHQ-9a (n = 284) 130 (45.78) 0.70 13 (44.83) 0.74 117 (45.88) 0.68
ASRMb (n = 29) – – 16 (55.17) 0.47 – –
ReQoL-20 PHQ-9a (n = 284) 130 (45.78) 0.76 13 (44.83) 0.82 117 (45.88) 0.75
ASRMb (n = 29) – – 16 (55.17) 0.39 – –
506 T.G och d eis e al.
1 3
Table 3 Concu en alidi y be ween ReQoL-10/ReQoL-20, clinical measu es (PHQ-9, ASRM) and measu es o heal h- ela ed quali y o li e
(EQ-5D-5L, EQ-VAS, SF-6D, MCS, PCS) o pe sons wi h mood diso de s
All co ela ion coe icien s we e s a is ically signi ican wi h p ≤ 0.001
PCC Pea son’s Co ela ion Coe icien
a One pe son had an unspeci ied mood diso de (F39)
PCC All pe sons wi h mood diso -
de s (F30-F39; n = 393a)
Pe sons wi h bipola a ec i e
diso de (F31; n = 36)
Pe sons wi h majo dep ession
and dys hymia (F32.2 and F34.1;
n = 356)
ReQoL-10 PHQ-9 (n = 383) − 0.69 − 0.70 − 0.67
ASRM (n = 36) – 0.53 –
EQ-5D-5L (n = 384) 0.55 0.55 0.54
EQ-5D-5L domain
anxie y/dep ession
(n = 387)
− 0.64 − 0.64 − 0.62
EQ-VAS (n = 391) 0.56 0.76 0.53
SF-6D (n = 358) 0.62 0.74 0.58
MCS (n = 349) 0.63 0.78 0.57
PCS (n = 349) 0.29 0.44 0.29
ReQoL-20 PHQ-9 (n = 383) − 0.76 − 0.80 − 0.75
ASRM (n = 36) – 0.52 –
EQ-5D-5L (n = 384) 0.57 0.53 0.56
EQ-5D-5L domain
anxie y/dep ession
(n = 387)
− 0.66 − 0.61 − 0.65
EQ-VAS (n = 391) 0.56 0.72 0.53
SF-6D (n = 358) 0.63 0.70 0.60
MCS (n = 349) 0.63 0.69 0.60
PCS (n = 349) 0.29 0.46 0.29
Table 4 Known-g oup alidi y o ReQoL-10/ReQoL-20 using cu -o poin s o clinical measu es (PHQ-9a, ASRMb) as well as gene ic measu es
(CGI-Sc, GAFd)
a PHQ-9 was dicho omized in o minimal o mode a e dep ession (PHQ ≤ 14) s. mode a ely se e e dep ession o se e e dep ession (PHQ-9 > 14)
b ASRM was dicho omized in o low p obabili y o a manic o hypomanic condi ion (ASRM < 6) s. high p obabili y o a manic o hypomanic
condi ion (ASRM ≥ 6)
c The se en o iginal ca ego ies o he CGI-S we e dicho omized in o less se e e (1 ≤ CGI-S ≤ 4) s. mo e se e e illness (4 > CGI-S ≤ 7)
d The GAF was dicho omized in o good (GAF > 50) s. poo global unc ioning (GAF ≤ 50)
e One pe son had an unspeci ied mood diso de (F39)
Cohen’s D All pe sons wi h mood diso -
de s (F30-F39; n = 393e)
Pe sons wi h bipola a ec i e
diso de (F31; n = 36)
Pe sons wi h majo dep ession
and dys hymia (F32.2 and F34.1;
n = 356)
ReQoL-10 PHQ-9 (n = 383) 1.39 1.65 1.34
ASRM (n = 36) – − 1.09 –
CGI-S (n = 393) 0.37 0.14 0.63
GAF (n = 393) 0.38 0.31 0.56
ReQoL-20 PHQ-9 (n = 383) 1.63 1.97 1.58
ASRM (n = 36) – − 1.11 –
CGI-S (n = 393) 0.36 0.09 0.59
GAF (n = 393) 0.36 0.19 0.56