STUDY PROTOCOL Open Access
Clinical Decision Making and Ou come in
Rou ine Ca e o People wi h Se e e Men al
Illness (CEDAR): S udy p o ocol
Be nd Puschne
1*
, Sabine S e en
1
, Mike Slade
2
, Helena Kaliniecka
2
, Ma io Maj
3
, And ea Fio illo
3
,
Po l Munk-Jø gensen
4
, Jens I a La sen
4
, Anikó Ége házi
5
, Zol an Nemes
5
, Wul Rössle
6
, Wol am Kawohl
6
,
Thomas Becke
1
Abs ac
Backg ound: A conside able amoun o esea ch has been conduc ed on clinical decision making (CDM) in sho -
e m physical condi ions. Howe e , he e is a lack o knowledge on CDM and i s ou come in long- e m illnesses,
especially in ca e o people wi h se e e men al illness.
Me hods/Design: The s udy en i led “Clinical decision making and ou come in ou ine ca e o people wi h se e e
men al illness”(CEDAR) is ca ied ou in six Eu opean coun ies (Denma k, Ge many, Hunga y, I aly, Swi ze land and
UK). Fi s , CEDAR es ablishes a me hodology o assess CDM in people wi h se e e men al illness. Speci ic
ins umen s a e de eloped (and psychome ic p ope ies es ablished) o measu e CDM s yle, key elemen s o CDM
in ou ine ca e, as well as CDM in ol emen and sa is ac ion om pa ien and he apis pe spec i es. Second, hese
ins umen s a e being pu o use in a mul i-na ional p ospec i e obse a ional s udy (bimon hly assessmen s
du ing a one-yea obse a ion pe iod; N = 560). This s udy in es iga es he immedia e, sho - and long- e m e ec
o CDM on c ucial dimensions o clinical ou come (symp om le el, quali y o li e, needs) by aking in o accoun
signi ican a iables mode a ing he ela ionship be ween CDM and ou come.
Discussion: The esul s o his s udy will make possible o delinea e quali y indica o s o CDM, as well as o speci y
p ime a eas o u he imp o emen . Ing edien s o bes p ac ice in CDM in he ou ine ca e o people wi h
se e e men al illness will be ex ac ed and ecommenda ions o mula ed. Wi h i s explici ocus on he pa ien ole
in CDM, CEDAR will also con ibu e o s eng hening he se ice use pe spec i e. This p ojec will subs an ially add
o imp o ing he p ac ice o CDM in men al heal h ca e ac oss Eu ope.
T ial egis e : ISRCTN75841675.
Backg ound
Se e e men al illness (SMI) subs an ially con ibu es o
disabili y and global bu deno disease.In hegene al
popula ion in Eu ope o adul age, he p e alence a e
(12 mon hs) o se e e men al illness is abou 2.2% [1]
indica ing ha abou 11 million people in he Eu opean
Union a e a ec ed by clinically and socially disabling
condi ions wi h a high need o in ensi e and long- e m
p o essional ea men .
While people wi h SMI in Eu ope ecei e p o essional
heal h ca e in di e en ea men se ings wi h he
majo i y being ca ed o by communi y-based se ices,
he e is a lack o knowledge on clinical decision making
(CDM) and i s ou come in ou ine ca e. This is espe-
cially dis u bing since du ing he las decades, men al
heal h esea ch has esul ed in a la ge numbe o in e -
en ions wi h p o en e icacy whose implemen a ion
equi es communica ion be ween pa ien and clinician
as well as ac ions ollowing hei in e ac ions. I is
unknownwhe he , owhichex en ,andhowposi i e
pa ien ou come ollowing such in e en ions depends
upon pa ien -clinician in e ac ion o CDM. We a gue
* Co espondence: be nd.puschne @bkh-guenzbu g.de
1
Depa men o Psychia y and Psycho he apy II, Ulm Uni e si y, Ludwig-
Heilmeye -S . 2, 89312 Günzbu g, Ge many
Full lis o au ho in o ma ion is a ailable a he end o he a icle
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
© 2010 Puschne e al; licensee BioMed Cen al L d. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e
Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and
ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
ha he majo eason o his lack o knowledge is ha
esea ch on CDM in heal h ca e has p ima ily ocused
upon well-de ined si ua ions in physical condi ions,
while he e a e only e y ew s udies on CDM in ou ine
ca e o people wi h SMI wi h i s high demands on
pa ien engagemen , ensu ing con inui y o ca e and
es ablishing s able he apeu ic ela ionships.
Con ex ualising clinical decision making
Gi en a gene al inc ease in in e es in pa ien -cen e ed
[2] o pa ien - ocused [3] app oaches, i has been sug-
ges ed ha esea ch on clinical decision making should
become a p io i y. E.g. he NIMH B idging Science and
P ac ice Repo [4] speci ically ecommended o encou-
age “ he de elopmen o me hods o s udy and inco po-
a e clinician and pa ien /consume decision making
p ocesses in o in e en ion esea ch”( ecommenda ion
#24), as well as “ he imp o emen o me hods o bo h
e alua ing clinician implemen a ion and pa ien /consu-
me adhe ence o ea men ecommenda ions and es i-
ma ing he consequences o hese a ia ionson he
e ec i eness o ea men ”( ecommenda ion #26).
Since he 1960s, concep s de eloped wi hin he ame-
wo k o decision heo y ha e been applied o heal h
ca e esea ch. Resea ch on CDM has d awn upon se -
e al concep ual app oaches such as in o ma ion p oces-
sing, social judgemen heo y, and expec ed u ili y
heo y. Un il he 1980s, his esea ch almos solely
ocused on clinician decision making [5]. The e is con-
side able deba e on wha cons i u es a “good”clinical
decision. In a well-de ined one- ime clinical decision
scena io, an ideal decision is concep ualised as being he
sole esponsibili y o he physician who decides ia a
a ional p ocess aking in o accoun scien i ic e idence
and clinical expe ience. In his scena io i is assumed
ha pa ien s would make he same decision i p o ided
wi h he same in o ma ion [6]. In o ma ion would be
communica ed o he pa ien , bu he con ex o deci-
sion making is a he unimpo an in such a scena io
[7]. CDM iewed his way is a a ional and linea p o-
cess lending i sel eadily o sys ema ic analysis [8].
Howe e , since CDM a ely akes place in such clea -
cu si ua ions, less decon ex ualised models o decision
making ha e been de eloped.
En wis le e al. [9] p oposed a con ex ualised sequence
o ac i i ies in decision making consis ing o : (a) ecog-
ni ion and cla i ica ion o a p oblem; (b) iden i ica ion
o po en ial solu ions; (c) app aisal o po en ial solu ions;
(d) selec ion o cou se o ac ion; (e) implemen a ion o
he chosen cou se o ac ion; and ( ) e alua ion o he
solu ion adop ed. Simila ly, Ro he e al. [10] de eloped
a gene al amewo k (see Figu e 1) which concep ualises
he decision making p ocess om he pa ien ’s pe spec-
i e, and shows how - ia he p ocess o decision
making - indi idual-le el a iables migh be ela ed o
indi idual- and se ice-le el ou comes.
This amewo k sugges s ha e ec i e decision mak-
ing depends on accu a e in o ma ion ega ding he isks
and bene i s as well as he likelihoods o ele an ou -
comes and an unde s anding o he alues ele an o
he decision. Decisions a e made a e p e e ences ha e
been o mula ed by combining in o ma ion and alues
and in u n a ec pa ien beha iou s (e.g. ea men
adhe ence) and ou comes. In o ma ion e e s o deci-
sion- ele an da a, e.g. isks and bene i s associa ed wi h
a gi en ea men , which should be easily accessible and
comp ehensible. Values ep esen he indi idual a ac-
i eness o heal h s a es aking in o accoun a gi en
ea men ’s nega i e aspec s, e.g. unwan ed side e ec s.
The con ex o decision making e e s o a ious aspec s
o a pe son’s e e yday li e including s uc u e and acces-
sibili y o he heal hca e sys em as well as na u e, du a-
ion and se e i y o he illness. P e e ences indica e
g ea e liking o one ea men op ion compa ed o
ano he and a e concep ualised as an in e ac ion
be ween in o ma ion and alues. Ou comes as he esul
o pa ien beha iou include pa ien heal h s a us o
cos s o ca e [5].
Types o clinical decision making
Fu he mo e, inc easing a en ion has been gi en o
pa ien in ol emen in CDM. Cha les e al. [11] p o-
posed h ee gene al ypes o ea men decision making
(see Table 1) in which h ee main ac i i ies (in o ma ion
ans e , delibe a ion, and deciding abou implemen ing
ea men ) a e being conside ed. Analogue ypes aking
in o accoun desi e o in o ma ion and o ea men
choice ha e been sugges ed [12]: (a) P o essional choice:
The clinician decides and he pa ien consen s; (b)
Sha ed decision making: In o ma ion is sha ed and bo h
decide oge he ; (c) Consume choice: The clinician
in o ms and he pa ien makes he decision.
Coul e [12] a gues ha di e en models may be
app op ia e a di e en imes. While sha ed decision
making has been ad oca ed as a p omising app oach in
o de o imp o e ma ching o ea men s o pa ien s,
pa ien sa is ac ion, and ou come, he ex en o decision
making in ol emen ha is necessa y o a sha ed deci-
sion making p ocess is unde deba e. E.g. he e is con-
sis en e idence o a high (and o en unme ) need o
ea men in o ma ion by pa ien s, bu i has also been
ound ha in some ins ances, pa ien s do no wan o
be esponsible o making ea men decisions.
Thus, in high-s ake decisions such as eme gency and
li e- h ea ening si ua ions, a pa e nalis ic app oach
migh be mo e easonable while in si ua ions whe e
ea men decisions a e mo e con o e sial, he sha ed
decision making o e en he consume choice model
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 2 o 11
migh be p e e able [13]. The e is no e idence on which
app oach is o be p e e ed in CDM si ua ions ega ding
he ca e o people wi h SMI.
Clinical decision making in ch onic illness
Much o he li e a u e on CDM has ocused on acu e
and se ious medical condi ions. In immedia e and high-
isk acu e ca e si ua ions, he clinician is o en seen as
he p ima y sou ce o medical knowledge and as he
sole au ho i y o deciding on ea men op ions, which
is mos o en es ic ed on whe he o no o comply
wi h his o he ecommenda ion [8]. Howe e , decision
making in ch onic condi ions such as SMI di e s om
decision making in acu e ca e in se e al aspec s. Wa
[8] in oduced a amewo k o unde s anding decision
making in ch onic condi ions which can also be applied
o SMI. The au ho delinea ed se e al ac o s impac ing
di e en ly upon decision making in acu e s. ch onic ill-
ness (see Table 2).
Acco ding o his amewo k, CDM in pe sis en con-
di ions such as SMI - as opposed o well-de ined acu e
ca e si ua ions - has o ake in o accoun ha : (a) ea -
men ocus is on long- e m disease managemen ; (b) a
high numbe o decisions ha e o be a i ed a e-
quen ly, o en oge he wi h mo e han one se ice p o-
ide and/o ca e s; and (c) pa ien s in gene al a e
highly knowledgeable abou hei illness. Due o
inc eased accessibili y o ea men - ele an in o ma ion
e.g. ia in e ne o sel -help sou ces, pa ien s migh e en
ha e mo e ecen and be e in o ma ion han hei se -
ice p o ide s.
Resea ch on clinical decision making in gene al
Resea ch on CDM has ocused on a ange o physical
condi ions, p edominan ly in well-de ined sho - e m
li e- h ea ening e en s (hea a ack, s oke), bu has
also looked a p olonged s a es o ill heal h, e.g. cance
and ib omyalgia. The esea ch ocus has been p ima ily
on o mal decision analysis in high isk- isk acu e ea -
men s such as one- ime acu e ea men choices in su -
ge y ia hypo he ical scena ios/ igne es. These
app oaches ha e been c i icised o being o e ly cogni-
i e and decon exualised, and also o lack o gene alisa-
bili y o esul s [5].
F om his ype o esea ch, decision ees and decision
aids ha e been gene a ed. While he numbe o decision
suppo applica ions has been inc easing apidly du ing
he las yea s, he e is s ill a ange o open ques ions
ega ding hei use, con en , and o ma [14]. Wi h he
excep ion o a decision aid o dep ession medica ion,
he e is cu en ly no publicly a ailable decision aid o
men al illness [5]. In o ma ion o pa ien decision mak-
ing a ies widely in i s quali y and comp ehensibili y
and is no always adequa ely accessible o pa ien s o be
use ul o decision making [15]. Some posi i e e ec s o
decision aids ha e been iden i ied, e.g. on pa ien s’
knowledge and unde s anding o hei condi ion, ea -
men op ions, and ou come p obabili ies, as well as on
ag eemen be ween pa ien p e e ences and subsequen
ea men decisions [16]. Howe e , a e iew o 200 deci-
sion aids has also shown decision aids ailed o imp o e
sa is ac ion wi h decision making, anxie y, and heal h
ou comes [17].
Resea ch on clinical decision making in men al heal h
I is doub ul whe he he gene al concep s o CDM
desc ibed abo e ca y o e well o men al heal h ca e
p o ision. On one hand, men al heal h is unique o
medicine in ha some pa ien s a e being ea ed agains
hei will [13]. The e o e, gene ic indings on CDM may
no be applicable in ce ain ins ances, e.g. among people
who ha e expe ienced in olun a y men al heal h ea -
men [18]. Also, he e migh be pa ien s who ail o pe -
cei e pe sonal con ol o choices as a eali y [5]. On he
o he hand, pa ien s a e inc easingly ecognised as key
Figu e 1 A simpli ied model o decision making (adap ed om
Wills e al. [5]).
Table 1 Models o ea men decision making
Pa e nalis ic model Sha ed decision making model In o med (pa ien ) model
In o ma ion
ans e
One-way (doc o o pa ien ) ans e o
minimum medical in o ma ion necessa y
o in o med consen
Two way: doc o p o ides all medical
in o ma ion needed o decision making. Pa ien
p o ides in o ma ion abou p e e ences
One way (doc o o pa ien )
ans e o all medical in o ma ion
needed o decision making
Delibe a ion Doc o alone, o wi h o he doc o s Doc o and pa ien (possibly wi h o he s) Pa ien (possibly wi h o he s)
Decision abou
implemen ing
ea men
Doc o Doc o and pa ien Pa ien
No e. Adap ed om En wis le e al. [9].
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 3 o 11
decision make s in men al heal h ca e, and i has gene -
ally been shown ha choice is impo an o pa ien s and
imp o es engagemen wi h se ices [13].
Some ecen s udies epo ha peoplewi hmen al
illness wan a say in hei ca e. Hamann e al. [19]
ound ha in pa ien s wi h schizoph enia he desi e
o decision making was sligh ly s onge han among
pa ien s in p ima y ca e [7]. Simila esul s ha e been
epo ed o communi y men al heal h pa ien s in Eng-
land [20]. The au ho s also showed ha he e was a
g ea a ia ion in he ex en o which pa ien s wan ed
o be in ol ed in decisions ega ding hei ca e.
Fu he mo e, low le els o pa ien in ol emen in
medical decisions we e obse ed in p ima y ca e con-
sul a ions o dep essi e pa ien s [21] while e ec s o
he decision p ocess on pa ien sa is ac ion and ea -
men ou come we e no assessed. A ecen RCT ound
ha sha ing medical decisions wi h acu ely ill people
wi h schizoph enia is easible. Howe e , e ec s o a
sha ed decision in e en ion we e only shown wi h
ega d o he le el o knowledge abou he illness
(which was highe ) and pe cei ed in ol emen in med-
ical decisions (which was inc eased), bu no o symp-
om le el [22].
A ew s udies on gi ing pa ien s a choice in selec ing
be ween a limi ed numbe o (mos ly wo) di e en
b oad ea men op ions (e.g. psycho he apy s. medica-
ion) ha e been conduc ed. While some posi i e e ec s
ha e been shown o ea men adhe ence (lowe d op-
ou a es o pa icipan s who we e gi en a choice) in
people wi h dep ession [23,24], esul s ega ding clinical
ou come a e mixed. No clea e ec s o pa ien p e e -
ence on ou come we e e ealed in s udies wi h cocaine
abuse s [25] and wi h people wi h dep ession in p ima y
ca e [26], whe eas e ec s ha e been shown in people
wi h alcohol abuse [27] and wi h phobia [28]. I has also
been shown ia con e sa ion analysis ha pa ien s wi h
SMI no only wan a say in hei ca e bu a e ac i ely
in ol ed in nego ia ing ca e [29].
Assessmen o clinical decision making and o ea men
ou come in men al heal h
In o de o sc u inise he ela ion be ween quali y o
CDM and ou come in he ca e o people wi h SMI, ea-
sible measu es wi h good psychome ic p ope ies
including sensi i i y o change a e necessa y o cap u e
i al elemen s o CDM and ea men ou come.
While some ins umen s o measu ing he quali y o
decision making in gene al heal h ca e ha e been pu
o h [7,30], ins umen de elopmen o assessing he
quali y o decision making in men al heal h condi ions
has begun only ecen ly [31]. Ins umen s p edominan ly
ocus on pa ien s’app aisal o hei in ol emen in
ea men decisions o on pa ien au onomy. Howe e ,
he e is a lack o ins umen s cap u ing c ucial basic ea-
u es o CDM in men al heal h ca e including: (a) cha -
ac e is ics o clinical decisions; (b) pa ien (and clinician)
sa is ac ion wi h clinical decisions; and (c) ac ual ( s.
p e e ed) pa ien in ol emen in making clinical
decisions.
On he o he hand, du ing he las yea s signi ican
p og ess has been made in measu ing men al heal h
ou comes. Fi s , s anda dised e sions in se e al Eu -
opean languages o ins umen s measu ing key ou come
domains in he ea men o people wi h se e e men al
illness ha e been p esen ed [32]. Second, i has been
shown ha con inuous assessmen o ea men ou -
come ia s anda dised ins umen s is easible in people
wi h men al illness [3,33]. Fu he mo e, ecen e idence
indica es ha people wi h men al illness a e well
equipped and able o use mode n communica ion ech-
nologies o ou come assessmen [34-36], and ha eli-
able and alid ou come a ings can be ob ained ia he
in e ne [37].
Resea ch need
While subs an ial e idence has been accumula ed ia
a he e ined and heo y-based me hods o CDM in
physical condi ions, esea ch on CDM in men al illness
Table 2 Fac o s in clinical decision making in acu e s. ch onic illness
Fac o s Acu e illness Ch onic illness
Na u e o illness Disc e e; ime-limi ed; ea able Pe asi e; long- e m; manageable
Decisions Cu e ocused Con ol ocused
Na u e Deal wi h cause; minimal side e ec s Symp om educ ion; sequellae p e en ion; side e ec s ade-o
Numbe Single Mul iple; epe i i e
E idence used Focused on illness Focused on illness plus li es yle; li le on mul iple ch onic condi ions
and hei in e ac ion
Decision making
ela ionship
Pa ien and ea men ocused; Pe mission o
p o ide o ac
Consume and symp om ocused; Pe mission o consume o ac
Decision making
en i onmen
Tempo a y dis up ion un il pa ien is well Pe manen ly al e ed o accommoda e symp oms and managemen
No e. Adap ed om Wa [8].
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 4 o 11
is s ill a an ea ly s age. By inco po a ing mul iple me h-
ods, esea ch on CDM should go beyond he labo a o y
se ing which is also consis en wi h calls o s udy men-
al heal h phenomena and in e en ions unde less han
con olled eal-wo ld condi ions [38].
Resea ch has ocused ei he on how o help pa ien s
make decisions, o on how o unde s and he deg ee o
in ol emen in decision making desi ed by he pa ien ,
bu no on he na u e (kind, numbe ) o CDM in e e y-
day li e. Key esea ch challenges in CDM in he ca e o
people wi h SMI include [5,8]:
•Desc ip i e esea ch and ins umen de elopmen
ocussing on how decisions a e ac ually made in ou ine
ca e, and how he p ocess o decision making ela es o
e e yday beha iou s and ou comes;
•Imp o emen o measu es o cha ac e ising decision
making p ocesses ha a e ma ched o s udy popula ions,
complexi y, and ype o decision making, especially in
people wi h se e e and long-s anding men al diso de
who a e obliged o make mul iple and epe i i e ea -
men decisions, o en in coope a ion wi h mo e han
one ea men p o ide ;
•In o ma ion abou he psychological impac o
pa ien pa icipa ion in making complex and s ess ul
decisions;
•Decision making s yles o bo h pa ien s and o p o i-
de s and how hese s yles a e enac ed in a a ie y o
CDM encoun e s;
•How decision making esul s in cong uen o con-
lic ing ou comes and how all pa icipan s e alua e such
ou comes.
Fu he mo e, quali y o CDM in he ca e o people wi h
SMI has ye o be s udied om an in e na ional pe spec-
i e which would yield insigh s in o commonali ies and
di e ences o CDM be ween di e en coun ies and men-
al heal h se ice sys ems. The mos impo an app oach
(i.e. bo h clinically ele an and c ucial o clinical go e n-
ance) would be a ocus on wha le el o pa icipa ion a
pa ien wan s in hei ca e, and whe he a good ma ch
be ween desi ed and expe ienced le el o pa icipa ion has
any impac on ei he sa is ac ion o ou come.
Resea ch ques ion
Main objec i e o his s udy is o de elop a me hodology
o assess he scope and quali y o clinical decisions in
he ca e o people wi h SMI om bo h he pa ien and
clinician pe spec i e, and o speci y how and o wha
deg ee CDM in ou ine ca e a ec s pa ien beha iou
and sho - and long- e m ea men ou come. Thus, he
main s udy hypo heses a e:
(1) P ima y
(a) The quali y o CDM can be adequa ely desc ibed
by aking in o accoun decision making s yles,
sa is ac ion wi h decision making, and ype o
decision making ("pa e nalis ic” s. “sha ed” s.
“in o med”) om bo h pa ien and clinician pe spec-
i e as well as hei cong uence o incong uence.
(b) The ype and quali y o CDM is posi i ely ela ed
o ea men ou come in he ou ine ca e o people
wi h SMI.
(c) Ac ual CDM in ou ine ca e depends on con ex
a iables, i.e. a ies o di e en ypes o decision
and is suscep ible o change o e ime.
(2) Seconda y
(a) The ela ion be ween quali y o CDM and ou -
come is a ec ed by a numbe o co a ia es a he
le el o
(i) he pa ien (sociodemog aphic s a us, clinical
cha ac e is ics, symp om se e i y),
(ii) he clinician (expe ience, expe ise),
(iii) hei in e ac ion,
(i ) he quali y o hei he apeu ic ela ionship,
( ) he cong uence o incong uence o CDM
p ocess om pa ien and clinician pe spec i e,
( i) he se ice sys em (a ailabili y o and access
o ea men ).
(b) The quali y o CDM is ela ed o se ice use, i.e.
mo e adequa e se ice use is o be ound in people
wi h a high quali y o CDM.
Me hods/Design
The s udy “Clinical decision making and ou come in
ou ineca e o peoplewi hse e emen alillness”
(CEDAR) will es a model o CDM in people wi h SMI
as shown in Figu e 2.
This model shows ha he ocus o CDM is he in e -
ac ion be ween pa ien and clinician who a e cha ac-
e ised by a numbe o a ibu es including decision
making s yle and o m a he apeu ic alliance o a ce ain
quali y. Bo h pa ien and clinician as well as hei alli-
ance a e a ec ed by aspec s o he se ice sys em, e.g.
whe he a gi en in e en ion is a ailable o a o dable.
This builds he con ex o clinical decisions which du -
ing a gi en pe iod di e in kind (e.g. ela ed o pha ma-
cological o psychosocial ea men s) and numbe , as
well as in ex en o which hey con ain elemen s o
“sha ed decision making”. In e media e consequences o
decision making a e sa is ac ion wi h decision making
om he pe spec i e o bo h pa ien and clinician and
pa ien beha iou (adhe ence). The esul o CDM is
clinical ou come which should cap u e di e en domains
(symp oms, quali y o li e, needs) and be obse ed om
di e en pe spec i es (pa ien , clinician, and indepen-
den a e ). The e a e a numbe o possible eedback
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 5 o 11

loops. The mos ob ious is an ou come-o ien ed adap a-
ion o clinical decisions ollowing e alua ion o adhe -
ence o and ou come o he p e ious decision.
Speci ically, as de i ed om his model, he ela ion-
ships be ween he ollowing a iables will be in es iga ed
in a p ospec i e mul i-cen e s udy:
(a) Pa ien desi e o and expe ience o in ol emen in
CDM;
(b) Cong uence o pe cep ion o CDM (CDM ype,
sa is ac ion) be ween pa ien and clinician;
(c) Quali y o he he apeu ic ela ionship;
(d) Pa ien sa is ac ion wi h hei in ol emen ( he
hypo hesis being ha a high ma ch be ween desi ed and
expe ienced in ol emen will be associa ed wi h highe
sa is ac ion);
(e) T ea men adhe ence (wi h he same hypo hesis);
( ) Ou come (needs, quali y o li e, symp oms).
Design and ec ui men
CEDAR is a na u alis ic p ospec i e longi udinal obse -
a ional s udy wi h bimon hly assessmen s du ing a
12-mon h obse a ion pe iod (T0-T6). Pa icipan s a e
being ec ui ed om caseloads o ou pa ien /commu-
ni y men al heal h se ices a six cen es h oughou
Eu ope: Depa men o Psychia y II, Ulm Uni e si y,
Ge many (coo dina ing cen e); Sec ion o Reco e y a
Ins i u e o Psychia y, London, U.K.; he Depa men
o Psychia y a Second Uni e si y o Naples, I aly; he
Depa men o Psychia y a Deb ecen Uni e si y,
Hunga y; he Uni o Psychia ic Resea ch a Aalbo g
Psychia ic Hospi al, Denma k; and he Depa men o
Gene al and Social Psychia y a Uni e si y o Zu ich,
Swi ze land.
Be o e he s a o ec ui men in No embe 2009, he
s udy p o ocol has been app o ed by all cen es’e hics
commi ees. Only subjec s will be included who p o-
ided alid in o med consen . Each po en ial pa icipan
in his esea ch p ojec , p io o consen , will be clea ly
in o med o i s goals, i s possible ad e se e en s and he
possibili y o e use o pa icipa e o o wi hd aw con-
sen wi hou any ad e se consequences. In o med con-
sen will be asked only o pe sons able o eely
unde s and and ques ion.
Inclusion and exclusion c i e ia
Sc eening o inclusion and exclusion c i e ia is ca ied
ou by quali ied esea ch wo ke s in close con ac wi h
clinical s a .
Inclusion c i e ia
•Adul age (18-60 yea s) a in ake;
•Men al diso de o any kind as main diagnosis es ab-
lished by case no es o s a communica ion using SCID
c i e ia;
•P esence o se e e men al illness (Th eshold
Assessmen G id ≥5 poin s and illness du a ion ≥2
yea s);
Figu e 2 Model o clinical decision making in he ca e o people wi h se e e men al illness o be es ed in CEDAR.
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 6 o 11
•Expec ed con ac wi h men al heal h se ices
(excluding inpa ien se ices) du ing he ime o s udy
pa icipa ion;
•Su icien command o he hos coun y’s language;
•Capable o gi ing in o med consen .
Exclusion c i e ia
•Main diagnosis o men al e a da ion, demen ia, sub-
s ance use o o ganic b ain diso de ;
•Cogni i e impai men se e e enough o make i
impossible o gi e meaning ul in o ma ion on s udy
ins umen s;
•T ea men by o ensic psychia ic se ices.
Ins umen s and da a collec ion
Using in ensi e li e a u e sea ch and ocus g oup me h-
odology [39], h ee ins umen s we e de eloped in he
cou se o p epa ing he s a o he s udy (Ap il -
Oc obe 2009):
(1) Clinical Decision Making S yle Scale (CDMS
CEDAR) in o de o comp ehensi ely assess na u e (p e-
e ences, au onomy, in o ma ion seeking) and s abili y o
pa ien s’and clinicians’decision making s yle bo h a
baseline and a one-yea ollow-up (21 i ems);
(2) Clinical Decision Making in Rou ine Ca e Scale
(CDRC CEDAR) in o de o measu e key aspec s o
CDM om he pa ien and clinician pe spec i es as hey
un old in ou ine ca e (4 i ems plus 3 ollow-up i ems
measu ing deg ee o implemen a ion o he decision
iden i ied a he las CEDAR assessmen and signi ican
li e e en s since hen).
(3) Clinical Decision Making In ol emen and Sa is ac-
ion Scale (CDIS CEDAR) in o de o assess subjec i e
sa is ac ion and in ol emen wi h clinical decision mak-
ing (7 i ems).
Table 3 lis s he ins umen s used in CEDAR o assess
ele an a iables by ime poin s o hei applica ion and
a e pe spec i e(s).
Uni o analysis o he CDM measu es is always he
decision a i ed a du ing he mee ing p io o he cu -
en assessmen poin as indica ed by he pa ien .
Fu he mo e, adhe ence o he decision indica ed a he
p e ious ime poin is sc u inized ia he espec i e
i ems in he CDRC ollow-up ("Ha e you implemen ed
he decision iden i ied in you las CEDAR assessmen
wo mon hs ago?”).
All ins umen s used we e made a ailable in all cen-
es’languages ia in ensi e o wa d and backwa d
ansla ion ollowing common s anda ds [40]. Da a is
collec ed ia ques ionnai es ( illed in by he pa ien o
his o he key wo ke ) o ia in e iews conduc ed by
he CEDAR s udy wo ke . Da a en y modes a e ia
compu e o pape -pencil o ms.
Sample size
Sample size calcula ion was pe o med o he analyses o
he p ima y ou come, i.e. whe he needs a ed ia he
CANSAS-P a e a ec ed by he quali y o decision making
du ing he one-yea obse a ion ime. Following Hedeke
e al. [56], assuming a cons an g oup e ec o e ime
wi h a andom-e ec s uc u e and au o-co ela ed esi-
duals, and es ima ing a panel a i ion o 5% a each mea-
su emen poin , a small e ec size (0.2 SD) should be
de ec ed wi h a powe o 0.80 a a wo- ailed signi icance
le el o 0.05 wi h a g oup sample size o N = 222 o six
ime poin s (and o N = 238 o eigh ime poin s).
Requi ed sample size o se en ime poin s as in ou
design was es ima ed ia in e pola ing he di e ence in N
om six o eigh ime poin s wi h all o he elemen s o
he equa ion emaining unchanged: N = (222 + 238)/2 =
230. Fo his analyses, pa icipan s will be g ouped in wo
ca ego ies o quali y o decision making, esul ing in a
o al equi ed g oup size a T0 o N = 460. This means
ha baseline sample size o be ec ui ed a each cen e is
N = 77 (a e ounding o he nex highe in ege ).
Va ia ion be ween cen es will be aken in o accoun
by ea ing cen es as clus e s. Acco ding o Donne
[57], he a iance in la ion ac o (o design e ec ) is
gi en by IF =1+(m-1)* ;whe em= clus e size,
and = ICC (in e -clus e co ela ion).
Wi h m = 77 and = .003, IF = 1,22, esul ing in an
adjus ed sample size o N = 561 (94 pe cen e).
P ocedu es
Desc ip i e epo sincludeabsolu eand ela i e e-
quencies o ca ego ical a iables, and means and s an-
da d de ia ions (and minimum, median and maximum
as well he 25%- and 75%-pe cen iles whe e applicable)
o con inuous a iables. Be ween-cen e di e ences will
be explo a i ely es ed by c
2
-Tes s o ac o s and by
T- es s o ANOVAs esp. o con inuous a iables. The
e ec o he in e en ion on needs, quali y o li e and
symp oma ic impai men will be es ed by means o
hie a chical linea models [41] wi h he ime a iable
(0, 2, 4, 6, 8, 10 and 12 mon hs). Random e ec s will be
obse a ions “wi hin”subjec o e ime, and ixed e ec s
will e ec s o ime, quali y o clinical decision making
and o he co a ia es (see Figu e 2) on he gi en ou -
come measu e. All a ailable da a will be used in he
da a analysis. Sum scales will be p o a ed in case o
missing alues on less han 80% o he single i ems mak-
ing up he sco e. Fu he mo e, clus e analyses will be
used o a i e a meaning ul ca ego ies o quali y o
decision making om he CDM measu es applied, and
chain modelling [42] will be used o d aw possible cau-
sal in e ences om he panel da a.
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 7 o 11
Discussion
Du ing he las decades, almos all EU membe s a es
ha e unde gone subs an ial psychia ic e o ms. In addi-
ion, men al heal h p ac ice and esea ch has p o ided a
la ge numbe o pha macological and psychosocial in e -
en ions wi h p o en e icacy and also e ec i eness o
imp o ing clinical ou come and quali y o li e among
people wi h SMI. S ill, as s a ed in he EC’s ecen
G een Pape [43], “men al heal h o he EU popula ion
can be conside ably imp o ed”(p. 3). We belie e ha
an e ec i e way o achie e his is no so much he
de elopmen o u he new in e en ions, bu o see o
ha exis ing e ec i e ea men a e being o e ed and
u ilised ia speci ying bes p ac ices o clinical decision
making in he ca e o people wi h se e e men al illness.
As desc ibed abo e, du ing he las decades conside -
able e idence has been accumula ed on CDM in physi-
cal condi ions, especially in well-de ined high- isk
si ua ions. Howe e , he e is a sho age o esea ch ind-
ings on CDM in he ou ine ca e o people wi h pe sis-
en diseases such as se e e men al illness.
High-quali y desc ip i e esea ch is needed in o de
o gain knowledge on he s uc u e and p ocess o
CDM in he ou ine ca e o people wi h SMI. Fu he -
mo e, mo e knowledge is needed on he immedia e
and long- e m e ec s o CDM on sa is ac ion wi h
decision making, pa ien beha iou , and mos impo -
an ly on ea men ou come. The igo ous sc u iny o
hese issues in a well-designed la ge mul ina ional p o-
spec i e obse a ional s udy will yield insigh s in o
gene al e ec i e ing edien s o CDM and in o speci ic
ing edien s applicable o speci ic men al heal h se ice
sys ems a indi idual (pa ien , clinician) and se ice
le el, bu also in o ac o s no eadily amenable o
change, and hus subs an ially ad ance he s a e-o -
he-a in he ield.
In he ollowing, CEDAR’sexpec ed impac s will be
ou lined in ela ion o he opics o he call (FP7-
HEALTH-2007-3.1-4: Imp o ing clinical decision mak-
ing, [44] p. 44).
De elop and alida e me hodology o measu e
he quali y o clinical decisions
Ins umen s o cap u e s uc u e, p ocess, and ou come o
decision making in he ca eo peoplewi hSMIwillbe
de eloped and empi ically alida ed: (a) S uc u e: Clinical
decision making s yle and key elemen s o clinical deci-
sions; (b) P ocess: Con ibu ion o pa ien and clinician o
CDM, pa ien beha iou (immedia e and long- e m); (c)
Ou come: Sa is ac ion wi h CDM, and clinical ou come
(immedia e and long- e m). Since he e is no cu en gold
s anda d o a clinical decision, CEDAR will con ibu e o
an ou come-o ien ed concep ualisa ion o CDM quali y:
Clinical decisions o a “good quali y”a e hose wi h a
s ong associa ion wi h good clinical ou come.
Apply me hodology o explain a ia ions o ca e
esul ing om clinical decision making
Th ough a mul i-cen e p ospec i e obse a ional s udy,
a comp ehensi e model o CDM will be es ed in people
Table 3 S udy ins umen s by pe spec i e and measu emen poin
Va iable Ins umen Pe spec-
i e
Measu emen
poin
0 1- 5 6
Clinical cha ac e is ics (diagnosis, illness
du a ion)
S uc u ed Clinical In e iew o DSM-IV on he basis o case no es (SCID
[45,46])
P
R
✓
Sociodemog aphic s a us, se ice use Clien Sociodemog aphic and Se ice Receip In en o y (CSSRI-EU [49]) P
R
✓✓
Illness se e i y Th eshold Assessmen G id (TAG [47]) P
R
✓✓
CDM S yle Clinical Decision Making S yle Scale (CDMS CEDAR) P/S ✓✓
CDM in Rou ine Ca e Clinical Decision Making in Rou ine Ca e Scale (CDRC CEDAR) P/S ✓✓✓
CDM In ol emen and Sa is ac ion Clinical Decision Making In ol emen and Sa is ac ion Scale (CDIS
CEDAR)
P/S ✓✓✓
Needs Cambe well Assessmen o Need Sho App aisal Schedule (CANSAS
[48,49])
P✓✓✓
Quali y o Li e Manches e Sho Assessmen o Quali y o Li e (MANSA[50]) P ✓✓
The apeu ic ela ionship Helping Alliance Scale (HAS [51]) P/S ✓✓✓
Symp oma ic impai men Ou come Ques ionnai e (OQ-45.2 [52]) P ✓✓
Heal h o he Na ion Ou come Scale (HoNOS [53]) S ✓✓
Func ioning Global Assessmen o Func ioning Scale (GAF [54]) S ✓✓
Reco e y S ages o Reco e y Ins umen (STORI-30 [55]) P ✓✓
No es. CDM: Clinical Decision Making; P: Pa ien ; S: S a ; P/S: Pa ien and S a ; P
R
: Pa ien , esea che -led, 0: baseline assessmen ; 1- 5: in e media e assessmen s
(2, 4, 6, 8, and 10 mon hs); 6: inal assessmen (12 mon hs).
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 8 o 11
wi h se e e men al illness in di e en coun ies wi h di -
e en men al heal h se ice sys ems. The objec i e is o
ex ac bes p ac ices o CDM, i.e. o iden i y s uc u e
and p ocess a iables wi h a subs an ial ela ion o clini-
cal ou come. Since quali y o CDM is jus one o many
ac o s impac ing upon clinical ou come, possible mod-
e a o s and media o s o he CDM-ou come ela ion
(e.g. sociodemog aphic cha ac e is ics, clinical a iables,
he apeu ic ela ionship, sa is ac ion wi h in ol emen )
will be comp ehensi ely included in he model. People
wi h SMI will se e as he popula ion o he es ablish-
men o bes CDM p ac ices. In case di e ences ela ed
o diagnoses should eme ge, hese p ac ices will be spe-
ci ied o di e en se e e men al diso de s (e.g. schizo-
ph enia and dep ession), and can se e as a model o be
ans e ed o o he pe sis en illnesses.
S eng hen he clinical go e nance p ocess o
imp o emen s in clinical decision making
Fu he mo e, by speci ying he ela ionship be ween
CDM and ou come, bes p ac ices o clinical decision
making in he ca e o people wi h se e e men al illness
will be made a ailable o s akeholde s (pa ien s, clini-
cian, heal h ca e unde s) and clinical go e nance will
be s eng hened. This will include he ques ion o
whe he and o wha ex en clinical e idence and guide-
lines a e pu in o p ac ice (in he CDM p ocess) in ou-
ine ca e o people wi h se e e men al illness. This will
also include a ho ough analysis whe he he pa ien /
use pe spec i e is ac i ely in eg a ed in he CDM p o-
cess in ou ine ca e. Thus, he s udy will con ibu e o
ou ine men al heal h ca e being based on an in eg a-
ion o p o essional and use pe spec i es.
The cen al a ge will be o p o ide a di e en ia ed
answe o he ques ion, “Wha amoun o pa ien in ol-
emen is mos bene icial (i.e. subs an ially ela ed o
pa ien sa is ac ion, pa ien beha iou , and clinical ou -
come)inwha kindo clinicaldecision?”This will be
done on a gene al le el, bu also ake in o accoun a ia-
ions in se ice p o ision be ween he pa icipa ing cen-
es in Ge many, UK, I aly, Hunga y, Denma k, and
Swi ze land. This will lead o a se o good p ac ice
poin s which will gi e guidance on how o imp o e
CDM in he se ice p o ision o people wi h SMI.
Op imising he deli e y o heal h ca e and ansla ing
he esul s o clinical esea ch in o p ac ice
Communica ion be ween clinicians and pa ien s builds
he con ex o he deli e y o men al heal h ca e in he
o m o speci ic ea men s a he pa ien -le el. While
being a ec ed by a wide backg ound o sys em le el
a iables (i.e. he ex en o which local men al heal h
policy and se ice p o ide o ganisa ions adequa ely
suppo he p o ision o e idence-based in e en ions),
CDM can be ega ded he p ima y means o ansla ing
he esul s o clinical esea ch in o p ac ice.
The e is a lack o knowledge on he scope and quali y
o CDM in he ca e o people wi h ch onic diseases
such as SMI. In addi ion, a small numbe o s udies
in es iga ing he e ec o in e en ions o imp o e
CDM in men al heal h ha e yielded mixed esul s, and
pa icula ly ha dly any on clinical ou come. A ho ough
examina ion o CDM and i s ou come in his ield ia a
mul i-cen e p ospec i e s udy will help o ill his gap.
By iden i ying elemen s o bes p ac ice CDM (i.e.
aspec s o CDM wi h a subs an ial ela ion o good
ea men ou come), CEDAR will p o ide e idence
di ec ly con ibu ing o op imising he deli e y o heal h
ca e o Eu opean ci izens. Fu he mo e, CEDAR will
pa e he way o he de elopmen o a ge ed in e en-
ions o imp o e CDM in men al heal h.
Lis o abb e ia ions
CEDAR: Clinical Decision Making and Ou come in Rou ine Ca e o People
wi h Se e e Men al Illness (s udy ac onym); CDIS CEDAR: Clinical Decision
Making In ol emen and Sa is ac ion Scale (ins umen ); CDM: Clinical
decision making; CDMS CEDAR: Clinical Decision Making S yle Scale
(ins umen ); CDRC CEDAR: Clinical Decision Making in Rou ine Ca e Scale
(ins umen ); SMI: Se e e men al illness; ANOVA: Analysis o Va iance
Acknowledgemen s
The CEDAR s udy is unded by a g an om he Se en h F amewo k
P og amme (Resea ch A ea HEALTH-2007-3.1-4 Imp o ing clinical decision
making) o he Eu opean Union (G an no. 223290).
CEDAR is a mul i-cen e collabo a ion be ween he Depa men o Psychia y
II, Ulm Uni e si y, Ge many; he Sec ion o Reco e y, Ins i u e o Psychia y,
King’s College London, U.K. (in he con ex o he NIHR Specialis Men al
Heal h Biomedical Resea ch Cen e a he Ins i u e o Psychia y, King’s
College London and he Sou h London and Maudsley NHS Founda ion
T us ); he Depa men o Psychia y, Uni e si y o Naples SUN, I aly; he Uni
o Psychia ic Resea ch, Aalbo g Psychia ic Hospi al, Aa hus Uni e si y
Hospi al, Denma k; he Medical and Heal h Science Cen e , Depa men o
Psychia y, Uni e si y o Deb ecen, Hunga y; and he Depa men o Gene al
and Social Psychia y, Uni e si y o Zu ich, Swi ze land.
The CEDAR g oup includes: Sabine S e en, Pe a Neumann, Ka in A nold,
Es a-Sul an A a , Nadja Zen ne (Ulm); Ha ie Jo dan, S ephen Williams
(London); Co ado De Rosa, Domenico Giacco (Naples); Zol án Nemes, Tibo
I ánka, Agnes Su eges (Deb ecen); Malene F økjæ K ogsgaa d Bo ding,
Helle Øs e ma k Sø ensen (Aalbo g); A le e Bä (Zu ich).
We a e g a e ul o he membe s o he CEDAR ad iso y boa d: Sue Es o
(Depa men o An h opology, Uni e si y o No h Ca olina a Chapel Hill,
U.S.A.), Ma ga e a Ös man (Facul y o Heal h and Socie y, Malmö Uni e si y,
Sweden), Di k Rich e (Depa men o Heal h, Be n Uni e si y o Applied
Sciences, Swi ze land), Is àn Bi e (Depa men o Psychia y and
Psycho he apy, Semmelweis Uni e si y, Budapes , Hunga y).
We also wish o hank Is àn Deg ell
†
o ini ia ing he CEDAR coope a ion
wi h Deb ecen Uni e si y, Hunga y.
Au ho de ails
1
Depa men o Psychia y and Psycho he apy II, Ulm Uni e si y, Ludwig-
Heilmeye -S . 2, 89312 Günzbu g, Ge many.
2
King’s College London, Ins i u e
o Psychia y, Box P029, De C espigny Pa k, London SE5 8AF, UK.
3
Depa men o Psychia y, Uni e si y o Naples SUN, La go Madonna delle
G azie, 80138 Naples, I aly.
4
Uni o Psychia ic Resea ch, Aalbo g Psychia ic
Hospi al, Aa hus Uni e si y Hospi al, Møllepa k ej 10, 9000 Aalbo g, Denma k.
5
Medical and Heal h Science Cen e , Depa men o Psychia y, Uni e si y o
Deb ecen, Nagye dei k . 98, 4012 Deb ecen, Hunga y.
6
Depa men o
Gene al and Social Psychia y, Uni e si y o Zu ich, Mili ä s asse 8, 8021
Zu ich, Swi ze land.
Puschne e al.BMC Psychia y 2010, 10:90
h p://www.biomedcen al.com/1471-244X/10/90
Page 9 o 11